This post was originally published on this siteAt Pure//Accelerate 2018 Cody Hosterman and I delivered a session on “Moving Data Between Cloud and On-Premises Virtualized Environments“....
When storage vendors release their new product offerings, the focus tends to be on traditional storage performance metrics like IOPS, throughput, latency, I/O size, etc. These metrics are typically generated by synthetic tools and difficult to translate into real world capability.
As a SQL Server database architect the question always comes down to how does my storage array perform relative to how much actual work is accomplished by the database engine.
We recently launched Pure Storage FlashArray//X, which is powered by NVMe using our proprietary DirectFlashTM technology. Utilizing NVMe in conjunction with our Purity Operating Environment allows native access to the flash, getting rid of traditional barriers such as the SSD drive firmware and controller.
Our Purity Operating Environment now delivers global flash management, allocation, and garbage collection at the system level. This eliminates redundant and inefficient processing layers in the SSD that take time such as flash translation, IO scheduling, wear leveling, over provisioning, and garbage collection.
These innovative technologies result in increased productivity for write intensive highly concurrent SQL Server workloads.
The following two SQL Server metrics were used to measure SQL Server database productivity with a real world OLTP type workload (read/write ratio 20/80) between our FlashArray//M and FlashArray//X products:
Page IO Latch Average Wait Time: Time it takes to retrieve a page from nonvolatile storage (flash) into volatile storage (DRAM).
Transactions per Second: Operations that create a single unit of work.
The efficiencies of FlashArray//X translate into a more productive SQL Server environment with real world transaction results. If you need fast performance for write intensive highly concurrent SQL Server workloads, FlashArray//X will meet your needs, improving on the already stellar performance of our FlashArray//M product.
Contact us to find out more!