Datera Inc. has become the latest storage startup to move out of stealth, promising an Amazon Web Services-esque cloud storage offering aimed at large enterprises and service providers.
In that respect Datera probably sounds familiar, not least because companies like Ceph, Gluster, SimpliVity and Nutanix all have their own cloud storage platforms. It’s such a busy space in fact that readers may well be wondering what on earth Datera has to offer that could entice anyone.
The answer is “a lot”. Datera’s Elastic Data Fabric product is described as scale-out storage software that turns standard, commodity hardware into a RESTful API-driven, policy-based storage fabric for large-scale clouds. It’s able to automatically compose scale-out storage on standard servers, and it’s already being used in production by a significant number of enterprise and service provider customers. The product is aimed at DevOps and cloud native apps use, runs on x86 servers and also has what Datera calls a “flash-first strategy”. There’s also iSCSI-based native integration with the likes of CloudStack, OpenStack, VMWare Inc.’s vSphere, as well as container platforms like Docker, Kubernetes and Mesos.
“Datera Elastic Data Fabric makes Infrastructure as Code a reality for DevOps environments who have struggled for too long against the cumbersome operations requirements of legacy storage infrastructure,” the company said in a statement. “Now enterprise and service provider clouds can enjoy the operational efficiencies that have given the likes of AWS and Google their competitive edge.”
At the same time as it emerges from stealth, Datera has landed $40 million in funds from a number of big name investors, including Khosla Ventures, Samsung Ventures and well-known Silicon Valley faces Andy Bechtolsheim and Pradeep Sindhu.
“Khosla Ventures invests in category-defining infrastructure companies that automate the agile IT era,” said Vinod Khosla, founder of Khosla Ventures. “Datera has assembled a world-class team to deliver the type of intelligent software we believe is emerging as the foundation of next-generation cloud data centers. [It is] changing the game for anyone needing to deploy cloud scale-out storage.”
Here’s some of the key features of Datera’s Elastic Data Fabric:
- API-first operations provides web-scale automation with full infrastructure programmability.
- Policy based configuration leaves the details to Datera Elastic Data Fabric so customers can optimize price/performance for each application without hand-crafting every LUN.
- Grow-as-you-go model so customers only buy what they need when they need it. The scale-out architecture of Datera Elastic Data Fabric means customers do not need to compromise on performance or capacity.
- Flash-first design delivers high efficiency and low latency (< 1ms) across distributed, diverse storage media types. This maximizes application performance and access density.
- Multi-tenancy and quality of service for cloud-native and traditional workloads optimizes asset utilization and simplifies operations across containers, VMs and bare-metal.
- Heterogeneous component support makes it possible for IT and DevOps to seamlessly scale across many configurations and generations of industry standard x86 servers with multiple storage media types.
SiliconANGLE Media’s Jeff Frick and Stu Miniman were lucky enough to sit down with Datera CEO Marc Fleischmann (pictured, right) during last month’s OCP U.S. Summit, where he explained how the company was focused on reworking and optimizing infrastructures in conceptual and practical forms.
“The basic business problem is that if you look at infrastructure as it moves forward, you basically want to flip it upside-down, so to speak,” he said, adding, “Cloud providers have shown us that’s not how to build infrastructures.”
Fleischmann explained that its the largest enterprises who’re setting the standards today, pointing to cloud data center trends like self-optimizing and self-healing as attributes to look for in the future of infrastructure. Even so, Fleischmann said there are many issues which have yet to be addressed, not leas the problem of hyper-scale efficiency.
“The problem with hyperscale is that at a certain level of scale, it is not efficient anymore,” Fleischmann told theCUBE hosts.
Check out more of Fleischmann’s interview and his plans for Datera in the video below.