Infinidat - Enterprise Data & Cloud Storage Solutions https://www.infinidat.com/rss.xml en The Myths and Realities of Software-Defined Storage https://www.infinidat.com/blog/myths-and-realities-of-software-defined-storage <span class="field field--name-title field--type-string field--label-hidden">The Myths and Realities of Software-Defined Storage</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/SDS%20blog%20thumbnail_0.jpg?itok=SXTc25NL" width="330" height="330" alt="" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/65" typeof="schema:Person" property="schema:name" datatype="">Drew Schlussel</span></span> <span class="field field--name-created field--type-created field--label-hidden">Fri, 06/28/2019 - 2:00pm</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>Software-Defined Storage (SDS) is not a new technology; it is a go-to-market strategy that decouples storage hardware from storage software purchases. Claimed SDS benefits include more agility, more scalability, more performance, more fault tolerance, a standardized hardware infrastructure that creates economies of scale, faster hardware refresh cycles, no vendor lock-ins, and lower purchase prices. However, the relative lack of success for SDS, outside of a handful of very specific use cases, highlights the fact that obtaining these promised benefits is not without risks and costs for both vendors and users.</p> <h5>Executive Summary</h5> <ul><li>SDS is not going to displace on-premises primary storage arrays supporting mission-critical applications for the foreseeable future</li> <li>Hyper-Converged Infrastructures (HCIs) will continue to account for the bulk of on-premises SDS usage for the foreseeable future</li> <li>Whether SDS becomes a significant share of the secondary storage market remains to be determined</li> </ul><p>Let's go through how I arrived at these and more conclusions...</p> <h5>Analysis</h5> <ul><li>User interest in SDS is driven by the desire to build software-defined infrastructures and to lower storage acquisition and ownership costs</li> <li>Vendor interest is driven by the need to: offer a cost-effective alternative to the cloud; create an appealing hybrid cloud infrastructure vision, and protect existing storage array revenue streams as well as protecting customers' investments in their technology</li> </ul><p>All storage arrays sold today are software-defined because their backend media is virtualized and abstracted into LUNs, file systems, or object stores. However, that is sidestepping the real questions about where and how SDS is being purchased and used. </p> <p>The three primary use cases for SDS are:</p> <ol><li>As a foundational technology within hyper-converged integrated systems (HCIS)</li> <li>At the edge of IOT infrastructures</li> <li>Building inexpensive scale-out secondary storage arrays</li> </ol><p>To understand why SDS is not displacing storage arrays in mission-critical environments, we need to understand the tension that exists between on-premises storage vendors' business needs and the market realities.</p> <h6>On-Premises Vendor Business Needs</h6> <ul><li>Vendors have no choice but to compete against HCIs, SDS, cloud providers, and other storage array vendors</li> <li>The need to protect existing storage array revenue streams and their customers' investments in these arrays has forced established storage vendors to sell SDS defensively. In other words, the user has to motivate their incumbent storage vendors to bid their SDS solutions</li> <li>Established storage vendors do not want to introduce new storage architectures into the market that will put them in the position of having to recompete for their customers</li> </ul><h6>Market Realities</h6> <ul><li>Businesses are rethinking their acquisition procedures to take full advantage of the CapEx, capacity on demand (COD) and consumption-based pricing models</li> <li>Many privately held SDS vendors lack the credibility, marketing, sales bandwidth, ecosystem support, and testing capabilities needed to displace arrays supporting mission-critical applications in petabyte-scale data centers</li> <li>CIOs, storage administrators, and operations directors are not going risk their ability to meet mission-critical SLAs when most data growth is in unstructured data that can be stored on inexpensive on-premises storage or in the cloud</li> </ul><h6>End User Realities</h6> <ul><li>Deploying storage arrays is a safer alternative to deploying standalone SDS solutions</li> <li>Large enterprises are evaluating the operational and efficiency advantages of building more cloudlike infrastructures </li> </ul><p><em><strong>For InfiniBox customers and petabyte-scale prospects ready to deploy InfiniBox arrays there is no need to read further because InfiniBox is delivering many of the promised benefits of SDS without the risks inherent in becoming your own systems integrator.</strong></em></p> <p>For those that need or want a deeper understanding of SDS pros/cons vs storage arrays, let's begin by acknowledging that storage failures are longer lasting and therefore more painful and expensive to recover from than compute or networking failures and that interest in SDS is driven primarily by the desire to lower storage acquisition and ownership costs. If an application aborts, it's restarted from a checkpoint. Recovering from a Software-Defined Networking (SDN) failure can take more time than recovering and restarting an aborted application, but it's still short compared to recovering from an SDS loss of data integrity that can result in hours to days of hobbled operations as data is restored or declaring a disaster and moving operations to a disaster recovery site.</p> <h6>Observations</h6> <ul><li>Established storage vendors will continue selling SDS defensively until they can sell SDS without damaging their revenue streams, or competition from on-premises and cloud providers forces them to sell SDS proactively</li> <li>Current storage capacity growth forecasts have made discounting SDS by more than 10% to 15% below their equivalent array solutions is a money loser for established storage vendors: the more they sell, the more they lose</li> </ul><h6>User Guidance</h6> <ul><li>Installing SDS requires that you become your own systems integrator. You are buying the hardware, managing software and firmware upgrades across all nodes, and integrating the hardware with licensed software</li> <li>Monetize additional support and downtime costs to ensure that SDS savings do not evaporate over time or with a single outage</li> <li>Do not buy SDS hardware and software from different vendors unless there is a compelling financial or strategic reason to do so because it can: <ul><li>Create finger pointing situations when problems arise</li> <li>It could force established storage vendors to try and claw back lost hardware margins by raising software prices</li> </ul></li> </ul><h5>Claimed SDS advantages</h5> <p><strong>Agility</strong> – Moving data between arrays or data centers is an inherently time-consuming, resource intensive, error-prone process. Moving bits from one location to another is real work and the reason why data generally stays where it was first stored even if it's no longer cost-effective. One of InfiniBox's advantages is that it moves data between DRAM, Flash, and HDDs in real time without human intervention to optimize price/performance and without consuming SAN bandwidth.</p> <p><strong>Scalability</strong> – For "SDS washed" array software SDS instantiations this claim is a non sequitur because it's the same software that runs the vendors' arrays, but for scale-out SDS instantiations this is a mostly true claim, but one built on many assumptions. Among the more prominent claims are:</p> <ul><li>Scale-out SDS has lower acquisition and ownership costs than storage arrays</li> <li>Performance will scale linearly and provide a consistent performance experience with high node counts</li> <li>Intermixing asymmetric and different generation nodes in the same cluster will not cause problems. This is almost inevitable if the organization is pursuing a just in time upgrade strategy because of the short marketing lives of servers, HDDs, and SSDs.</li> <li>Scale-up arrays cannot satisfy the organization's performance and capacity needs.</li> <li>SDS vendors, without the gross margin of hardware sales, can build and sustain an effective support organization and fund R&amp;D at a high enough level to maintain large compatibility support matrices and product competitiveness.</li> </ul><p>While InfiniBox cannot compete on price against small SDS configurations, at petabyte-scale the cost comparisons between scale-out SDS and InfiniBox demand closer analysis. InfiniBox's scale-up integrated mixed-media architecture lowers InfiniBox controller and media costs vs a scale-out SDS solution that relies heavily on Flash to deliver performance. InfiniBox arrays also reduce total ownership costs by automating data placement and being complemented by a broad set of productivity tools that includes InfiniVerse, InfiniMetrics, and Host Power Tools.</p> <p><strong>Performance</strong> – More performance sounds great, but without a baseline reference it's meaningless; furthermore, if it's 30% or more past the organizations' worst case forecasted need over the planned service life of the storage solution, it doesn't make a difference. If maintaining consistent performance requires frequent and episodic tuning, the higher ownership costs can easily wash away any initial acquisition savings. InfiniBox DRAM centric data flows, coalesced writes, and QoS capabilities essentially eliminate the creation of hotspots and scaling to up to 4 PB of usable capacity and up to 10 PBs of effective capacity when best practice configuration techniques are followed.</p> <p><strong>More Fault Tolerance</strong> - Since scale-out architectures generally have more components than scale-up arrays of comparable performance and capacity, they need more fault tolerance to match a scale-up arrays usable availability. However, moving beyond this simple, common-sense observation are more fundamental and important observations. More specifically:</p> <ul><li>HDD and SSD failures account for the bulk of hardware failures in a scale-up or scale-out array. Hence it is the MTBFs of the HDDs and SSDs, the number of failures that can be tolerated while still guaranteeing data integrity, and rebuild times that will most influence the mean time between data losses of an array. InfiniBox HDD groups can tolerate 2 HDD failures before losing the ability to guarantee data integrity and its rebuild times are measured in minutes, not hours or days</li> <li>Human errors and software bugs account for approximately 80% of all downtime <ul><li>InfiniBox is self-management capabilities and superb ergonomics reduce the opportunities for human errors</li> <li>InfiniBox's low frequency of patch activity reflects high software quality</li> <li>InfiniBox software is market validated by our customers</li> </ul></li> </ul><p><strong>Standardized Hardware Infrastructure Creates Economies of Scale</strong> - This is a true but limited statement because server margins are lower than storage array margins and there are limits to discounting whether buying servers or storage arrays. Stated differently, going beyond buy quantities that deliver the best possible prices does not result in even better prices. Since Infinidat's go to market strategy is focused on install base growth, rather than protecting existing revenue streams, their use of price/performance as a value proposition greatly diminishes SDS's purported price advantage.