PEAK:AIO extends its lead within the AI storage market, introduces new breakthrough GPU storage performance reducing the cost of ultra-performance AI data storage by 70%
PEAK:AIO, a provider of a transformative software defined storage platform designed to meet the needs of artificial intelligence (AI) applications, announced the launch of its latest version of the AI Data Server, which doubles previous performance, reaching a breakthrough 80GB/sec from a single storage node as small as 1U and reduces the cost of ultra-performance AI data storage by up to 70%. Over the last few years PEAK:AIO has established itself as a leading storage provider within the AI market, with a large user base ranging from startups to world leading projects including the high-profile UK NHS AI deployment trials.
PEAK:AIO, founded in 2019, is an established vendor within the EMEA AI market. Its new software range, deployed on server technology, is made in collaboration with and support from vendors such as DellEMC, Supermicro, Gigabyte, and will be distributed by PNY Technologies.
CIO INFLUENCE News: Brivo Partners With Sanico USA to Bring Improved Security to Commercial and Industrial Facilities Nationwide
With the performance of the latest generation of CPUs, PCIe5 and NVIDIA’s networking technology, PEAK:AIO has achieved what traditionally would be considered HPC level performance, within a single cost-effective node, enabling resellers to provide blisteringly fast AI solutions at a realistic price point.
Legacy storage solutions have evolved for traditional IT applications, rich in enterprise features unnecessary for AI. Parallel filesystems are designed for HPC environments. Both options are complex and expensive, designed for much larger solutions and impact the budget for GPUs, draining resources away from innovation. PEAK:AIO dramatically reduces the cost, provides faster results for a more effective overall solution, and allows the user to spend funds where they best serve the project, on GPU servers and not storage.
PEAK:AIO Founder and CEO Mark Klarzynski, spearheaded the evolution of software defined storage (SDS), pioneering the development of a leading iSCSI, Fibre Channel, and InfiniBand SDS framework (SCST) still used and licensed today by leading storage vendors. Pushing the then boundaries of performance, Mark went on to develop the initial ultra-performance all flash storage arrays in partnership with vendors such as Fusion-io.
CIO INFLUENCE News: MemComputing Selected to Join 2023 NSIN Propel Hawaii Accelerator
“We started the development of PEAK:AIO when it was clear that AI simply did not fit the business model of the existing storage vendors. The user had to compromise between overly expensive flash storage, or a slower solution,” said Mark Klarzynski, Founder and CEO of PEAK:AIO. “AI is less about ‘big data’ and more about ‘good data’. Why would we expect storage designed for the world of big data, HPC or the enterprise to work perfectly for the new world of AI? It needed new thinking, not just new solutions, but a completely new type of solution. To any vendor that was listening, the message was loud and clear from the users and resellers. We listened and PEAK:AIO developed exactly what was asked for and was an instant success.”
As PEAK:AIO scaled through EMEA, a more system integrator friendly solution was developed enabling the reseller to provide a complete solution using their skills and experience with their chosen hardware.
The latest version of PEAK:AIO, publicly available Q2, provides HPC level performance with the simplicity of a traditional NAS at a cost never seen before. PEAK:AIO revolutionizes the way the AI industry works with storage by enabling larger and faster GPU solutions to be priced for a much wider market.
PEAK:AIO solutions start at under $8K and scale from 50TB to 700TB of NVMe flash per single node and is currently available to resellers for beta testing.
CIO INFLUENCE News: Arcserve Global Research Reveals Cloud Investments Grow While Data Protection Lags Behind
[To share your insights with us, please write to sghosh@martechseries.com]
