Purity Store

Rely on Safe, Efficient, Storage Capacity

Have the confidence of knowing that your data is stored in a way that ensures data security and availability. Efficiencies in design minimize the infrastructure to create an environment that keeps up with the rate of data growth and adoption.

Data Storage Capacity Optimisation

Get the storage capacity to meet your business needs on-demand, economically. This means you can buy less capacity on-premises or in the cloud than you would otherwise need, without sacrificing the features and flexibility of cloud storage. Growing and shrinking capacity according to real demand enables you to avoid CAPEX spending and build your business on an OPEX* model.

Pure’s Agile Data Services model delivers the industry’s most granular and complete data reduction, which powers Pure’s industry-leading total efficiency—savings that other all-flash vendors can’t touch.

We use five data-reduction technologies:

Pattern Removal: Pattern removal identifies and removes repetitive binary patterns, including zeroes. In addition to capacity savings, pattern removal reduces the volume of data to be processed by the dedupe scanner and compression engine.

512B Aligned, Variable Dedupe: High-performance, inline deduplication operates on a 512-byte aligned, variable block size range from 4K to 32K. Only unique blocks of data are saved on flash, which removes even the duplicates that fixed-block architectures miss. Best of all, you get these savings delivered without the need for any tuning.

Inline Compression: Inline compression reduces data to use less capacity than the original format. Append-only write layout and variable addressing optimise compression savings by removing the wasted space that fixed-block architectures introduce. Combined with deep reduction, compression delivers data reduction of 2x to 4x and is the primary form of data reduction for databases.

Deep Reduction: Data reduction doesn’t stop at inline compression. Additional, heavier-weight compression algorithms are applied post-process to increase the savings on inline-compressed data. Most other all-flash products lack the use of multiple compression algorithms, so they simply miss these savings.

Copy Reduction: Copying data on a Pure Storage® FlashArray™ only involves metadata. Leveraging the data-reduction engine, Purity provides instant pre-deduplicated copies of data for snapshots, clones, replication, and xCopy commands.

Be Vigilant of Data Reduction Claims

Many vendors will claim that their storage efficiency rates—which usually only include thin provisioning and snap savings—are the same as good data reduction. That’s why Pure is clear about our reduction rates: We tell you what our inline and always-on data-reduction technologies are delivering, globally across our entire installed base. And we tell you what the total efficiency numbers look like with thin provisioning included.

How Much Data Reduction Can I Expect?

Data reduction works on a wide variety of applications and data types, but the only way to know how it functions on your data is to try it. The averages below are typical for our most common use cases. Here’s the real live data reduction our customers are seeing today, as average. This is data deduplication and compression.

Environment	Data Reduction
Virtual Server Environments	4-6:1	VMware or Hyper-V, consolidated virtual server environments with mixed applications.
Database Environments	2-4:1	OLTP or OLAP, even databases get surprising amounts of data reduction, most of it via compression.
VDI Environments	5-10:1	Virtual desktops (both persistent and non-persistent) are one of the most reducible workloads in the data centre.

Always Available and Always On

Ensure business continuity by reducing the risk of downtime while keeping mission-critical applications and data online and accessible by those who rely on them.

Designed for flash, RAID-HA protects against the three flash failure modes:

Device Failure: RAID-HA protects against concurrent dual-drive failure and also re-builds around failures automatically within minutes.
Bit Errors: RAID-HA uses independent checksums and dedicated parity to detect and heal around bit-error issues.
Performance Variability: RAID-HA treats performance variability just like a failure, using parity to work around potential bottlenecks to deliver consistent latency.

Intelligent QoS for Continuous Infrastructure Tuning

Always-on QoS prevents workloads from using more than their fair share of resources on the array by efficiently throttling noisy neighbors. And without artificial limits on workloads, it’s always possible to get maximum utilization of the array and you’re assured of full performance for all your workloads.

Some of the ways FlashArray maintains 100% performance include:

Active-Active High Availability: Clustered controller design allows for the complete failure of a controller or any controller component without impacting operations.

Mirrored NV-RAM: Write IOs are persisted to NV-RAM modules, ensuring in-flight writes are protected against power loss and device failure.

Hot-swappable Components: DirectFlash™ modules, NV-RAM modules, and controllers are hot-swappable for continuous operation, even when recovering from a failure.

Stateless Controller Architecture: Simply unplug a failed controller, cable up a new one, and the FlashArray is back to full-availability—without any performance loss.

Always-on Encryption

Always-on simplicity of our “encrypt everything” approach eliminates additional tasks. You don’t need user intervention/key management to provide enterprise-grade, built-in data security. Defend against ransomware with data backup protection, backup software integration, and high-performance recovery. Maintain regulatory compliance and help achieve GDPR compliance with FIPS 140-2 validated encryption, and impact-free, AES-256 data-at-rest encryption. EncryptReduce enables full encryption from FlashArray to host servers with full data reduction.

Non-disruptive Everything

Downtime isn’t an option when your storage array hosts hundreds of applications. That’s why ensuring that storage-management activities are non-disruptive is a critical part of our product design. For instance:

Non-disruptive Capacity Expansion: Add flash capacity online and it becomes instantly available.

Non-disruptive Controller Upgrades: Expand performance with no downtime, even across generations.

Non-disruptive Hardware Replacement: Perform maintenance or replacement of any failed component online.

Non-disruptive Software Upgrades: Easily enable new features and flash-technology advancements with software upgrades that are not only non-disruptive to availability, but also to the performance of applications.

Only Pure delivers everything with 100% performance. Although the IO handling of FlashArray is active-active from all ports on both controllers, the system reserves the performance of one controller on the back end. This enables all maintenance operations to be performed with zero performance loss.

*OPEX treatment is subject to customer’s auditor review