StorPool Storage adds file storage, NVMe/TCP and AWS support in 20th version of their storage platform

The Bulgaria-based scale-out provider has always specialized in providing high-value around a limited number of use cases, but with this release, is expanding the use cases that they support.

Boyan Ivanov, CEO of StorPool Storage

StorPool Storage, a Bulgarian scale-out block storage provider, has announced the official release of the 20th major version of their StorPool Storage platform.  The new capabilities include the addition of support for the NVMe/TCP protocol, availability on AWS, which is targeted primarily at large customers with complex use cases, and new support for running NFS Servers inside StorPool storage clusters for specific use cases.

StorPool is headquartered in Bulgaria, and was designed to replace million-dollar Tier One boxes with software on standard x86 servers, pooling the storage on these servers to create a single shared block storage pool. While the scale-out storage market has taken much longer to develop than many had foreseen, including some failed early entrants, and is now being confronted with a more negative economic environment, StorPool thinks these factors favor a company like themselves.

“We are in a very strong position because we are one of the very few profitable storage companies, and when capital is not free any more, that helps us,” said Boyan Ivanov, CEO of StorPool Storage. “Our brand is not that visible and we are not raising money, but we are building our business. We are a European company that is organically grown. Today, the increasing interest rates and inflation delays investment decisions and increases uncertainty. The Ukraine war has meant that some deals have been put on hold. But we expect more market share because we are resilient, and we are seeing more demand from Global 2000 customers.”

Ivanov emphasized that the use cases they support are focused around extreme reliability and low latency.

“When we designed this, it became an issue of either having a solution that can cover many use cases, versus being the best in the world at those we do cover,” Ivanov said. “We are at the forefront of use cases that we target, but we don’t do all use cases.”

With this release, however, they have expanded in this respect. StorPool has introduced support for running highly available NFS Servers inside StorPool storage clusters for specific use cases. These include video rendering, video editing and heavily loaded web applications, as well as moderate-load use cases like configuration files, scripts, images, email hosting, and supporting cloud platform operations. The cumulative provisioned storage of all shares exposed from each NFS Server can be up to 50 TB. NFS file storage on StorPool is not suitable for IOPS-intensive file workloads like virtual disks for virtual machines.

“With the NFS File Storage, in order to address more Global 2000 complex use cases, we ‘legoized’ it and did some product work to address the needs of those with very fast NFS systems in terms of throughout,” Ivanov said. “It can also scale down to second and third tier storage systems, so expands the capability of our data management platform. It’s not meant to compete with 10 TB cases, but will address maybe half the file cases out there.”

StorPool now also supports NVMe/TCP [NVMe over Fabrics, TCP transport] the next-generation block storage protocol that leverages the TCP/IP communication protocols, using the standard NVMe/TCP initiators available in VMware vSphere, Linux-based hypervisors, container nodes, and bare-metal hosts.

“Historically we have competed against competitors who did NVMe over Fabrics, but who didn’t have the data management features we did, and that customers want, and we also competed with them on performance and scalability,” Ivanov said. “We are now adding in TCP which has superseded NVME over Fabrics in terms of performance.”

Finally, StorPool Storage can now be deployed on sets of three or more i3en.metal instances in AWS, to deliver 1.3M+ balanced random read/write IOPS to EC2 r5n and other compatible compute instances like m5n, c6i, and r6i. Ivanov stressed that this isn’t something that was done to allow customers to just port their existing solutions to the AWS, which is still less than 50% of the market in terms of dollars spent.

“The interesting part of this is that while many companies port to the public cloud and advertise it, we have a different story,” he said. “We have very large customers who don’t live on the public cloud, because it’s not good enough for them. This isn’t the mass market. It’s the top 1% of customers , who found that what they wanted to be was impossible to do in the public cloud before from a cost/performance ratio perspective. AWS actually came to us to solve this problem.”