Top 8 Reasons to Choose Pavilion Data’s Hyperparallel Flash Array

With listels all the rage these days, I thought it would be interesting to summarize the key points from 2019’s previous blogs and cover a few more in this BuzzFeed friendly format.  While it’s not exactly my style to boil deep technical insight into a top 8 listel, it is good to be concise.

  • Customers are voting with their pocketbooks. 

Since the beginning of 2019, Pavilion Data has shipped dozens of systems to customers around the globe and closed multiple 7-figure deals.  Most deal with several things in common.  First, a never-ending thirst for speed that only NVMe can quench.  Second, analytics workloads.  While these range from Artificial Intelligence to Machine Learning and Deep Learning, they span applications including IBM Spectrum Scale, various flavors of NoSQL Databases and the Pivotal™ Greenplum Database and Pivotal™ Cloud Foundry as well as Dense K8s and VM environments currently running on Direct-Attached Storage (DAS).  Most are upgrades to traditional NVMe scale-out where DAS has reached a point of diminishing returns, and proprietary AFAs can’t deliver on NVMe performance at scale.

  • Node Recovery.

DAS Aggravation is a real thing.  As drive makers continue to increase the size of an NVMe SSD, DAS implementations take even longer to recover from a node failure impacting application performance.  With SWARM recovery, Pavilion Data’s array has zero impact on application performance and can recover data to a new drive in as little as 5 minutes per TB.

  • OpenChoice Storage™.

In his blog, “3 Questions to ask your big storage vendor,” our CEO, Gurpreet Singh, summarizes Pavilion Data’s disruptive business model where flexibility meets investment protection.  Customers benefit from reuse of existing standard 2.5” NVMe SSDs or by procuring the latest SSDs directly from drive makers, dramatically reducing procurement costs and future-proofing the platform.

The human eye can blink twice in the latency it takes for a 128KB block to move from a host, across an RDMA-based NVMe Fabric, through the Pavilion array to an OPENCHOICE of NVMe SSDs in a RAID-6 volume.

Jeff Sosa tackles one “little” metric that should be part of every conversation with Big Storage vendors.  How many Rack Units does it take for your preferred flash array to achieve 90GB/sec write, 120GB/sec read, 20M IOPS at 40 microseconds of latency?  If it is even possible, the best number on the market today is 80 RU.  For Pavilion Data – 4 RU.

  • NVMe-oF is HOT!

Of course, we are biased, but with announcements from every significant storage system provider and a few startups, NVMe-oF might be the storage technology of the year for 2019.  We tackle the DirectFlash™ Fabric announcement by Pure Storage head-on in two blogs – Man In The Arena Pavilion and Award Pure a Participation Trophy.  Both worthy of a read, or reread.  Feedback coming from customers (see item 1 above) show that traditional AFAs just don’t cut it.  In fact, in one transaction, Pure required 14 systems to meet the performance objectives that Pavilion Data’s Hyper-Parallel Flash Array handled with just two systems.

  • We are a recognized leader.

In 2018, Gartner gave us “cool vendor” status.  This year, there were only three startups referenced in G00364795 – Competitive Landscape: Solid-State Arrays, Worldwide.  And just last week, we were an IDC Innovator in NVMe over TCP.

  • Procurement.

We haven’t mentioned this yet, but our system is now available through Dell Technologies.  That’s right!  In the past few weeks, Dell has loaded Pavilion SKUs into its ordering systems.  You can procure both the product and support directly through Dell and its channel partners.  The Dell relationship, in addition to several vertically-focused resellers, makes it easier than ever to get the world’s fastest NVMe-oF system.

Of course, we are far from done.  Stay tuned to this blog and follow us on social media at and as we continue to enhance our technology and go to market.