In this storage vendor profile, we look at Pure Storage, which among all the big six storage players comes closest to having been a startup in recent times.
This means that Pure Storage joined the market unencumbered by legacy technology and benefits from late entrant advantage. That said, it has barely turned a profit in a decade or so of existence, but that doesn’t seem to be stopping it.
What we find in Pure Storage is an array player that’s a flash specialist, with its own array/drive technology that can credibly claim much better flash storage density than its rivals. It targets rapid access use cases, not just for structured mission critical data, but also in unstructured data workloads. In addition, it possesses a coherent container-focused hybrid cloud strategy and well-developed consumption models of purchasing.
In this article, we look at Pure Storage, its origins, its key storage products and its approach to the cloud, containers and consumption models of storage purchasing.
Where did Pure Storage come from?
Every other big six storage player has its origins pre-2000, but not Pure Storage, which is a product of the flash era. It was initially formed in 2009 under the code name Os76 Inc in the offices of private equity firm Sutter Hill Ventures by John “Coz” Colgrave and former Yahoo! CTO John Hayes.
The company worked in stealth until 2010 and went public in 2015, with venture capital injections along the way. Early on, the company’s founders set to work on maximising capacity and endurance from the then emerging multi-level cell (MLC) flash media.
Also core to the Pure story is that it initially used proprietary flash drives, but had introduced its own DirectFlash Modules by 2017 that were theoretically able to go to much higher capacities than off-the-shelf solid state because they worked in tandem with array-level systems.
Pure Storage has strongly espoused the view that these capacity increases will translate to lower per-GB prices for its flash than is possible with spinning disk and the consequent obsolescence of the latter.
How does Pure Storage rank against other storage players?
By the second quarter of 2021, IDC ranked Pure Storage last among the big six storage array makers with market share of 4.1% – however, that was in the same ball park as Hitachi and IBM (both just under 5%). The big dog is still Dell EMC on 27% market share, with HPE (11%), NetApp (10%) and Huawei (9%) behind that. Dell was ranked 34 in the 2023 Fortune 500.
Pure Storage revenue was $440m in 2015, the first year it went public, and that figure had reached $2.76bn by 2023. But during that period, it only made a net profit during three quarters by the time of writing.
What are Pure Storage’s key storage products?
Core to Pure’s flash storage hardware products are its DirectFlash Modules. Currently, these come in at 75TB capacity, with capacity of 300TB planned for 2026. DirectFlash modules come in triple-level cell (TLC) for performance applications and quad-level cell (QLC) for capacity.
The modules are designed to communicate directly with the array, so media management functions – such as mapping data to physical media, garbage collection, wear levelling, and so on – are built into system-wide functionality. That’s unlike commodity solid state drives that do this at individual-drive level with dedicated media, such as costly DRAM, on the drive.
FlashArray is Pure’s product line aimed at primarily at structured data and block storage, although file storage is also possible.
FlashArray//X is aimed at mission-critical workloads up to 3.3PB; FlashArray//XL, which is the extended footprint version with capacities up to 5.5PB per 5U appliance; FlashArray//C targets “business-critical” workloads – i.e., less performance-hungry – up to 8.9PB; while FlashArray//E provides QLC capacity options with lower performance up to 4PB.
FlashBlade is Pure’s fast file and object storage family, which provides rapid access to large-capacity storage. It is aimed at what have traditionally been secondary storage use cases – backup, archive, analytics datastores – but which have evolved to require rapid input/output (I/O) for analysis, recovery and so on.
FlashBlade//S comes in two models, with the S200 emphasising capacity, with data reduction, while the S500 goes for performance. FlashBlade//E provides capacity-focussed QLC media aimed at unstructured data use cases where performance is less important.
What markets and workloads does Pure Storage target?
Pure Storage aims at datacentre workloads that range from the most performance-hungry, mission-critical operations to workloads that would once have been deemed secondary, such as backups and less frequently used data. Pure’s argument is that the fast access speeds of flash storage are needed in these use cases because organisations need to, for example, recover data quickly after a ransomware event, or transfer and access data rapidly during AI-type analytics.
How does the cloud fit Pure Storage strategy?
Pure Storage is well-oriented towards hybrid cloud operations and modern cloud-native applications, based on the idea that customers will want to work across all locations with the ability to operate between cloud and on-premise.
It has Pure Cloud Block Store, an enterprise-grade block storage service delivered natively in the public cloud, with access to the same protocols, snapshots and replication capabilities of Pure’s storage software.
At the same time, it has Portworx – Pure’s biggest acquisition, bought in 2020 for $370m – which is a cloud-native data management platform. Portworx is centred on provision of persistent storage and data protection for containerised applications across cloud and on-premise, but also data services such as databases, event handling and messaging platforms such as Cassandra, Kafka, PostgreSQL, RabbitMQ, Redis and Zookeeper.
Pure Storage also has Fusion, a cloud-like control layer that manages storage across on-site, cloud and colocation options. Fusion provides a cloud-resident control plane on top of existing Pure-as-a-Service and on-site storage deployments. V Admins can manage Fusion via application programming interfaces (APIs) to handle deployment, workload placement and mobility and rebalancing.
What is the Pure Storage container strategy?
As mentioned above, Pure bought Portworx in 2020 and it forms the core of its container strategy. Portworx allows orchestration of Kubernetes and other container platforms across cloud and on-prem locations, with persistent storage and data protection and provisioning and management of container clusters.
What consumption models of purchasing does Pure Storage offer?
Pure Storage offers consumption models of purchasing its Evergreen portfolio, which includes: Evergreen Forever, which is buy-outright with lifetime upgrades; Evergreen Flex, where hardware is paid for and capacity is pay-as-you-go; and Evergreen One, which is just pay-as-you-go.
