From day one, the mission behind FlashBlade has been to provide the engineers and scientists of tomorrow with an all-flash platform that enables innovation, insight and discovery across industries and segments.
We had key insights into an expanding data set, and we knew that modern computing frameworks had become much more distributed, giving rise to increasingly complex, high-performance analytics and valuable data. This was the primary motivator behind our mission to build an all-flash platform that was big, fast and easy to use — a product purpose-built for effortless high-performance across all industries and segments.
Today, thanks to the amazing work by the FlashBlade team we are announcing our next milestone. Three and a half years in the making, we are incredibly proud to announce that FlashBlade is now in General Availability (GA).
For the past six months we’ve been shipping FlashBlade in Directed Availability (DA) to ensure we hardened the product, perfected our internal processes and learned first-hand from customers the incredible range of uses they anticipated for FlashBlade.
We’ve had incredible success with customers using FlashBlade in production across a number of industries and widely varied workloads. We have customers using FlashBlade in production across the globe, including North and South America, across Europe and into the Asia-Pacific region.
In particular, we have learned that FlashBlade excels in analytics — both classic application suites like Oracle RAC running data warehouses, and the emerging Big Data 2.0 stacks running large scale Apache Spark clusters for iterative queries, machine learning or SQL query processing.
We are now at an inflection point for many industries. Distributed programming paradigms, also known as “the new stack,” are approaching critical mass. Arguably, the first industry that jumped on this architecture were hyper-scale cloud companies. These players established this approach as the core architecture available that could meaningfully cope with massive growth and scale.
We observed, together with some of the leading cloud companies and other industry pioneers, that FlashBlade meaningfully accelerated insights, discoveries and analysis of data in ways that previously would have been complicated or unattainable. It has been an amazing journey, together with our customers, and we treasure the relationships and improvements we’ve forged.
Along this journey, we found the new stack is far more common, maybe even approaching standard, across more widely varied industries than we previously imagine. FlashBlade is currently deployed across a range of use cases, and we’ve found significant technology and deployment similarities across them.
We’ve observed that analyzing genomes for clinical diagnostics is, from a workload perspective, very similar to the way geophysicists use clusters of computers to perform geophysical mapping in oil and gas. Simultaneously, we realized how similar this flow is to the way a data scientist uses Apache Spark for business analytics or to create scenarios for machine learning. This in turn is very similar to the way software houses build, test and integrate agile software at scale. The list goes on and on.
What became obvious to us is that these patterns repeat themselves, particularly in larger cloud and SaaS environments. We have deployed FlashBlade into clouds for core banking and insurance, as well as clouds for personal entertainment and gaming. Across all these industries we observe the ways in which FlashBlade’s core design principles and modern approach to storage accelerate value creation from data.
Our key takeaway from all of this? We are in the midst of a core technology transition that is happening here and now. And FlashBlade is best positioned to not only take advantage of that shift, but to lead it.
We are entering into a period in which a major redesign of application architectures is underway, where we leave client/server and n-tier architectures and drive towards distributed computing and vertical scale. This transition changes fundamental requirements for everything in the underlying infrastructure. Our most notable successes to date are, in essence, the same regardless of industry — enabling the transition to a modern application infrastructure with storage that can scale effortlessly, that is incredibly dense and energy efficient, and that is easy to use and never requires tuning or complex setups even for the most complex demands.
What’s really enlightening is what FlashBlade enables our customers to achieve. The UC Berkeley Genetics and Genomics department, for example, is conducting breakthrough research on genomics. They are using FlashBlade to conduct complex analysis and run data-intensive visualization in 3D. The results have been astounding — queries on Apache Spark that took 12 hours now take just 30 minutes. Ultimately, UC Berkeley is conducting this research to adjust clinical diagnostics in real-time, which in turn are improving patient outcomes and saving lives.
On the sports and entertainment front, the Mercedes-AMG Petronas Motorsport Formula 1 team is using FlashBlade to gain real-time insight into what’s happening on the track to win races. The team installs a FlashArray trackside, acting as a mobile data center to store and process data from nearly 250 sensors across their cars. Meanwhile, back at home base, the team uses FlashBlade to run simulations and testing on new car design. This is a prime example of how analytics provide a competitive edge in professional sports.
Customer feedback thus far has been both motivating and inspiring:
Do you have a use for FlashBlade? Would you like to know more about what we learned, or are you planning an aggressive foray into analytics? Are you a part of a data science team, or an IT professional? Do you work as a DBA or do you care about design patterns for modern infrastructure stacks? If so, talk to Pure. Let’s see what we can build together to make the future better!