<br /> Because data tends to stay where it first landed the flexibility of moving data between servers within a standardized infrastructure is of more academic and practical interest because moving data between arrays or storage nodes is a time consuming, error-prone process high impact process. </p> <p><strong>Faster Hardware Refresh Cycles</strong> – These are dictated by the vendor, not the user, unless the user is willing to run uncertified hardware. Asset management and internal qualification policies can also retard the adaption of new hardware platforms. Many enterprises will not buy a new storage solution until it has been market validated, meaning that it has been used actively in the market for 6 to 12 months. Then, if there is an internal qualification procedure, that could add 3 more months before deploying into production, with the result that users know that they will be 9 to 15 months behind the newest technology. Infinidat as a matter of historical record has done two hardware refreshes since launching InfiniBox in early 2013, and on May 7, 2019, guaranteed that its newest InfiniBox is both NVMe-oF and Storage Class Memory (SCM) ready. </p> <p><strong>No Vendor Lock-Ins</strong> – This is an unobtainable fantasy because once you buy hardware or software from a vendor, you create a relationship that requires resources to break. The lock-ins can be weak or strong, technical, financial, procedural, visible or invisible, emotional, or legal. Array replication technologies have historically been the most difficult to break because changing them can break so many things. The right business decision is to accept them where they provide needed capabilities or create competitive advantage. That said if we can acknowledge that NFS is perhaps the least sticky protocol in ubiquitous use, InfiniBox's NFS implementation can help users minimize lock-ins and potentially delay or avoid converting applications to S3 or other RESTful object protocols because it can store billions of files without performance imploding.</p> <p><strong>Lower Purchase Price</strong> – Without lower ownership, upgrade, and refresh costs, a low-cost storage solution can easily become an expensive storage solution. Since scale-up arrays usually just add media and enclosures when adding capacity vs. SDS scale-out arrays which usually add nodes, (complete with HBAs, microprocessors, DRAM, power supplies, blowers, and media) their cost of goods (COGs) is generally going to be higher than the COGs for an equivalent scale-up architecture such as an InfiniBox. This means that acquisition and maintenance costs are going to be set by the vendor's go-to-market strategy, competitive pressures, and the user's negotiating skills – ditto for upgrade costs. </p> <p>However, ownership costs are determined by more than acquisition, maintenance, power and cooling costs. Personnel, backup/restore, downtime, and lost opportunity costs (i.e., the lack of agility) are also significant contributors, and these are very heavily influenced by array automation, scripting capabilities, GUI ergonomics, data flows, code quality, and fault tolerance – all areas of InfiniBox excellence. While downtime costs are always hard to calculate, reframing the question into "How many extra downtime events need to occur before the inexpensive storage solution becomes a false saving?" If the answer is "one," it's probably not the best decision.</p> <h5>Bottom Line Conclusions</h5> <ul><li>SDS is NOT going to displace on-premises primary storage arrays supporting mission-critical applications for the foreseeable future</li> <li>HCIs account for the bulk of on-premises SDS usage for the foreseeable future</li> <li>Whether SDS becomes a significant share of the secondary storage market remains to be determined</li> <li>SDS instantiations of on-premises storage arrays, running on shared cloud infrastructures, will become a common hybrid cloud deployment model because it solves the technology translation problem at the infrastructure level that plagues current hybrid cloud deployment models</li> <li>Deployments at the edge will remain a popular SDS use case because of its ability to improve data availability while reducing environmental footprints and costs</li> <li><strong>InfiniBox, InfiniGuard, and Neutrix Cloud are delivering many of the values SDS promises, without any of the accompanying risks, WITH satisfaction that has 99% of Infindat customers proclaiming that they recommend Infindat storage to their peers</strong></li> </ul></div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="458" role="article" about="/node/458" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/458" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Stanley Zaffos</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Stanley is the Sr. VP of Product Marketing at Infinidat.<br /> Prior to joining Infinidat, he was a Research VP with Gartner focused on Infrastructure and Operations Management. His areas of expertise cover storage systems, emerging storage technologies, software-defined storage, hyper-converged infrastructure, and hybrid cloud infrastructure. He's worked with numerous clients to develop messaging and collateral that maximizes the impact of their product announcements and sales training, as well as helping to define roadmaps that ensure ongoing competitive advantage. </p> </div> </div> </article> </div> Fri, 28 Jun 2019 18:00:00 +0000 Drew Schlussel 475 at https://www.infinidat.com IBM XIV and A9000 Users - Don’t worry, you’re not abandoned, Infinidat is here to rescue you and your data! https://www.infinidat.com/blog/infinidat-data-rescue-program <span class="field field--name-title field--type-string field--label-hidden">IBM XIV and A9000 Users - Don’t worry, you’re not abandoned, Infinidat is here to rescue you and your data!</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/Data%20Rescue%20Program.png?itok=SUUhVWIG" width="330" height="330" alt="Infinidat Data Rescue Program" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/65" typeof="schema:Person" property="schema:name" datatype="">Drew Schlussel</span></span> <span class="field field--name-created field--type-created field--label-hidden">Tue, 06/04/2019 - 8:30am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>If the thought of replacing your XIV or A9000/R systems with a storage architecture that you already rejected when you chose your XIV or A9000/R system is less than appealing, we are ready to help! We are offering you a free migration onto Infinidat’s InfiniBox with our <a href="https://info.infinidat.com/data-rescue-web.html" title="Infinidat Data Rescue Program">Data Rescue Program</a>, and the super-easy option of using our InfiniBox FLX program to move storage from CapEx to OpEx. Our InfiniBox FLX allows you to only pay for the storage that you need, when you need it, and only for as long as you need it.</p> <p>In the early days of computer systems, the few original vendors that were in the market tended to allow new and rapidly-changing technology to define each new system that they would introduce, each one usually with new software, and a new product name. IBM’s rise to early dominance in the computer industry was primarily driven by an innovative change: the concept of an architecture, and specifically the System/360. The S/360 used a microcode abstraction layer and allowed for different models to be built with different technologies to implement the same architecture, enabling software investment and staff skills to be protected and carried forward. </p> <p>When it comes to systems, IBM’s track record for architectural investment protection has been excellent - not only evolving the S/360 into today’s Z Series but doing the same with the AS/400 into the iSeries and the RS/6000 into the Power series. However, when it comes to storage, IBM customers have experienced a very different legacy and reality. </p> <p>If we just restrict ourselves to the era of RAID protected storage arrays, IBM has launched and withdrawn the following mix of acquired and organically developed storage systems: </p> <ul><li>RAMAC 1, 2, and 3</li> <li>RVA</li> <li>RSA</li> <li>9337</li> <li>7135</li> <li>7137</li> <li>7133 (SSA)</li> <li>FasT500, FasT700</li> <li>VSS (Tarpon)</li> <li>MSS (with Compaq!)</li> <li>ESS (Shark)</li> <li>DS4000, DS5000 </li> <li>n-series </li> <li>XIV</li> </ul><p>Do you remember some major IBM publicly-touted storage projects that never became what was originally described at all, such as <a href="https://www.theregister.co.uk/2003/10/13/ibms_storage_tank_arrives_missing/">Storage Tank</a> and <a href="https://searchstorage.techtarget.com/news/1263835/IBM-washes-hands-of-IceCube-storage-project">IceCube</a>?</p> <p>On February 27, 2018, <a href="https://www-01.ibm.com/common/ssi/ShowDoc.wss?docURL=/common/ssi/rep_ca/8/897/ENUS918-028/index.html&amp;request_locale=en">IBM announced the withdrawal of XIV</a> systems, with a suggested replacement going forward to be the A9000. Now, just over a year later, it appears that the A9000 as well may not have much of a future, with <a href="https://en.globes.co.il/en/article-ibm-israel-laying-off-dozens-in-storage-division-1001286937">IBM confirming staff reductions</a> at a facility that supports it, and with last year’s announcement of the newer 9100 product.</p> <p>If you are one of IBM’s existing XIV customers, there are probably some things you really like about XIV. XIV revolutionized storage interfaces with its simplicity and graphical elegance. XIV found a way to get unexpected higher performance out of inexpensive SATA disks and could rebuild a failed drive faster than any system the industry has ever seen before.</p> <p>At Infinidat, we know what made the XIV popular. Moshe Yanai, our founder, and much of our original core team that created our InfiniBox were the same people that created XIV as well as EMC’s Symmetrix. They’ve taken the concepts people liked about XIV (and Symmetrix) and expanded it massively with InfiniBox. It’s like having the best of both worlds without the limitations, and more. Consider moving up to multi-PB scale in a single rack, 100% availability, faster-than-flash performance, all with the industry’s best TCO &amp; ROI.</p> <p>We’d never want to see anyone or their data abandoned, so let’s make this easy for you. </p> <p><strong>If you are an existing IBM XIV or A9000 customer, we’d like to suggest you move up to our Infinidat InfiniBox, with free data migration, with the option of using our InfiniBox FLX subscription as a low-risk strategy for leaving IBM storage behind and joining the Infinidat family.</strong></p> <p>Please contact your Infinidat partner, or contact us directly at Infinidat to learn more about the <a href="https://info.infinidat.com/data-rescue-web.html" title="Infinidat Data Rescue Program">Data Rescue Program</a>, <a href="https://www.infinidat.com/products-technology/infinibox" title="Infinidat InfiniBox">InfiniBox</a>, or <a href="https://www.infinidat.com/sites/default/files/resource-pdfs/InfiniBox_FLX%20DS-INFBXFLX-190331-US.pdf">InfiniBox FLX</a>.</p> <p>When it comes to protecting your investment with enduring storage architecture and innovative features, at Infinidat, we “get it” and we won’t let you down!<br />  </p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="430" role="article" about="/node/430" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/430" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Ken Steinhardt</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Ken Steinhardt serves as Field CTO at Infinidat. He has held many different roles in the IT industry over the last 45 years, having started as a mainframe applications programmer and systems analyst, part-time when he was still in high school. After receiving his BS degree in Computer Science from WPI, Ken was at Digital Equipment Corporation for 16 years, EMC Corporation for 20 years, and most recently joining Infinidat from Pure Storage. His longest-tenured role was as EMC's Vice President of Enterprise Storage &amp; CTO for Global Product Sales.