Advanced Computing in the Age of AI | Friday, March 29, 2024

Excelero Launches From Stealth With NVMesh 1.1 

March 8 -- Excelero, a disruptor in software-defined block storage, is launching out of stealth today and announcing version 1.1 of its NVMesh Server SAN software for exceptional Flash performance for web and enterprise applications at any scale. Funded in part by Fusion-io's founder David Flynn, Excelero enables enterprises to deploy next-generation data centers modeled after the standard hardware-based hyperscale IT architectures of technology giants like Facebook, Amazon and Google. With NVMesh, customers can leverage the full performance of NVMe flash at data center scale, with predictable application performance and astounding price/performance levels for even the most demanding applications.

NVMe holds the promise of unlimited performance for hyperscale environments, but current inefficient storage and application architectures suffer from low utilization rates – as little as 25%.  In nearly three years of development, Excelero has achieved 11 patents or patent-pending innovations for the ability to virtualize the NVMe devices and unify the capacity into a single pool of high-performance storage - to flexibly provision storage to applications as they need it.

NVMesh is a 100% software-only solution that enables customers to benefit from the performance of local flash with the convenience of centralized storage while avoiding proprietary hardware lock-in and reducing overall storage TCO. Designed to meet storage requirements for applications of any scale, NVMesh is the only NVMe sharing technology that scales performance linearly at near 100% efficiency by shifting data services from centralized CPU to complete client-side distribution. Its flexibility allows for both physically converged or disaggregated deployments to create a virtual, distributed non-volatile array. In early use by GE, NASA Ames, and other household-name enterprises, NVMesh is well suited for Internet of Things (IoT), Big Data analytics and Artificial Intelligence as well as web-scale enterprise loads.

"NVMe storage is a game changer for the machine learning industry. For the Industrial Internet of Things, it enables us to process more data faster -- but challenges exist in using NVMe at scale with conventional storage architectures," said Darren Haas, head of cloud engineering at GE. "Excelero has created a unique software-only architecture that allows the industry to use NVMe to the full potential, at cloud scale."

"Excelero's approach makes data locality irrelevant, and that's a huge technology advantage," said Howard Marks, founder and chief scientist of DeepStorage, LLC, a hands-on IT analyst and testing practice. "With the NVMesh RDDA protocol, the latency to a remote device is so small that the application won't notice the difference between local and remote. It enables applications to effectively share NVMe, which will only become more significant as new non-volatile memory technologies like 3D XPoint come to market."

With today's general availability of NVMesh 1.1, enterprises can design Server SAN infrastructures for the most demanding enterprise and cloud-scale applications, leveraging standard servers and multiple tiers of flash. Excelero's patented Remote Direct Drive Access (RDDA) allows for true convergence by logically disaggregating storage from application servers – keeping valuable CPU available for applications.

NVMesh adds a mere 5µs latency (mostly from network latency) over local NVMe drive latency, allowing customers to leverage full performance of their hardware at data center scale. This enabled NASA Ames, for example, to produce 140GB/s of aggregated throughput and 30M random 4k IOPS. This was achieved on 128 x86 servers. Observed average latency was below 200µs and the CPU overhead was negligible.

To future proof deployments and ultimate flexibility, NVMesh also supports the NVMe over Fabrics (NVMf) transport, and is an active member of NVMe Express consortium that is helping to shape future refinements of this standard.  The company is an ardent supporter of other open initiatives, such as LinkedIn's Open19 and Facebook's Open Compute Project, that can assist enterprises in transitioning infrastructures for maximum efficiency and flexibility.

"The innovation in Flash storage is impressive: the cost per IOP is at an all-time low. But customers cannot leverage the full performance at scale. As a result, they need to compromise on utilization, which takes away their competitive advantage," said Lior Gal, CEO and co-founder of Excelero.  "Excelero has worked diligently to enable customers to leverage the full potential of high-performance flash at scale. With NVMesh, customers achieve much closer to 100% capacity utilization without giving up local performance characteristics effectively, which lowers the storage cost dramatically."


Source: Excelero

EnterpriseAI