“Space: the final frontier. These are the voyages of the starship Enterprise. Its five-year mission: to explore strange new worlds, to seek out new life and new civilizations, to boldly go where no man has gone before.”    Many of us remember these words and the speed of the starship USS Enterprise (NCC-1701) and its crew.   Star Trek was a groundbreaking series in many ways.  One of the most important features of Star Trek was its concept of “warp” speed.   Warp speed is multiples of the speed of light and was how the starship Enterprise jetted around the universe. …This warp technology opened new world’s not previously imaginable… Now, FlexPod has a warp-like technology for your data center called Memory Accelerated FlexPod!

 

 

Like Star Trek and its warp speed, FlexPod® from NetApp® and Cisco is setting a new standard in performance. What would make a new warp-like milestone in performance look like in the context of converged infrastructure? For FlexPod it is composed of industry-leading technologies from NetApp, Cisco and Intel that include NetApp MAX Data software, Cisco UCS servers and Nexus switches, NetApp AFF storage arrays and Intel Optane persistent memory.

 

Without MAX Data software, FlexPod is already a record-setting converged infrastructure:

  1. Cisco UCS servers are the compute in FlexPod and have taken an astonishing 150+ world records to date.
  2. NetApp AFF storage arrays deliver breakthrough SPC-1 performance.  Performance is but one benefit of NetApp AFF arrays.
    1. A few other NetApp AFF arrays benefits minimized footprint with compaction, compression and dedupe, and the ability to unify both SAN and NAS on the same platform.

A new feature of FlexPod that further increases performance is memory accelerated FlexPod powered by MAX Data, a futuristic NetApp server-side storage technology that works with Intel Optane memory to accelerate applications. MAX Data presents Optane persistent memory as a traditional POSIX compliant file system enabling existing Linux based apps to work natively with Optane without rewriting the applications.  MAX Data is not a cache, it’s a tiered filesystem that delivers a persistent storage layer, that combines server side persistent memory tier with an AFF storage tier.  More importantly MAX Data includes integrated data protection and cross-server replication to protect application data against faults.  You might say that memory accelerated FlexPod with MAX Data is exploring “new worlds” of converged infrastructure performance.  FlexPod with MAX Data software brings the performance of persistent memory to many of the applications you run today.  FlexPod can now deliver orders-of-magnitude faster transactions for business applications with MAX Data and Intel Optane DC persistent memory.

 

Like the warp drives of the starship Enterprise which could reach greater-than-the-speed-of-light speeds, the MAX Data difference is that it delivers remarkable performance and data services.   The warp speed is the improvement form the blistering performance of NetApp AFF storage arrays with UCS UCS compute alone!

 

The standard performance threshold for a single database on traditional all-flash performance is about one millisecond. MAX Data performance cuts the latency for those critical I/Os from one millisecond closer to 10 microseconds, more than an order of magnitude improvement!  Cutting down the latency means that the database can drive more transactions, using fewer compute resources and that user queries complete faster.  These benefits, like the Starship Enterprise acceleration to warp 9, translated to getting more done faster and fewer software licenses, as licensed CPUs can be more efficiently utilized.   Now, you can explore new worlds of infrastructure performance at lower software costs!

 

 

Fortunately, as you are getting faster workload performance, you don’t have to make any compromises in your complete data protection. MAX Data includes seamless ONTAP integration that stores snapshot-based backups of your database for quick and reliable backup and recovery, including comprehensive disaster recovery.  Memory accelerated FlexPod powered by MAX Data enables data resilience with minimal impact on performance and provides last-write safety. The NetApp Memory Accelerated Recovery (MAX Recovery) feature enables you to mirror and protect Optane persistent memory in a server and to use Snapshot copies for fast data recovery. The most relevant data is always in the memory tier (Optane), while warmer data is tiered to the storage layer (an ONTAP based AFF system), which leverages all the rich data management capabilities in ONTAP, including high availability, cloning, snapshot copies, backup, and disaster recovery. This MAX Data tiering capability provides enormous near-DRAM capacity. Essentially, you’re getting the best of all worlds: performance of Optane, the capacity and value of flash and industry-leading NetApp data services.

What new performance and software worlds can you explore with FlexPod and what are some of the practical benefits.

    1. Get two times the VM density:
      • Please read the latest Memory Accelerated FlexPod white paper that shows VMs shrinking their resources usage in ½ and still providing significant performance benefits
      • Now you can stretch your VM density and increase the amount VMs per server, lowering your application and VMware licensing costs significantly
      • VMware testing showed that a VM with half of the CPU and DRAM budget achieved 2.4 times the throughput improvement over the more expensive baseline VM (remember the VM was shrunk by ½ and the performance increased!).
    2. Leverage your database licenses:
    3. Power real-time applications and radically accelerate transaction intensive applications.
      • A recent bare metal test showed Memory Accelerated FlexPod capable of 5 times more I/O operations with 25 times less latency than an already record breaking AFF array/UCS server combination!
    4. Read the Cisco blog for a fantastic metaphor to learn about memory accelerated FlexPod!

FlexPod converged infrastructure provides an ideal platform to deliver memory acceleration by taking groundbreaking technologies and making them safe and consumable for enterprise organizations.  The reasons are explained in two sister blogs, here and here.  To summarize, FlexPod has become a standard in converged infrastructure similar to what electricity has become.    This FlexPod standard can handle both large, mission-critical applications and edge workloads. Exciting innovations like AI/ML or EPIC and MEDITECH EHR healthcare solutions plug into the same industry-leading standardized components, like NetApp® FAS and AFF storage arrays, Cisco Unified Computing System™ (Cisco UCS®) servers, and Cisco Nexus® switches. FlexPod is constantly evolving and modernizing adding new Cisco Validated Designs (CVDs) and NetApp Validated Architectures (NVAs) to its family of 170+ standard solutions, ensuring the performance of modern enterprise applications that run the largest and most complex, mission-critical enterprises like Oracle RAC, Microsoft SQL Server, and SAP.  The FlexPod standard is also continuously adding new applications, workloads, and technologies (like MAX Data software) to ensure that customers that they have a long roadmap of both application and technology improvements. FlexPod is unique in that this assurance has been built over years, over thousands of customers, hundreds of applications and countless technologies.  If you want a rock-solid, reliable, thoroughly tested vehicle to run your MAX Data software, FlexPod is the ticket.

 

So, to learn more about the MAX Data for the FlexPod standard, “beam” over to the NetApp(327), Cisco(1427) or Intel(1039) booths (add booth numbers at VMWorld).

 

For more information, please visit:

Bruno Messina

Bruno Messina joined NetApp in 2018 and works in product marketing for FlexPod. His previous experience includes a career in product marketing of UCS servers for Cisco Systems and Solaris server marketing and competitive analysis at both Oracle and Sun Microsystems, where he joined in 2000. Bruno spent ten years in various roles of competitive analysis and product management at Sun Microsystems, leading analysis in both the workgroup and enterprise servers. Prior to Sun Microsystems, Bruno spent time finishing his MBA education and worked for two years at Cadence working on product marketing for both board-level and board timing tools. Bruno holds both a BSEE and MBA from Rensselaer Polytechnic Institute in Troy, N.Y..