</p> </div> </div> </article> </div> Tue, 04 Jun 2019 12:30:42 +0000 Drew Schlussel 462 at https://www.infinidat.com 12+1 Reasons That I Joined Infinidat https://www.infinidat.com/blog/12plus1-reasons-i-joined-infinidat <span class="field field--name-title field--type-string field--label-hidden">12+1 Reasons That I Joined Infinidat</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/stanley%20zaffos%20infinidat%20headshot.jpg?itok=0R-yb5sv" width="330" height="330" alt="" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/65" typeof="schema:Person" property="schema:name" datatype="">Drew Schlussel</span></span> <span class="field field--name-created field--type-created field--label-hidden">Wed, 05/22/2019 - 3:33pm</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>How does an industry veteran - someone with 40+ years experience, the last 22 as a VP of Research at Gartner, find himself back working with a vendor? The short answer to the question is that I became a believer in Infinidat: its people, culture, vision, products, development and support capabilities. It also helped a lot that the position they offered me sounded like it would be a lot of fun. </p> <p>So, what led to me becoming a believer? I will walk you through my thought process by listing the insights, market and technology trends, and end-user buying behaviors that led to my decision to join Infinidat. </p> <p><strong>1. Ease-of-use, support effectiveness, ecosystem size, cloud integration, scalability, and financials will increasingly dictate future infrastructure refresh decisions</strong></p> <p><strong>Why it matters</strong>: As traditional measures of product attractiveness (i.e. performance, availability, local and remote replication, multi-tenancy support, et al) increasingly fail to provide significant product differentiation, they are being replaced by the factors listed above. Performance in RAID or erasure protected storage arrays is a more scalable enabler than a differentiator, in its own right.  Ease of failover and failback capabilities are now more important than differences of minutes per year in predicted downtime.  Ecosystem size becomes a form of risk mitigation and deployment flexibility. Managing fewer, bigger arrays instead of many, smaller ones improves staff productivity. Lowering ownership costs frees resources to improve an organization’s competitiveness.<br />   <br /><strong>Advantage Infinidat</strong> - By any or all of the following measures of product and supplier attractiveness: AFA like ease of use and performance with PB scale and HDD economics, superb Gartner Peer Insights reviews, non-disruptive data mobility (NDM), and near autonomic integration with the cloud, an NFS implementation that can store billions of files without performance imploding that could enable users to delay or avoid having to invest in object storage and converting applications to S3 or another RESTful protocol. Capacity on Demand (COD) and consumption-based pricing models (InfiniBox FLX and InfiniGuard FLX) that lower customer costs and risks.<br />  </p> <p><strong>2. Inertia, asset management policies, and budget constraints limit most customers’ ability to absorb new technology on a broad basis</strong></p> <p><strong>Why it matters</strong>: With one exception, I’ve never spoken to a client that had more storage, staff or budget than they needed. I have talked with hundreds, possibly thousands, of clients that were pushed into infrastructure refreshes by high maintenance costs, performance and scale constraints, and architectural obsolescence. Large established vendors frequently restrict new features and functions to new arrays or delay the availability of these new capabilities to drive install base churn.<br />  <br /><strong>Advantage Infinidat</strong> – InfiniBox’s 3-controller architecture has enabled Infinidat customers to benefit from the functional enhancements and code optimizations that characterize new versions of InfiniBox software. We do not use new software releases to obsolete our customers’ investments in our arrays. Nor do we use expensive capacity upgrades or expensive maintenance to drive install base churn. Our focus is on growing the install base and satisfying our customers’ needs better than our competitors.<br />  </p> <p><strong>3. Today’s product advantages are tomorrow’s standard features</strong></p> <p><strong>Why it matters</strong>: There is a growing gap between new storage arrays’ active marketing lives and customers planned service lives. New arrays typically have a three-year active marketing lives because competitive pressures and technology improvements demand it, whereas customers now frequently plan on 5 or more years of service for their arrays. This creates a multi-year window of architectural obsolescence risk, especially for customers of large established vendors that depend upon install base churn for much of the revenues.<br />   <br /><strong>Advantage Infinidat</strong> – Infinidat’s practice of allowing older (n-1, n-2 generation)  systems to run current software effectively protects our customers from architectural obsolescence for 5 to 7 years, or up to two times longer than our established competitors. And that is before considering the protections against architectural obsolescence provided by our FLX consumption model, which provides free hardware refreshes, and non-disruptive data mobility(NDM) between Infinidat arrays. With the addition of the Infinidat Elastic Data Fabric and transparent data mobility, customers will have immortal data without worrying about their storage hardware.<br />  </p> <p><strong>4. Unstructured data accounts for 80% or more of forecasted storage data growth</strong></p> <p><strong>Why it matters</strong>: Data analytics and AI/ML applications need lots of data to provide useful insights and learn. The trade-off between IOPS, bandwidth, and cost often favors HDDs because these applications are more bandwidth intensive than IOPS sensitive. HDDs provide lots of low-cost bandwidth and capacity.</p> <p> <strong>Advantage Infinidat</strong> – 25GB/s host visible bandwidth with our new F63xx, sub-millisecond latency, HDD economics, an NFS implementation that can store billions of files without performance imploding, and petabyte scale satisfy a huge percentage of the requirements for analytics and AI/M without forcing the customer to utilize the cloud.<br />  </p> <p><strong>5. Approximately 66% of data residing on primary storage arrays is stale data, meaning that it has not been accessed in 60 to 90 days</strong></p> <p><strong>Why it matters</strong>: Storing stale data in AFA storage wastes money and is a luxury that most organizations can no longer afford. Deploying an archiving solution adds complexity to ongoing operations and D/R testing. Both increase costs, which can make hosting these applications in the public cloud an appealing alternative. However, if over time it resulted in the loss of critical IT skills, it could become a poor long term business decision.</p> <p> <strong>Advantage Infinidat</strong> – Storing both active and inactive data in the same array with HDD economics removes or greatly reduces the economic justification for deploying an archiving storage solution, even as it delivers on the previously listed benefits of scale, performance, and ease of use. It also buys time for new technology to solve the archiving, migration, and cost problems associated with retaining stale data.<br />  </p> <p><strong>6. Clever software running a hybrid array configured with a massive second level cache creates a storage array that can deliver flash performance, HDD economics, and scale to petabytes without complexity by offloading data traffic between SSDs and HDDs from the SAN</strong></p> <p><strong>Why it matters</strong>: InfiniBox beats AFAs on scale, economics, and protection against architectural obsolesce. It is competitive against AFAs [aka SSAs] on ease of use, IOPS, latency, and bandwidth. The only areas where hybrid arrays are arguably deficient relative to AFAs are in hostile environments, characterized by high vibration, large temperature swings, poor humidity control, lots of airborne contaminants, and dirty power: military vehicles, manufacturing floors, and edge deployments are some examples.</p> <p><strong>Advantage Infinidat</strong> – Self-evident in data center environments.<br />  </p> <p><strong>7. Disk will maintain at least a 5:1 to 10:1 raw $/TB advantage over flash through 2027</strong></p> <p><strong>Why it matters</strong>: Implementing data reduction technologies down to the HDD tier in hybrid arrays takes away the ability for AFAs to compete with hybrid arrays on an effective $/TB basis because of the HDD cost advantage relative to NAND flash. Other risks to data reduction include the limited value of deduplication outside of virtualization, VDI, and as a backup target. The fact that server-side encryption and rich media (video, medical imagery, audio) is essentially irreducible eliminates the benefits of deduplication.<br />  <br /><strong>Advantage Infinidat</strong> – self-evident in data center environments.<br />  </p> <p><strong>8. Artificial Intelligence for IT Operations is eroding the staff productivity advantages enjoyed by public cloud providers</strong></p> <p><strong>Why it matters</strong>: Tightly integrating AI Ops with an intelligent infrastructure will further narrow the skills gap and improve usable availability by continuously monitoring infrastructure performance.</p> <p> <strong>Advantage Infinidat</strong> – InfinBox is an intelligent storage system. It manages data placement and performance tuning for itself, and per previous “Why it matters” insights, it’s a very highly available petabyte (PB) scale, affordable speed demon that has few, if any, surprise “gotchas” vis-à-vis competitors with ad hoc collections of dissimilar storage arrays or storage services.<br />  </p> <p><strong>9. Hyperconverged integrated systems are generally not being deployed at PB scale </strong></p> <p><strong>Why it matters</strong>: HCIs are providing users with very good ease-of-use and availability. However, they do suffer two architectural shortcomings compared to HCI or reference architectures: their inability to scale performance and capacity independently and their inability to cost-effectively scale to PB capacity because of higher hardware costs and hypervisor licensing fees.</p> <p> <strong>Advantage Infinidat</strong> - InfiniBox’s near autonomic operation and quality software provide users with an ease-of-use that is competitive with market leading HCI solutions. A consistent sub-millisecond performance that doesn’t depend upon the locality of reference or internode bandwidth; PB scalability, and the ability to scale compute and storage separately without incurring additional licensing fees favors the Infinidat approach.<br />  </p> <p><strong>10. The perceived cost advantages of public clouds are eroding</strong></p> <p><strong>Why it matters</strong>: Much of the early interest in the cloud was driven by the desire to lower costs and shift CapEx to OpEx. Today it's driven by early access to new platforms, agility, and scale. Cost savings has declined as a reason to move to the cloud because of competitors’ responses to the cloud and lower than anticipated savings due to variable costs, complex bills, and the relative immaturity of cloud data services.</p> <p><strong>Advantage Infinidat</strong> – InfiniBox’s COD and FLX consumption-based pricing models, AFA like performance, and ease of use, make Infinidat an easy choice for mission-critical and exploratory applications. InfiniBox’s scale and high performance make it possible to solve the ETL (Extract Transform, and Load) or integration problems that often plague data analytics projects by enabling them to easily share storage at scale.<br />  </p> <p><strong>11. Microprocessor performance improvements are improving faster than data transfer rates for  SSDs or HDDs</strong></p> <p><strong>Why it matters</strong>: The delta between microprocessor and storage media bandwidth improvements gives scale-up storage an inherent cost of goods advantage relative to scale-out arrays by driving down the microprocessor to capacity ratio needed to avoid performance bottlenecks, even at PB scale.