Pure Storage Makes Huge Bet On Hybrid On-Prem, Cloud Data Management

All-flash storage technology developer Pure Storage on Monday signaled its intention to be a force in the multi-cloud data management business with the introduction of two initiatives to its high-performance storage capabilities natively to the cloud.

Pure Storage introduced Cloud Block Store for AWS, a technology that allows the Mountain View, Calif.-based company's storage software to run natively on AWS for managing block data across on-premises and Amazon Web Services cloud environments.

The company also introduced CloudSnap for AWS, a technology for natively sending data backups from its flagship FlashArray all-flash arrays to AWS for cloud-based data protection and recovery.

[Related: CRN Exclusive: Pure Storage CEO Giancarlo On A.I., Innovation, And M&A Strategy]

The result is a whole new way of thinking at Pure Storage, said Matt Kixmoeller, vice president of strategy.

"In many ways, this is the start of a second chapter for Pure," Kixmoeller told CRN. "It's the first time we partnered with AWS on a set of cloud services."

Pure Storage believes in a data-centric architecture that offers high-performance shared storage, provides capacity delivered on demand, and enables a hybrid world where the data can sit on-premises and in the cloud, Kixmoeller said.

Pure Storage has been building its cloud capabilities for well over a year. The company last year at its Pure//Accelerate conference introduced a hybrid cloud solution for AWS, and is offering customers direct connections to the Microsoft Azure and Amazon Web Services clouds.

Starting now, and over the next few months, Pure Storage will be getting closer to AWS in a cloud-native fashion.

The company on Monday unveiled Cloud Block Store for AWS that allows the company's software to run natively on the AWS cloud and offer the full range of data services, Kixmoeller said.

"This is industrial-strength block storage," he said. "Often, block storage on the cloud doesn't have the needed services. But we're not just taking our on-premises software, putting it in a virtual machine, and running it in the cloud. We've optimized our software to run on the cloud. When the software runs on-prem, it talks to our hardware and NAND memory. When it's running on AWS, it talks to Amazon as the hardware platform."

Cloud Block Store for AWS, on the other hand, includes deduplication, compression, and thin provisioning, Kixmoeller said. Thin provisioning is the ability to provision additional capacity as needed and not have to do so in advance.

"Cloud applications are normally thick-provisioned, so you have to pay more for compute and capacity," he said. "We're more efficient."

Cloud Block Store for AWS also includes full data resiliency services, including active-active resiliency between multiple availability zones, along with a full suite of other data protection capabilities, he said.

The Pure1 storage operating system also extends across on-premises and cloud architectures, Kixmoeller said. "You can use VMware's vRealize on-premises to spin up Cloud Block Store on Amazon and replicate data to the cloud," he said. "You can spin up an instance in a half-hour, then turn it off, all automated via APIs."

The second major change at Pure Storage is a new modernization of data protection that takes advantage of on-premises flash-based storage for performance combined with the cloud for secondary storage or disaster recovery.

This is becoming a big issue for a lot of customers who are actually buying Pure Storage's high-performance FlashBlade arrays for data backups, Kixmoeller said.

"We were confused about this at first," he said. "But we saw people needed fast restores. Restore speeds has become a real pain."

Pure Storage is modernizing its data protection capabilities in two ways.

First, the company introduced CloudSnap which takes snapshots of data on FlashArray arrays to move to another FlashArray or FlashBlade for fast access, and then to Amazon for long-term data protection, Kixmoeller said. "Clients can use the backup copy to recover data to a Cloud Block Store instance if needed," he said.

The second is taking advantage of Pure Storage's acquisition this Summer of StorReduce, a small developer of software for deduplicating, managing and migrating data.

Using the StorReduce technology, which provides the industry's first cloud-based deduplication engine, customers can back data from both Pure Storage and non-Pure Storage arrays to the cloud without modifying their existing backup software or data protection environment, Kixmoeller said.

"Just point your existing environment to StorReduce, and it takes advantage of both flash storage and the cloud," he said.

Cloud Block Store for AWS, CloudSnap, and StorReduce are some pretty cool and unique offerings, said Cindy Kennedy, president of the western region of Converge Technology Partners, a Greenwood Village, Colo.-based solution provider and long-time Pure Storage channel partner.

"Our customers are all going to the cloud," Kennedy told CRN. "They all have a hybrid strategy. Maybe a few will go all-cloud. But they're nearly all looking at a hybrid on-premises and cloud strategy."

Amazon Web Services is the public cloud most of Converge's clients are working with, Kennedy said. However, there was some hesitation on the part of clients who also like Pure Storage because they would lose many of the services the vendor's software offered, she said.

For instance, she said, Pure Storage was the first to make flash storage economical by introducing data reduction in flash storage for production environments, and not just for data protection.

"This is the same strategy Pure has been doing all along: Make it simple," she said. "If customers are going to the cloud, why should they need to focus on the IT side? The future is artificial intelligence and IoT, and not in managing storage."

"And the ability to automate snapshots in the cloud gives customers the ability to think about storage like I think of my iPhone on the iCloud," she said. "I don't care where the data is coming from. I just want simplistic data management and access to my data."

Pure Storage's new cloud-focused data management push comes at the right time for customers, said Rich Falcone, president of sales at Computer Design & Integration, or CDI, a New York-based solution provider and managed services provider and long-time channel partner of Pure Storage.

"Many of our Pure Storage clients are also AWS clients," Falcone told CRN. "They've been chomping at the bit to leverage both."

Falcone termed what Pure Storage is doing with AWS as "hybrid hardware," which he said was his term for letting software running on on-premises hardware natively leverage data on and communicates with AWS.

"Pure Storage is doing it at the block level and the snap level with single-pane-of-glass management," he said.

There are a couple of important use cases that Cloud Block Store for AWS will impact, Falcone said. By running Pure Storage's FlashArray in the cloud for a secondary site, customers can use it for bursting data, disaster recovery, and test and development, he said.

Clients will also be able to utilize CloudSnap to store their data snapshots in a bucket in AWS S3 in an incredibly efficient way without the need to buy more flash storage for their on-premises architectures, Falcone said.

"Over a year or two of protecting data, clients can end up with a ton of physical disks," he said. "Some of that data will be deleted and archived, but clients are still stuck with all those disks. With CloudSnap, they can take advantage of Amazon S3 for efficient storage, and then use Cloud Block Store to natively restore data on Amazon, and then bring it back on-prem if necessary."

CloudSnap is already available, Kixmoeller said. Cloud Block Store for AWS and StorReduce are now in limited beta testing. StorReduce is slated for general availability in the first half of 2019, while Cloud Block Store is expected to be in general availability in mid-2019, he said.