In early testing, NVMesh speeds access to beamline microscopy data on GPU servers, bringing greater storage efficiency, agility, and ROI.

Excelero, a disruptor in software-defined block storage for AI/ML/deep learning and GPU computing, said that its partner HPCNow!, a provider of advanced supercomputing and high performance computing (HPC) services for scientific and research organizations, is helping major life science researchers around the world to accelerate research on COVID-19 and other diseases through more efficient data storage with Excelero’s NVMesh® software.

HPCNow! has performed a pilot at the ALBA Synchrotron to demonstrate how existing storage technologies can effectively deploy elastic NVMe based data storage to support the acquisition and processing of the massive volumes of scientific data generated by its high performance beamlines, which use soft and hard X-rays’ intense light source beams to help to characterize materials, their properties and behavior. The pilot report will be published in a market consultation of the CELLS in 2020. Over 2,000 researchers in biosciences, condensed matter (magnetic and electronic properties, nanoscience) and materials science obtain unique information from ALBA’s beamlines each year. With a new initiative to support small to medium enterprises performing COVID-19 research, in addition its many other scientific data processing and analysis workloads, ALBA’s IT Systems department needs an efficient scale-out storage to enable timely analysis and conclusions.

Excelero’s NVMesh™ is software-defined distributed block storage that delivers Elastic NVMe for high performance computing workloads as well as AI/ML/deep learning, data warehouses and containers. Customers benefit from the performance of local flash with the convenience of centralized storage while avoiding proprietary hardware lock-in and reducing the overall storage TCO. Its low-latency, high throughput distributed block storage makes it fast and easy to scale up storage capacity as datasets grow – ensuring HPC applications don’t become bottlenecked, and providing the agility, elasticity, price/performance and ROI that are often hard to achieve at scale.

“Top research institutions need far more throughput than traditional storage systems can provide to power data-intensive microscopy and other research, and we’ve been tremendously impressed with the high IOPS and low latency that Excelero delivers with its NVMesh software,” said David Tur, CEO at HPCNow!, who directed the ALBA proof of concept project. “We are already in the early stages of exploring several large-scale deployments where scale-out storage such as Excelero’s NVMesh can help researchers achieve their goals faster and at lower cost.”

“As a company that sells 100% through partners, Excelero is totally focused on enabling partner success, and we’ve been thrilled at the innovative approaches that HPCNow! is bringing to organizations that need scale-out storage for complex HPC applications,” said Kirill Shoikhet, CTO at Excelero. “We are delighted to help HPCNow! serve its research customers with more efficient, cost-effective storage that helps to accelerate their life-saving work.”

HPCNow! provides its customers with solutions and technologies for dealing with the most complex problems in High Performance Computing (HPC). The company has experts in multiple areas such as cluster design, supercomputer administration, and user support who help organizations with HPC deployments and related IT applications including Big Data, artificial intelligence, and cloud computing. With headquarters in Barcelona and offices in Auckland, New Zealand, HPCNow! helps customers take full advantage of the most advanced technologies in HPC along with providing exceptional customer and user support.

About Excelero Excelero is the market leader in distributed block storage software. Founded in 2014 by a team of storage veterans, the company delivers Elastic NVMe software that enables partners to build end-to-end, high-performance storage solutions for AI training and analytics workloads at any scale. With its partners, Excelero enables customers to massively improve ROI across the entire datacenter, using standard servers, maximizing component utilization (NVMe, GPU), minimizing overhead and reducing software license costs.

Excelero’s NVMesh® is distributed block storage that connects CPUs and GPUs to NVMe flash to create a significant improvement in price/performance, from entry level to any scale. NVMesh was designed as a storage layer that eradicates data bottlenecks so teams can access data at any speed in any location. NVMesh delivers 20x faster data processing for multi-server, multi-GPU compute nodes when working with massive datasets for machine learning, deep learning and complex analytical workloads.