</p> <p><strong>Advantage Infinidat</strong> – Dropping the controller to capacity ratio lowers frequency of repair activities and advantages scale-up arrays because fewer controllers also mean there are fewer electronics to support - HBAs, power supplies, blowers, etc. - than scale-out arrays of equivalent capacity. This advantage generally increases with capacity because scale-out arrays frequently add capacity by adding nodes that include controllers and their support electronics. These and the environmental advantages of scale-up vs scale-out arrays help explain the continued success of scale-up arrays in the marketplace.<br />  </p> <p><strong>12. Hyperscale cloud providers putting their technology into customer data centers is an unmistakable indication that on-premises infrastructure is not going away and that hybrid cloud infrastructures will become a common deployment topology</strong></p> <p><strong>Why it matters</strong>: it removes the illusion that large on-premises IT operations will give way to the cloud; that the relationships between on-premises and cloud providers will be anything but contentious for the foreseeable future; and that solid-state data centers are just around the corner.</p> <p><strong>Advantage Infinidat</strong> – Infinidat has created storage solutions with high product attractiveness and the ability to compete on price as necessary, a simple comprehensible hybrid infrastructure vision (EDF), and a highly satisfied customer base. Cloud providers entering the on-premises storage market accustomed to supporting relatively few data centers will have to invest heavily in their support infrastructure to handle hundreds or thousands of customer data centers.<br />  </p> <p><strong>13. Co-location facilities and consumption-based pricing models are giving organizations a cloud-like cost structure without the security and variable cost risks that are inherent in cloud usage</strong></p> <p><strong>Why it matters</strong>: Co-lo facilities enable users to avoid infrastructure refreshes, upgrades, and/or replacements. Co-lo facilities also provide facilities on a consumption-based pricing model,  reduce the time needed to allocate space, power, and communications links to new hardware, and the ability to reduce communication costs for many colo clients and in the process reduce bring up times.  </p> <p><strong>Advantage Infinidat </strong>– InfiniBox’s COD and FLX consumption-based pricing models, EDF vision, Neutrix Cloud, InfiniSync replication technology, near autonomic operation, and use of standard 19” NEMA racks make InfiniBox  an appealing storage solution to install in a co-lo rack or cage, particularly if the co-lo facility is also hosting Amazon, Azure, or Google because it eliminates latency caused by distance and lowers bandwidth costs.<br />  </p> <p><strong>Conclusion</strong>: Time and economics are on Infinidat’s side because we are doing a better job in the important measures of product attractiveness, customer satisfaction, and enabling the economic transformation that gives a competitive advantage to our customers.</p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="458" role="article" about="/node/458" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/458" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Stanley Zaffos</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Stanley is the Sr. VP of Product Marketing at Infinidat.<br /> Prior to joining Infinidat, he was a Research VP with Gartner focused on Infrastructure and Operations Management. His areas of expertise cover storage systems, emerging storage technologies, software-defined storage, hyper-converged infrastructure, and hybrid cloud infrastructure. He's worked with numerous clients to develop messaging and collateral that maximizes the impact of their product announcements and sales training, as well as helping to define roadmaps that ensure ongoing competitive advantage. </p> </div> </div> </article> </div> Wed, 22 May 2019 19:33:42 +0000 Drew Schlussel 459 at https://www.infinidat.com Storage Class Memory Demo - Faster than All-Flash, Part 3 https://www.infinidat.com/blog/storage-class-memory-demo-faster-all-flash-part-3 <span class="field field--name-title field--type-string field--label-hidden">Storage Class Memory Demo - Faster than All-Flash, Part 3</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/InfiniBox%20SCM%20Ready.png?itok=xaM1pbSJ" width="330" height="330" alt="" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/65" typeof="schema:Person" property="schema:name" datatype="">Drew Schlussel</span></span> <span class="field field--name-created field--type-created field--label-hidden">Wed, 05/15/2019 - 8:00am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>The first two posts in this series were about why InfiniBox is faster than all-flash arrays. This post is about an InfiniBox being faster than itself. Thoroughly confused? Read on!</p> <p>InfiniBox systems are guaranteed SCM ready, and our engineering team was looking for a cool way to demonstrate this capability. Around this time we learned there are certain applications that require 100% of IOs to come off silicon - when it is an immutable requirement from the application vendor. These customers consolidate 99.9% of their workloads onto InfiniBox, but then have to go buy, and then manage, a small hardware AFA for that .1% of their workloads.</p> <p>We think there’s a better way, and it’s also a perfect way to demonstrate SCM support: <strong>a software-defined solid-state-array</strong> running inside InfiniBox.</p> <p>Seeing is believing, here’s a video demo:</p> <script src="https://fast.wistia.com/embed/medias/hg0uo0fjo5.jsonp" async=""></script><script src="https://fast.wistia.com/assets/external/E-v1.js" async=""></script><div class="wistia_responsive_padding" style="padding:56.25% 0 0 0;position:relative;"> <div class="wistia_responsive_wrapper" style="height:100%;left:0;position:absolute;top:0;width:100%;"> <div class="wistia_embed wistia_async_hg0uo0fjo5 videoFoam=true" style="height:100%;position:relative;width:100%"> <div class="wistia_swatch" style="height:100%;left:0;opacity:0;overflow:hidden;position:absolute;top:0;transition:opacity 200ms;width:100%;"><img alt="Image removed." data-entity-type="" data-entity-uuid="" onload="this.parentNode.style.opacity=1;" src="https://www.infinidat.com/core/misc/icons/e32700/error.svg" style="filter:blur(5px);height:100%;object-fit:contain;width:100%;" title="This image has been removed. For security reasons, only images from the local domain are allowed." height="16" width="16" class="filter-image-invalid" /></div> </div> </div> </div> <p>  </p> <p><img alt="" data-entity-type="" data-entity-uuid="" height="225" src="https://www.infinidat.com/sites/default/files/uploads/images/SCM%20latency%20graph.png" width="1271" /><em>Volume 1 with the Quality of Service Policy for 100% of I/O from SCM</em></p> <p>This capability is an engineering preview, mostly to demonstrate our SCM roadmap.</p> <p>However, given InfiniBox’s AFA-crushing performance, we’re not convinced there is yet a huge market for this feature, and there are many other roadmap features competing for near-term QA cycles. </p> <p>We’re super interested in customer feedback on this feature; if you have any opinions on this, please let your account team know, or shoot an email over to <a href="mailto:product@infinidat.com">product@infinidat.com</a>.</p> <p><em>This is part 3 of a 3-part series. Read part 1 <a href="https://www.infinidat.com/blog/all-flash-is-not-fast">here</a> and part 2 <a href="https://www.infinidat.com/blog/dell-please-stop-calling-powermax-worlds-fastest-storage-array">here</a>.</em></p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="52" role="article" about="/node/52" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/52" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Brian Carmody</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Brian is Chief Technology Officer at Infinidat.<br /><a href="https://twitter.com/initzero">Follow Brian on Twitter</a>.</p> </div> </div> </article> </div> Wed, 15 May 2019 12:00:00 +0000 Drew Schlussel 438 at https://www.infinidat.com Dell, Please Stop Calling PowerMax the World’s Fastest Storage Array - Faster than All-Flash, Part 2 https://www.infinidat.com/blog/dell-please-stop-calling-powermax-worlds-fastest-storage-array-faster-all-flash-part-2 <span class="field field--name-title field--type-string field--label-hidden">Dell, Please Stop Calling PowerMax the World’s Fastest Storage Array - Faster than All-Flash, Part 2</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/InfiniBox%20NVMe-oF%20Ready.png?itok=X0NVqTjQ" width="330" height="330" alt="" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/65" typeof="schema:Person" property="schema:name" datatype="">Drew Schlussel</span></span> <span class="field field--name-created field--type-created field--label-hidden">Wed, 05/08/2019 - 8:06am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p dir="ltr">If you follow what has been going on in the world of computer storage over the last few years, you’ve certainly been hearing the <a href="https://blog.purestorage.com/a-new-era-of-storage-with-nvme-nvme-of/">growing</a> <a href="https://www.brighttalk.com/webcast/13935/314707/dell-emc-powermax-a-deep-dive-into-the-all-new-nvme-based-all-flash-platform">excitement</a> <a href="https://www.netapp.com/us/info/what-is-nvme.aspx">about</a> <a href="https://www.youtube.com/watch?v=BXcxOrpBoj8">Non-Volatile</a> <a href="https://blog.architecting.it/end-to-end-nvme/">Memory</a> <a href="https://www.nextplatform.com/2019/02/26/bringing-nvme-over-fabrics-into-the-storage-big-tent/">Express</a>, or NVMe. It might have seemed like almost everyone associated with storage has been talking about NVMe - vendors, analysts, press, and of course the source itself; the nonprofit industry organization that oversees NVMe, <a href="https://nvmexpress.org/">nvmexpress.org</a>.</p> <p>What’s the big deal? In the most basic of terms, NVMe is a protocol that was designed to enable higher performance for communication between applications (or host systems) and solid state components. In <a href="https://nvmexpress.org/resources/faqs/">the words of the NVM Express organization</a>: “NVMe was formed to remove the bottlenecks in legacy storage infrastructure designed for hard drives, with a streamlined protocol, scalable performance, and industry standard software and drivers.”</p> <ul><li dir="ltr"> <p dir="ltr">It’s faster</p> </li> <li dir="ltr"> <p dir="ltr">It’s more efficient</p> </li> <li dir="ltr"> <p dir="ltr">It’s parallel</p> </li> <li dir="ltr"> <p dir="ltr">It has well-defined industry standards</p> </li> <li dir="ltr"> <p dir="ltr">It was architected for memory technologies rather than disk</p> </li> <li dir="ltr"> <p dir="ltr">It can work well with existing open interconnect technologies</p> </li> <li dir="ltr"> <p dir="ltr">It is extensible to work with future technologies</p> </li> <li dir="ltr"> <p dir="ltr">Most major IT vendors are using it</p> </li> </ul><p dir="ltr">But what will it take for NVMe to truly show its potential? NAND flash technology, which is commonly found in most SSDs, has a minimum latency somewhere around 100 microseconds. So when all of the other inherent latencies required to deliver an I/O in most all-flash arrays are considered, the total minimal latency is probably close to about 200 microseconds.