Flash storage systems maker Pure Storage wants everybody to know that when they see the term “shared accelerated storage” being used to describe new-gen flash arrays, they created the idea first.
The Mountain View, Calif.-based company, which this week is hosting its annual Accelerate users conference at the Bill Graham Auditorium in San Francisco, has expanded its FlashArray product line to add—you guessed it—all-NVMe Shared Accelerated Storage for every workload.
NVME stands for non-volatile memory express, which is an open logical device interface specification for accessing non-volatile storage media attached via a PCI Express (PCIe) bus. PCI Express is a high-speed serial computer expansion bus standard which has replaced the older PCI, PCI-X, and AGP bus standards.
Speeds Up Everything in an On-Premises IT System
The new Pure product line, called FlashArray//X, uses all of the above to speed up everything in an on-premises IT system: databases, virtualized and containerized environments, test/dev initiatives and web-scale applications, at no additional cost over FlashArray//M, based on effective capacity, Pure said.
Storage and storage networking has long been the big bottleneck in causing latency in IT systems, and this is all designed to make that slowness a thing of the past.
“In a lot of ways, it feels like eight years ago when flash was perceived as this exotic thing and we came in and kind of democratized it,” Pure Storage Vice-President of Strategy Matt Kixmoeller told eWEEK. “Our goal is to do the same with NVME.”
The all-NVMe FlashArray//X lineup features five configurations, from the NVMe-ready //X10 for small application deployments to the all-NVMe //X90, Pure’s densest, fastest FlashArray to-date. FlashArray//X represents a new generation of Shared Accelerated Storage solutions with the potential to unite SAN and DAS into a single, consolidated, shared and fundamentally more efficient data-centric architecture.
Companies Can Continue to Use Older Arrays
So one of these systems, Kixmoeller said, can help companies maintain their investments in legacy arrays, because Pure’s NVME systems can make them all look the same in one controller.
Featuring up to 3PBs of effective capacity in a 6U package, the //X90 delivers latency as low as 250 µs. An updated version of the flash-only Purity Operating Environment, version 5.1, is available as a non-disruptive upgrade for all FlashArray//M and FlashArray//X customers and features up to 20 percent better data reduction than past versions, Kixmoeller said.
A data-centric architecture enables diskless, stateless compute to flexibly connect to a fast, reliable and dense pool of Shared Accelerated Storage via fast networking, Kixmoeller said. It decouples storage and compute, which enables them to scale and innovate independently while maintaining performance and efficiency, he said.
FlashArray//X, like all FlashArrays before it, is built on Pure’s Evergreen Storage architecture, which ensures that customer investments are protected with non-disruptive upgrades, no planned downtime, and no data migrations, Kixmoeller said. This enables existing FlashArray customers to upgrade non-disruptively to FlashArray//X and experience all-NVMe Shared Accelerated Storage at any time.
Here’s a rundown of Pure’s other new products:
Expanded FlashArray//X lineup:
- //X90: Pure’s densest, fastest array to date. 3 PBs with effective capacity in 6U, up to 2X faster than the market-defining FlashArray//M and lowest cost array per GB.
- //X70 and //X50: 100 percent NVMe arrays that scale to 1.3 PBs and 650 TBs of effective storage respectively.
- //X20 and //X10: Entry-level NVMe-ready arrays for smaller capacity and remote office needs. These come configured with SATA flash by default, but feature controllers fully-capable of supporting NVMe DirectFlash modules out-of-the-box.
Pure1 management software:
- Purity 5.1, the latest release of Pure’s fully-parallel flagship storage operating environment, features enhancements to compression, delivering up to a 20 percent improvement in overall data reduction efficiency, especially for compression-centric application workloads. This non-disruptive software update will be available for all FlashArray//M and FlashArray//X customers.
- New Pure1VM Analytics enables full-stack visibility into VMware vSphere environments. By integrating natively with vCenter, performance issues can be easily visually-correlated from VM → Server → Datastore → Network → Volume → Array.
- The expanded Pure1 Workload Planner, powered by the Meta AI engine,helps customers optimize their environments by better understanding the impact of workload changes and also enables FlashArray users to easily explore “what if” scenarios for hardware upgrades.
- The Pure1 Global Dashboard has been expanded, enabling fleet-wide visibility of all FlashArray and FlashBlade arrays globally, available on both the Pure1 SaaS portal as well as on the Pure1 Mobile app.
FlashArray//X is available now; new arrays and upgrades will be shipping in early June. Purity 5.1 is available now, the company said. Pure1 VM Analytics, Workload Planner and Global Dashboard updates will be made available soon. For more information, go here.