</p> <h3 dir="ltr">Second Time Around</h3> <p>Two years and one month ago, I wrote a <a href="https://www.infinidat.com/blog/all-flash-is-not-fast">blog post </a>showing InfiniBox delivering much higher performance than two wildly popular all-flash-arrays. I’ve been asked not to curse, so let’s say there was a lovestorm on social media and our web server got the hug of death. Today, with an upgraded CDN, I’m delighted to share some news around NVMe-oF. The latest generation of InfiniBoxes are NVMe-oF ready, and we plan GA roll out for NVMe-oF as a non-disruptive software update within a year. For customers who are not part of the early access program, here’s a video demonstration of what’s coming:</p> <script src="https://fast.wistia.com/embed/medias/q476wyu0v4.jsonp" async=""></script><script src="https://fast.wistia.com/assets/external/E-v1.js" async=""></script><div class="wistia_responsive_padding" style="padding:56.25% 0 0 0;position:relative;"> <div class="wistia_responsive_wrapper" style="height:100%;left:0;position:absolute;top:0;width:100%;"> <div class="wistia_embed wistia_async_q476wyu0v4 videoFoam=true" style="height:100%;position:relative;width:100%"> <div class="wistia_swatch" style="height:100%;left:0;opacity:0;overflow:hidden;position:absolute;top:0;transition:opacity 200ms;width:100%;"><img alt="Image removed." data-entity-type="" data-entity-uuid="" onload="this.parentNode.style.opacity=1;" src="https://www.infinidat.com/core/misc/icons/e32700/error.svg" style="filter:blur(5px);height:100%;object-fit:contain;width:100%;" title="This image has been removed. For security reasons, only images from the local domain are allowed." height="16" width="16" class="filter-image-invalid" /></div> </div> </div> </div> <p> </p> <p dir="ltr">That’s right - <strong>32 uSec reads and 38 uSec writes</strong> - roundtrip, full stack, measured by fio. Woof.</p> <p data-entity-type="" data-entity-uuid="" style="text-align: center;"><span><img alt="" data-entity-type="" data-entity-uuid="" height="330" src="https://www.infinidat.com/sites/default/files/uploads/images/NVMe-oF%20Performance.png" width="619" /><span title="Click and drag to resize">​</span></span></p> <p>Let’s compare this with our old friends and add NetApp, since the A800 is a great hardware-based array that NetApp customers I speak with absolutely love.</p> <p data-entity-type="" data-entity-uuid="" style="text-align: center;"><span><img alt="" data-entity-type="" data-entity-uuid="" height="163" src="https://www.infinidat.com/sites/default/files/uploads/images/nvme_latency_table.png" width="626" /><span title="Click and drag to resize">​</span></span></p> <p data-entity-type="" data-entity-uuid="" style="text-align: center;"><span><span title="Click and drag to resize">​</span></span></p> <p class="small">*  <a href="https://www.dellemc.com/resources/en-us/asset/data-sheets/products/storage/h16891-powermax-family-ds.pdf"> Dell EMC PowerMax Data Sheet</a> <br /> ** <a href="https://www.netapp.com/us/media/ds-3931.pdf">NetApp AFF 800 Data Sheet</a><br /> *** <a href="https://www.purestorage.com/content/dam/purestorage/pdf/datasheets/ps_ds_flasharray_03.pdf">Pure FlashArray//X Data Sheet</a> (pre-NVMe-oF, claiming 250 uSec) and <a href="https://blog.purestorage.com/pure-delivers-directflash-fabric-nvme-of-for-flasharray/">Pure Blog claiming up to 20% faster with NVMe-oF</a><br /> Note: per the public sources listed above, these are what we believe to be the best performance floor latency claims currently listed by these vendors for these products.</p> <p class="small"> </p> <h3>Surprise, InfiniBox wrecks the all-flash-arrays, again.</h3> <p>I want to be clear that InfiniBox is not mopping the floor with PowerMax and friends (again) because we write better code; it’s because of physics. No matter what anybody does in software, a NAND cell can’t recall its data any faster than around 100 uSec, and an all-flash-array therefore cannot deliver IO latency less than 100 uSec + fabric + software stack latency. This is why, for over five years when customers ask when Infinidat is going to build an all-flash-array, we say “never” because NAND flash is 1000x slower than the DRAM that powers InfiniBox.</p> <p data-entity-type="" data-entity-uuid="" style="text-align: center;"><span><img alt="" data-entity-type="" data-entity-uuid="" height="562" src="https://www.infinidat.com/sites/default/files/uploads/images/latency%20chart.png" width="461" /><span title="Click and drag to resize">​</span></span></p> <p>AFA vendors know this too, and that’s why everybody’s racing to implement Storage Class Memory (SCM), such as 3DXpoint, as the next “big thing” to improve their latency performance. And it probably will reduce latency compared to NAND flash. But it still won’t catch up to DRAM. By using DRAM cache as our primary technology to send and receive I/Os to and from storage networks and servers, Infinidat will continue to have a tangible performance advantage over AFAs - even those with SCM.</p> <p><strong>NVMe-oF is a perfect match for InfiniBox</strong>, and we’re expecting rapid adoption, not only by Infinidat customers but by the industry as a whole. We still have plenty of work to do before NVMe-oF on InfiniBox will be ready for GA, but I hope you enjoyed the demo, and stay tuned for updates!</p> <p dir="ltr"><em>This is part 2 of a 3-part series. Read part 1 <a href="https://www.infinidat.com/blog/all-flash-is-not-fast">here</a> and part 3 <a href="https://www.infinidat.com/blog/faster-all-flash-part-3-storage-class-memory" title="Faster Than All-Flash, Part 3 - Storage Class Memory">here</a>.</em></p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="52" role="article" about="/node/52" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/52" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Brian Carmody</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Brian is Chief Technology Officer at Infinidat.<br /><a href="https://twitter.com/initzero">Follow Brian on Twitter</a>.</p> </div> </div> </article> </div> Wed, 08 May 2019 12:06:35 +0000 Drew Schlussel 436 at https://www.infinidat.com Where Have All The Wizards Gone? https://www.infinidat.com/blog/where-have-all-wizards-gone <span class="field field--name-title field--type-string field--label-hidden">Where Have All The Wizards Gone?</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/wizard.jpg?itok=pJSxBLgi" width="330" height="330" alt="Where Have All The Wizards Gone?" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/3" typeof="schema:Person" property="schema:name" datatype="">Eric Klinefelter</span></span> <span class="field field--name-created field--type-created field--label-hidden">Fri, 05/03/2019 - 10:38am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>As Dell Technologies World is going on this week, it seems appropriate to reflect back on one of its foundations - EMC World - and even more so on the foundation of EMC World itself - The EMC Enterprise Wizards Conference, better known in the day as simply “Wizards”. Using the defunct Digital Equipment Corporation DECUS events as a loose model, the first Wizards event in 2001 was meant to be an opportunity for technical customers to engage and interact directly with EMC engineering. The objective was to allow engineering to share the deep technical details about products, product direction, and to actively solicit and gather information from the customers to influence future product features and direction. It was an event run by technologists, for technologists.</p> <p>Dell Technologies World has become a glitzier, more marketing-oriented event for a more general audience that now includes analysts and the press. In years past, at least you could expect there to be a flurry of new, interesting storage-related product announcements.</p> <p>You be the judge as to whether that expectation was met this year, but in our opinion that didn’t happen, and I’m a bit saddened and genuinely surprised by that. It’s almost as if it was more about what WASN’T announced rather than what was.</p> <p><img alt="" data-entity-type="" data-entity-uuid="" height="392" src="https://www.infinidat.com/sites/default/files/uploads/images/image5.png" width="463" /></p> <p>Here are some things that many expected to be announced, but didn’t happen:</p> <p><strong>1. No announcement of pricing or availability for Storage Class Memory (SCM) or NVMe-oF for PowerMax</strong></p> <p><strong>Why was it expected?</strong><em> </em>There have been many public statements by Dell EMC starting one full year ago that PowerMax was “SCM ready” and “NVMe-oF ready”, and that these features <a href="https://www.dellemc.com/resources/en-us/asset/briefs-handouts/solutions/h16738-powermax-top-ten-ho.pdf" target="_blank"><em>planned to GA with PowerMax early 2019</em></a>”. Sometime recently, the word “early” has been quietly removed from almost all these statements in online documents. In a document that has been removed completely, the expected date was even more specific, stating that availability was planned for “Q1”. It appears that Dell EMC has done a good job of scrubbing their public documents to remove all previous references to these publicly-stated shipment expectations, now opting for a more generic date of “2019”. But if you’re an existing Dell EMC customer, you know the truth about what Dell EMC said, and what you were told. Fortunately, it’s a lot harder to get an industry analyst to redact or retract what they’ve already written, even when Dell EMC sponsored their piece, <a href="https://www.emc.com/collateral/white-paper/idc-nvme-unlocking-next-gen-tier-0-data-storage.pdf" target="_blank">so here’s a note from IDC</a> that still says “early 2019”, at least for now. It’s also tricky to remove something from a video, so it’s still mentioned <a href="https://www.dellemc.com/en-us/storage/powermax.htm#collapse&amp;video-overlay=5773299265001" target="_blank">here too</a>. I’m genuinely surprised that these things were not announced this week.</p> <p><strong>Why wasn’t anything announced?</strong> Without inside information we can only speculate, but none of the hypotheses are good for users doing storage infrastructure refreshes this year. Among the more plausible possible explanations, are:</p> <ol><li>A business decision that NVMe-oF and SCM memory are not needed now</li> <li>A change in business direction or priority that required the redeployment of development resources elsewhere</li> <li>Development delays caused by architectural limitations, technology problems, or resource shortages</li> </ol><p><img alt="InfiniBox" class="float-right" data-entity-type="" data-entity-uuid="" height="494" src="https://www.infinidat.com/sites/default/files/uploads/images/image4.png" width="178" /><strong>Why does it matter?</strong> One of the big lead marketing messages associated with Dell EMC’s announcement of PowerMax last year was the term “Smart”, regarding the ability to add value through “<a href="https://corporate.delltechnologies.com/en-us/newsroom/dell-technologies-powers-up-performance-and-efficiency-for-the-modern-data-center.htm" target="_blank">machine learning</a>”. It sure looks like what Dell EMC was referring to was mostly their legacy FAST-VP software carried over from VMAX. There’s just one problem with that – since PowerMax is an all-flash array currently, all of that legacy FAST-VP intelligence isn’t really being used, as all the back-end storage devices are still just NAND Flash. There aren’t any different tiers of storage to move data to than the one that they’re already stored on. PowerMax is indeed likely able to evaluate millions of data segments and make billions of decisions as Dell claims, but those decisions today are likely to effectively move nothing. That is, until (if) SCM eventually does get announced and shipped by Dell EMC, at which point all of that legacy FAST-VP “intelligence” may at least be able to partially achieve some of the kind of tiering value that it once had in its previous glory days running on multi-tiered hybrid VMAX configurations.</p> <p>But even when/if that does happen, SCM is still an order of magnitude slower for performance than DRAM. The intelligent caching (vs. Dell EMC’s “tiering”) on today’s <a href="https://www.infinidat.com/products-technology/infinibox">InfiniBox</a> using DRAM will still provide higher performance for most customers and their real-world applications.</p> <p>As for NVMe-oF, Dell’s statements that this technology is still in the early phases of deployment by customers are true in our opinion. But we believe that NVMe-oF is going to be popular for all the right reasons and coming to market VERY quickly - possibly ramping up to 50% or more of performance-oriented storage shipments over the next 3 to 5 years. Some storage vendors have already announced availability of NVMe-oF, and we expect that it will become a useful, open, and pervasive technology for front-end connectivity for most storage vendors going forward.</p> <p><strong>2. No Announcement of “Destination New Product/MidRange.NEXT”</strong></p> <p><strong>Why was it expected?</strong> As they say, sometimes a picture is worth a thousand words. The slide below was part of a Dell EMC analyst briefing in September 2018.</p> <p><img alt="" data-entity-type="" data-entity-uuid="" height="407" src="https://www.infinidat.com/sites/default/files/uploads/images/image3.png" width="725" /></p> <p>Source: <a href="https://investors.delltechnologies.com/static-files/cb255da9-52e6-4f36-8e67-1d6890207b30" target="_blank">Dell Technologies</a>, slide #39</p> <p>The absence of any significant visible efforts to rationalize Dell’s current portfolio of architecturally dissimilar storage arrays suggests that there may be customer retention and revenue issues or development concerns that are precluding rationalization. This despite several press articles where Dell EMC’s Jeff Clarke was quoted talking about this consolidation and future transition, such as <a href="https://www.crn.com/news/storage/new-container-based-dell-emc-midrange-storage-debuts-in-2019">here</a>, <a href="https://searchstorage.techtarget.com/blog/Storage-Soup/Dell-EMC-midrange-convergence-due-in-2019">here</a>, and <a href="https://www.crn.com/news/storage/300104581/dell-counters-hpe-storage-products-will-be-supported-throughout-their-lifetime-period.htm">here.</a></p> <p><strong>Why does it matter?</strong> Customers expect to be given a clear, detailed, and specific, roadmap to support their decisions for why they should buy products today. Instead of announcing new Unity capabilities that improve agility, availability, or ease of management Dell gave us what many students of the industry might call a boring “mid-life kicker” technology refresh. And instead of actually adding NVMe to Unity, we got more talk. As journalists and analysts start publishing their thoughts on this, it appears that <a href="https://searchstorage.techtarget.com/news/252462544/Dell-EMC-Unity-XT-expands-midrange-storage-overlap" target="_blank">we’re not the only ones</a> that are a bit confused by this while <a href="https://www.crn.com/news/storage/crn-exclusive-dell-emc-opens-up-about-new-midrange-storage-coming-this-year" target="_blank">waiting for the next “Power something”</a>. We are already a third of the way through 2019, and public word of this transition to a new product first <a href="https://www.crn.com/slide-shows/storage/300103821/5-things-you-need-to-know-about-dell-emcs-next-generation-storage-product-strategy.htm/2" target="_blank">surfaced about one year ago</a> in May 2018. This announcement of enhancements to the Unity family may imply that the “new” product that was supposed to replace Unity (and probably SC, and probably XtremIO) may be farther away than the “2019” date that has been stated by Dell in the articles linked above. Since it appears that Dell may have already missed their target dates for SCM and NVMe-oF as highlighted above, this may be the most logical explanation - that the new product may be so far out into the future that it was easier to just add some minor tweaks to the existing Unity platform and see if anyone notices, or cares. We assume we’ll all know the real story eventually.</p> <p>At Infinidat we have a <a href="https://www.infinidat.com/products-technology">single, simple hardware architecture</a> with triple-redundancy of all key components that has already proven that it can extend and evolve with new technologies and into new areas rather than being made obsolete by them.</p> <p><img alt="" data-entity-type="" data-entity-uuid="" height="398" src="https://www.infinidat.com/sites/default/files/uploads/images/image6.png" width="769" /></p> <p><strong>3. No announcement of Flash for Data Domain</strong></p> <p><strong>Why was it expected?</strong> The backup/restore market is changing rapidly, and customers have come to realize that the only reason we do backups is so if something goes wrong we can restore the data. Customers have come to realize that the speed at which the restore takes place (RTO) is the most critical aspect of a backup/restore solution. Since the architecture of Data Domain doesn’t seem to have been designed to use advanced intelligent caching for restore performance, an assumed relatively easier way to address the issue specifically for Data Domain would be to use flash on the back end, as some other vendors have done successfully. Of course this assumes that host visible bandwidth is not constrained by compute resources.</p> <p><strong>Why does it matter?</strong> To their well-deserved credit, Data Domain became a huge industry factor a decade ago by proving that using disks instead of tape could improve performance while also reducing cost through effective data deduplication. At Infinidat, we recognize the achievements of Data Domain for initially paving the way on <img alt="" class="float-right" data-entity-type="" data-entity-uuid="" height="304" src="https://www.infinidat.com/sites/default/files/uploads/images/image2.png" width="232" />those issues, but a truly effective backup/restore solution needs to also be able to restore, really, really fast as an even greater priority. Dell doesn’t seem to want to talk much about restore performance statistics.</p> <p>Our <a href="https://www.infinidat.com/products-technology/infiniguard">InfiniGuard</a> product is based upon and exploits the core architectural features of our InfiniBox storage platform, and adds deduplication and integration with popular third-party backup software. Our customers can achieve outstanding backup performance, but more importantly, they achieve restore performance that is only marginally less than the backup performance (!) and typically well above all competitive alternatives for both backup AND restore performance...</p> <p><strong>So what <em>did</em> Dell announce at Dell Technologies World relative to storage?</strong></p> <p>In our opinion, much less than in any previous year.</p> <ul><li>Unity upped the cache, storage capacities, and performance, mostly by changing to more-current versions of commodity technologies.</li> <li>Isilon got a new version of code that allows for a larger cluster</li> <li>A new entry-level IDPA system for small and medium-sized businesses</li> </ul><p>I suspect that like many of our storage competitors, we were anticipating having to take some time to analyze an expected mountain of new information and then craft a response. But this year, there really isn’t anything that we thought was important enough to be worthy of such an effort.</p> <p>In Dell’s keynote on Tuesday there did seem to be a lot of excitement on the main stage regarding the lead products that were mentioned first, before storage. I heard a lot about cool Dell laptops, how little a new Dell laptop weighs compared to other competitive laptops, laptop cameras, and Alienware. And the importance of the “edge” of the computing world - not to be confused with the “core” (data center) or “cloud”.</p> <p>Draw your own conclusions about where the priorities lie, and how much “innovation” is being applied to storage.</p> <p>At <a href="https://www.infinidat.com">Infinidat</a> we put our priorities on designing and delivering the best storage products that help our customers above all else. With an “engineering first!” culture of innovation, founded by storage industry legends, we have (so far) resisted big marketing events and campaigns, mostly by allowing <a href="https://www.gartner.com/reviews/market/general-purpose-disk-arrays/vendor/INFINIDAT" target="_blank">our customers’ voices and successes</a> to speak loudest for us. Infinidat’s value proposition is incredibly simple: Faster than all-flash performance, seven-nines of availability, multi-Petabyte Scale, incredibly easy to manage, at a substantially lower cost and TCO – all without compromise.</p> <p>How things have changed from that first EMC Wizards Conference in 2001! The original 3,000 attendees (half estimated to have been EMC employees) heard that according to a study by the University of California at Berkeley that 12 Exabytes of data existed in total in the world at that time. Infinidat’s worldwide storage capacity shipped in just the last few years since coming out of stealth is already over a third of that number, and growing rapidly.</p> <p>So, where have all the wizards gone? The answer might surprise you: they never left Massachusetts, they just joined Infinidat, in Waltham. More of the original EMC Fellows are working at Infinidat (3) than remain at Dell (0). Don’t get us wrong, the Dell successor program, Dell Corporate Fellowships, is a prestigious honor, and the storage industry certainly needs more innovators. So on behalf of the Wizards of Waltham, will Michael Dell please recruit and elevate more <a href="https://en.wikipedia.org/wiki/Radia_Perlman" target="_blank">Radia Perlmans</a>! We all benefit from innovation.</p> <p>But most importantly, we’d love to learn more about YOUR environment and discuss how we can help you scale to win with what our customers tell us is superior technology that gives them tangible advantages. You can <a href="http://info.infinidat.com/contact_us.html" target="_blank">contact us here</a>, since we’re still light on doing big marketing events, preferring to talk directly with our customers more intimately.</p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="430" role="article" about="/node/430" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/430" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Ken Steinhardt</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Ken Steinhardt serves as Field CTO at Infinidat. He has held many different roles in the IT industry over the last 45 years, having started as a mainframe applications programmer and systems analyst, part-time when he was still in high school. After receiving his BS degree in Computer Science from WPI, Ken was at Digital Equipment Corporation for 16 years, EMC Corporation for 20 years, and most recently joining Infinidat from Pure Storage. His longest-tenured role was as EMC's Vice President of Enterprise Storage &amp; CTO for Global Product Sales.</p> </div> </div> </article> </div> Fri, 03 May 2019 14:38:14 +0000 Eric Klinefelter 431 at https://www.infinidat.com SMB Revealed https://www.infinidat.com/blog/infinidat_reveals_smb_for_infinibox <span class="field field--name-title field--type-string field--label-hidden">SMB Revealed</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/BlogImages-1-SMB%20Revealed.jpg?itok=crhmmJaT" width="330" height="330" alt="SMB Revealed" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/3" typeof="schema:Person" property="schema:name" datatype="">Eric Klinefelter</span></span> <span class="field field--name-created field--type-created field--label-hidden">Tue, 02/26/2019 - 10:48am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>SMB is here! So what is it <em>really</em>?</p> <p>For some of the lucky ones, SMB is the Seven Mile Beach on the Grand Cayman in the Cayman Islands. For the rest of us working in IT, it is Server Message Block, a protocol widely used for sharing files, printers, etc. – mostly in the Windows OS environment.</p> <p>SMB protocol is not new. Its history starts in the mid-80s when it first appeared as “IBM PC Network SMB Protocol”. Microsoft adopted it in its LAN Manager product and has continued to drive the development of the protocol ever since. Although the first version of SMB, SMB 1.0, has been discontinued for years (see <a href="https://blogs.technet.microsoft.com/filecab/2016/09/16/stop-using-smb1/">one of Ned Pyle’s great blog posts</a>), modern versions are still extremely popular and widely used.</p> <p>Many of our customers like the InfiniBox unified storage capabilities – simultaneously supporting FC, iSCSI and NFSv3 within the same system. But what about SMB support? </p> <p>Many customers address the need for SMB by deploying physical or virtual Windows servers in front of InfiniBox block storage, essentially creating their own SMB solutions. They benefit from the full Windows Server functionality, but also have to deal with the day-to-day operation and maintenance of these dedicated file servers. They would be glad to keep the functionality but get rid of all the extra management burden.</p> <p>We certainly hear our customers requests and work hard to address their requirements. But first, we have to understand what customers really mean when they ask about SMB support. Do they need integration with Active Directory? Support for multiple AD domains? Perhaps they need access-based enumeration? Previous versions? Replication?</p> <p>All of these features and plenty of others are not exactly part of the SMB file access protocol – but most of the customers assume the availability of this functionality when they’re talking about their Windows file shares. We clearly see that the real demand is for a real Windows file sharing experience, not just the SMB protocol.</p> <p>Wait no more!</p> <p>The new 4.0.40 software release and SMB Node for the InfiniBox F4000 offers the SMB protocol up to version 3.1.1 with many additional features – multiple tenants, previous versions support, full integration with InfiniBox UI, etc. As always with INFINIDAT, it will be fully supported and monitored by our global support team.</p> <p><img alt="Screenshot" data-entity-type="file" data-entity-uuid="dc09fdb1-a994-4e15-baa4-5f79cc7ee73f" src="https://www.infinidat.com/sites/default/files/uploads/images/screenshot.png" /></p> <p>Screenshot of SMB file system management with the InfiniBox GUI</p> <p>InfiniBox is a unified storage system. Customers looking for a file sharing solution for home directories or departmental file shares can use InfiniBox SMB file systems, while Hyper-V or MS SQL environments can be deployed using InfiniBox FC and/or iSCSI volumes.</p> <p>As the Product Manager responsible for delivering this new capability, I’m thrilled to see initial feedback from our customers and I expect plenty of constructive feedback as we continue the roll-out of this new InfiniBox functionality.</p> <p>In my future posts I’ll share more details about deployment and use cases for our newest protocol family member. Stay tuned, and happy file sharing!</p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="15" role="article" about="/node/15" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/15" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Gregory Touretsky</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Gregory Touretsky (@gregnsk) is a Senior Director, Product Management at INFINIDAT. He drives the company’s roadmap around NAS, cloud and containers topics. Before that Gregory was a Solutions Architect with Intel, focused on distributed computing and storage solutions, data sharing and the cloud. He has over twenty years of practical experience with distributed computing and storage. Gregory has an M.S. in Computer Science from Novosibirsk State Technical University and an MBA from Tel-Aviv University.</p> </div> </div> </article> </div> Tue, 26 Feb 2019 15:48:16 +0000 Eric Klinefelter 14 at https://www.infinidat.com Fun with Neutrix Cloud – Spawning Multiple DB Instances Using Snapshots https://www.infinidat.com/blog/fun-with-neutrix-cloud-spawning-multiple-db-instances-using-snapshots <span class="field field--name-title field--type-string field--label-hidden">Fun with Neutrix Cloud – Spawning Multiple DB Instances Using Snapshots</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/BlogImages-2-NeutrixCloud%20Spawning%20Multiple.jpg?itok=BQ6ybVQA" width="330" height="330" alt="Fun with Neutrix Cloud – Spawning Multiple DB Instances Using Snapshots" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/3" typeof="schema:Person" property="schema:name" datatype="">Eric Klinefelter</span></span> <span class="field field--name-created field--type-created field--label-hidden">Wed, 11/21/2018 - 1:01pm</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>I like learning from our customers about the various ways they’re using our products, especially when it comes to Neutrix Cloud. I’ve recently written about how Neutrix Cloud can be used for <a href="https://www.infinidat.com/blog/fun-with-neutrix-cloud-from-rendering-to-devtest-to-spark-tasks/">rendering a movie in a multilcloud environment</a>, as well as <a href="https://www.infinidat.com/blog/running-spark-tasks-on-amazon-elastic-mapreduce-with-neutrix-cloud/">how it can work with AWS Elastic Map Reduce</a>. In this post, I’ll cover a dev&amp;test scenario, in particular as it applies to a database environment.</p> <p>Let’s assume you are running a large scale database in a public cloud or in your datacenter, consuming terabytes of block storage. This could be Oracle, MS SQL, or mySQL. You want to spawn a copy or two of the production environment as a pre-production testing environment. How long would it take you to create several copies of this data using a native, public cloud storage solution? How much would it cost you?</p> <p>Now let’s assume a database is still running in a public cloud, but the data is stored on Neutrix iSCSI volumes.</p> <p><img alt="Neutrix Cloud" data-entity-type="file" data-entity-uuid="404c2e4a-1e31-4ce7-b964-c9262ad58d12" src="https://www.infinidat.com/sites/default/files/uploads/images/Neutrix_Cloud_1.png" class="align-center" /></p> <p>First, you’ll have to take snapshots of your production environment. This is as simple in the cloud as it is in your on-premises environment:</p> <p>snap create --uuid LUNid</p> <p>These snapshots are created instantly, have no performance impact on your production, but even more important – when initially created, and until you change any data, they’re totally free. They don’t cost you anything.</p> <p><img alt="Neutrix Cloud" data-entity-type="file" data-entity-uuid="26ac66df-26d6-454a-a092-57521b82406d" src="https://www.infinidat.com/sites/default/files/uploads/images/Neutrix_Cloud_2.png" class="align-center" /></p> <p>These snapshots can be made writeable and become instant copies of the original LUNs. The best part – you’re only paying for the changes made on these snapshots:</p> <p>snap modify --uuid Snapid --writable true</p> <p>These snapshots can now be accessed from a separate set of DB servers, offering instant, affordable provisioning of your dev&amp;test environments. As soon as the environment is not needed anymore, it can be instantly removed.</p> <p>There many other use cases available with Neutrix Cloud storage as a key component. I’ll cover some more of them in future posts. Stay tuned, and explore Neutrix Cloud on your own. All you need to do is to contact your INFINDIAT account team.</p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="15" role="article" about="/node/15" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/15" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Gregory Touretsky</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Gregory Touretsky (@gregnsk) is a Senior Director, Product Management at INFINIDAT. He drives the company’s roadmap around NAS, cloud and containers topics. Before that Gregory was a Solutions Architect with Intel, focused on distributed computing and storage solutions, data sharing and the cloud. He has over twenty years of practical experience with distributed computing and storage. Gregory has an M.S. in Computer Science from Novosibirsk State Technical University and an MBA from Tel-Aviv University.</p> </div> </div> </article> </div> Wed, 21 Nov 2018 18:01:25 +0000 Eric Klinefelter 17 at https://www.infinidat.com Fun with Neutrix Cloud – Running Spark Tasks on Amazon Elastic MapReduce with NFS Backend https://www.infinidat.com/blog/running-spark-tasks-on-amazon-elastic-mapreduce-with-neutrix-cloud <span class="field field--name-title field--type-string field--label-hidden">Fun with Neutrix Cloud – Running Spark Tasks on Amazon Elastic MapReduce with NFS Backend</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/BlogImages-3-NeutrixCloud%20Running%20Spark.jpg?itok=fkUhxyrT" width="330" height="330" alt="Fun with Neutrix Cloud – Running Spark Tasks on Amazon Elastic MapReduce with NFS Backend" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/3" typeof="schema:Person" property="schema:name" datatype="">Eric Klinefelter</span></span> <span class="field field--name-created field--type-created field--label-hidden">Mon, 11/19/2018 - 9:01am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>Amazon Web Services has done a great job simplifying provisioning, scaling and deprovisioning of Hadoop and Hadoop-like environments with their Elastic MapReduce (EMR) service. A fully configured cluster with all required components can be brought online via a single command (or a few mouse clicks in the console) and can be used to process the data easily. The AWS EMR provisioning process allows specifying bootstrapping actions for every new cluster instance. With this mechanism, one can easily mount a Neutrix NFS file system on every EMR cluster node.</p> <p>A typical bootstrap action for this would be:</p> <p>$ cat neutrix-emr-bootstrap.sh set –e sudo yum -y -q update sudo /usr/bin/yum -y -q install nfs-utils rpcbind sudo /bin/mkdir -p /mnt/neutrix sudo /bin/mount $NFSIP:$NFSEXPORT /mnt/neutrix</p> <p>This script has to be stored in AWS S3, so it can be easily retrieved by every EMR cluster node:</p> <p>$ aws s3 cp neutrix-emr-bootstrap.sh s3://neutrix-emr-config/ upload: ./neutrix-emr-bootstrap.sh to s3://neutrix-emr-config/neutrix-emr-bootstrap.sh</p> <p>To launch an EMR cluster, specify a path to this bootstrap action script:</p> <p><img alt="EMR Cluster" data-entity-type="file" data-entity-uuid="c9b37787-6d7c-463a-8fef-e2eb2a46fb7c" src="https://www.infinidat.com/sites/default/files/uploads/images/EMR%20Cluster.png" class="align-center" /></p> <p>This command would bring up a fully-configured cluster with Hadoop, Spark, Hive and Pig environments using a shared Neutrix Cloud NFS file system providing the backend storage.</p> <p><img alt="NFS" data-entity-type="file" data-entity-uuid="d88d4c7f-9f80-4b74-ad46-335afe3aafd8" src="https://www.infinidat.com/sites/default/files/uploads/images/NFS.png" class="align-center" /></p> <p>Once this is completed, a user may submit a Spark task (“step”) to the cluster, using Neutrix Cloud to store the input, output and the script itself:</p> <p>$ aws emr add-steps --cluster-id j-3HB834POEIRB --steps Type=SPARK, Name="Neutrix Spark Program", Args="[--deploy-mode,cluster,--class,FlightSample, file:///mnt/neutrix/flightdata/flight-project_2.10-1.0.jar, file:///mnt/neutrix/flightdata/output]"</p> <p>As with the <a href="https://www.infinidat.com/blog/fun-with-neutrix-cloud-from-rendering-to-devtest-to-spark-tasks/">previous example</a>, this file system could also be replicated between on-premises InfiniBox storage and Neutrix Cloud as well as shared among multiple clouds as needed to better control costs.</p> <p>This is the second of three posts covering Neutrix Cloud use cases. You can find the first post <a href="https://www.infinidat.com/blog/fun-with-neutrix-cloud-from-rendering-to-devtest-to-spark-tasks/">here</a>. In the next and final post, I’ll cover <a href="https://www.infinidat.com/blog/fun-with-neutrix-cloud-spawning-multiple-db-instances-using-snapshots/">spawning multiple DB instances using Neutrix Cloud snapshots</a>.</p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="15" role="article" about="/node/15" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/15" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Gregory Touretsky</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Gregory Touretsky (@gregnsk) is a Senior Director, Product Management at INFINIDAT. He drives the company’s roadmap around NAS, cloud and containers topics. Before that Gregory was a Solutions Architect with Intel, focused on distributed computing and storage solutions, data sharing and the cloud. He has over twenty years of practical experience with distributed computing and storage. Gregory has an M.S. in Computer Science from Novosibirsk State Technical University and an MBA from Tel-Aviv University.</p> </div> </div> </article> </div> Mon, 19 Nov 2018 14:01:10 +0000 Eric Klinefelter 19 at https://www.infinidat.com Fun with Neutrix Cloud – From Rendering to Dev&Test to Spark Tasks https://www.infinidat.com/blog/fun-with-neutrix-cloud-from-rendering-to-devtest-to-spark-tasks <span class="field field--name-title field--type-string field--label-hidden">Fun with Neutrix Cloud – From Rendering to Dev&amp;Test to Spark Tasks</span> <div class="field field--name-field-blog-image field--type-image field--label-hidden field__item"> <img src="/sites/default/files/styles/medium/public/blog-images/BlogImages-4-NeutrixCloud%20Rendering%20to%20Dev%26Test_0.jpg?itok=kKZmHDii" width="330" height="330" alt="Fun with Neutrix Cloud – From Rendering to Dev&amp;Test to Spark Tasks" typeof="foaf:Image" class="image-style-medium img-fluid" /> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span lang="" about="/user/3" typeof="schema:Person" property="schema:name" datatype="">Eric Klinefelter</span></span> <span class="field field--name-created field--type-created field--label-hidden">Thu, 11/15/2018 - 9:06am</span> <div class="clearfix text-formatted field field--name-field-blog-body field--type-text-with-summary field--label-hidden field__item"><p>I am writing this blog post while flying to Japan to meet with some of our great customers. All I see out the window is the cloud-filled horizon. These endless clouds as far as the eye can see make everything looking white and fluffy…</p> <p>There is a similar view in the IT industry as well – the endless compute clouds covering the world, and we can see a growing demand among our customers to move some or all of their IT operations to these public clouds. While they appreciate the simplicity, agility and endless capacity, they are not willing to give up the functionality and reliability they enjoy in their private data centers. This is especially true for data storage, a critical component of any infrastructure deployment, whether it happens on-premises, in the clouds, or some mixture of both.</p> <p><img alt="Clouds" data-entity-type="file" data-entity-uuid="e22c4fae-f066-4c56-bad9-dfc80cb8c5db" height="145" src="https://www.infinidat.com/sites/default/files/uploads/images/Clouds_1.jpg" width="109" class="align-left" />At INFINIDAT, we certainly care about data storage. We see tremendous growth in our on-premises deployed capacity, now exceeding 4 exabytes, and we are talking to customers every day about how to help them with their public cloud journey. Our solution for storage in the public cloud is Neutrix Cloud – a cloud-adjacent storage-as-a-service offering, powered by our proven InfiniBox storage.</p> <p>There are three major usage models for Neutrix:</p> <ul><li>Replicating data from on-premises environments to Neutrix Cloud, potentially as a DR target</li> <li>Consuming data over NFS and/or iSCSI interfaces from virtual machines running in one or more public clouds, as a feature-rich alternative to native cloud storage solutions</li> <li>A combination of the above – replicating data between on-premises and Neutrix, and accessing it from one or more public clouds.</li> </ul><p>Over the next few blog posts, I’d like to review a few real use cases for Neutrix Cloud. We’ll use the Neutrix Command Line Interface in the examples below, but the same operations can also be done using the Neutrix API and our new GUI, both of which will be on display and available for demonstrations at this year’s AWS re:Invent, booth #1503. Let’s start with…</p> <p>Rendering a movie</p> <p>A design company is planning to run a rendering farm on the cloud. They have limited compute capacity in the data center, with enough resources only to perform interactive work. However, rendering requires much more compute power – and it might be very expensive. The public cloud is a well-known solution for this kind of challenge.</p> <p>All major public clouds offer an option to buy excess compute capacity at a lower price, what they call “spot instances”. At scale, it makes sense to compare pricing between different public clouds and find the cheapest offering.</p> <p>However, data sharing among clouds becomes a gating factor. If your input files are stored in a Cloud A, transferring them to a Cloud B might be costly due to egress charges applied by every cloud vendor.</p> <p>Neutrix Cloud comes to the rescue.</p> <p>Let’s look at a simple example – using blender tool to render frames in the public cloud.</p> <p>The first step would be to create a blender input file. This is an interactive process, which can be done within a data center – or even on a laptop by a designer.</p> <p>Let’s assume a movie designer has an NFS file system on their on-premises InfiniBox, where the input files are stored. The InfiniBox name is MyMovieIbox and the file system name is NeutrixBlenderDemo.</p> <p><img alt="NFS" data-entity-type="file" data-entity-uuid="cb0a8cf0-a8bd-4295-a1f0-441b19668609" src="https://www.infinidat.com/sites/default/files/uploads/images/NFS_0.png" class="align-center" /></p> <p>First, this file system should be replicated to Neutrix Cloud to make it accessible for the public cloud. Let’s assume the replication link has been already established between the customer’s data center and Neutrix Cloud. The next step would be to create a replica between an on-premises file system and the Neutrix environment:</p> <p><a href="mailto:awsdemo@ims">awsdemo@ims</a>-dc &gt; replica create --source MyMovieIbox --target US-EAST --dataset NeutrixBlenderDemo --sync_interval 20 --description BlenderDemo Create replica for NeutrixBlenderDemo succeeded. Target dataset ID: ds_4b4ae144ee1d4623982784c05bb30335_tgt</p> <p>While the blend file can be pretty big, the changes within it are often very small on each iteration. The very efficient InfiniBox replication mechanism will transfer only the changed blocks between the source and the target to keep them in sync.</p> <p><span>So a replica of the InfiniBox file system in Neutrix Cloud will contain all content of the source file system, including the input data – in close proximity to the public cloud regions.</span></p> <p><img alt="Replication" data-entity-type="file" data-entity-uuid="cf4eba09-0e17-4bef-8ff3-52788a55765e" src="https://www.infinidat.com/sites/default/files/uploads/images/Replication.png" class="align-center" /></p> <p><span>Now, the designer may turn the Neutrix file system into a master copy and the on-premises file system into a target, reversing the replication direction:</span></p> <p><a href="mailto:awsdemo@ims">awsdemo@ims</a>-dc &gt; replica change_role --remote_system MyMovieIbox --remote_dataset NeutrixBlenderDemo Change role replica for NeutrixBlenderDemo succeeded, source is now in the cloud</p> <p> <span>The file system in Neutrix can now be accessed from any adjacent public cloud or clouds. As of today, we have direct integration with Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform, IBM Cloud (SoftLayer), and VMware Cloud on AWS (VMC) environments. We’re willing to consider other public cloud connectivity per request.</span></p> <p>Assuming the designer has an account in each of these clouds, they can access the Neutrix file system using NFS over a dedicated low-latency, high-bandwidth connection from any of these clouds:</p> <p>sudo mkdir -p /mnt/NeutrixBlenderDemo sudo mount $PRIVATE_NFS_ENDPOINT:$PRIVATE_NFS_EXPORT /mnt/NeutrixBlenderDemo</p> <p>They can dynamically shift the rendering workload among public clouds to achieve the lowest cost based on the best spot instances’ pricing available. All their virtual machines will have concurrent read-write access to the same file system from all supported clouds.</p> <p><img alt="Neutrix Cloud" data-entity-type="file" data-entity-uuid="f076da7d-b11f-4ec8-a6d9-f5d4bdebccd7" src="https://www.infinidat.com/sites/default/files/uploads/images/Neutrix%20Cloud.png" class="align-center" /></p> <p>Running a separate blender process on each of the participating virtual machines will distribute the workload and speed up the overall rendering process, while storing all generated frames on the same Neutrix NFS file system.</p> <p>blender -b /mnt/NeutrixBlenderDemo/infinidat_cube.blend -o /mnt/NeutrixBlenderDemo/${CLOUD}_`hostname`/frame_ -s $START -e $END -a</p> <p>The Neutrix file system is being replicated back to the on-premises InfiniBox, and by the time all blender processes finish, the frames will already be available back in the datacenter.</p> <p>At this stage the designer can deprovision compute resources in the cloud and reverse the replication direction once again, making the on-premises file system a source:</p> <p><a href="mailto:awsdemo@ims-dc.infinidat.com">awsdemo@ims-dc.infinidat.com</a> &gt; replica change_role --remote_system MyMovieIbox --remote_dataset NeutrixBlenderDemo Change role replica for NeutrixBlenderDemo succeeded, source is now on-premises</p> <p><img alt="Neutrix Cloud" data-entity-type="file" data-entity-uuid="21ce5577-f451-4733-85af-ea076c09b806" src="https://www.infinidat.com/sites/default/files/uploads/images/All%20Clouds.png" class="align-center" /></p> <p>From here it’s possible to glue all the frames into a movie:</p> <div class="video-embed-field-provider-html-5 video-embed-field-responsive-video"><video controls="" width="100%"><br /><source src="https://s3.amazonaws.com/neutrix-emr-config/InfinidatCube1024x768.mp4" type="video/mp4"></source><br /></video></div> <p><span>This is just one example of how to leverage Neutrix Cloud in a multicloud scenario. I’ll cover <a href="https://www.infinidat.com/blog/running-spark-tasks-on-amazon-elastic-mapreduce-with-neutrix-cloud/">Spark tasks on Amazon Elastic MapReduce</a> and spawning multiple database instances in the next couple of posts…</span></p> <p> </p> </div> <div class="field field--name-field-blog-author field--type-entity-reference field--label-hidden field__item"> <article data-history-node-id="15" role="article" about="/node/15" class="node node--type-person node--view-mode-full"> <h2> <a href="/node/15" rel="bookmark"><span class="field field--name-title field--type-string field--label-hidden">Gregory Touretsky</span> </a> </h2> <div class="node__content"> <div class="clearfix text-formatted field field--name-field-person-bio field--type-text-long field--label-hidden field__item"><p>Gregory Touretsky (@gregnsk) is a Senior Director, Product Management at INFINIDAT. He drives the company’s roadmap around NAS, cloud and containers topics. Before that Gregory was a Solutions Architect with Intel, focused on distributed computing and storage solutions, data sharing and the cloud. He has over twenty years of practical experience with distributed computing and storage. Gregory has an M.S. in Computer Science from Novosibirsk State Technical University and an MBA from Tel-Aviv University.</p> </div> </div> </article> </div> Thu, 15 Nov 2018 14:06:18 +0000 Eric Klinefelter 20 at https://www.infinidat.com