FREMONT, Calif., Aug. 8, 2016 /PRNewswire/ -- NVXL comes out of stealth mode to announce a revolutionary storage and compute acceleration platform. "With the massive shift in enterprise data performance driven by deep learning and machine learning, we are responding to and exceeding the need for more storage, acceleration, and efficiency," said company President Jeffery Bunting. "Our high-performance non-volatile storage and compute acceleration modules integrate into a flexible platform, The Plug-in Performance Platform."

The Plug-in Performance Platform is built to plug-in as much non-volatile SSD storage or compute acceleration to the platform as needed, for a lower cost in dollars and power consumption, enabling tremendous flexibility. Running on standard hardware, the platform handles low-level implementation details so developers and data scientists can be more productive.

Two critical technologies that support the NVXL platform were unveiled: FlexAccel and FlexDurance.  FlexAccel leverages FPGA and ASIC acceleration with a unique software layer that allows deep learning, machine learning, and big data developers to use industry-standard APIs, allowing modules to dynamically optimize for peak compute acceleration or faster SSD performance.

FlexDurance extends the write endurance of the 3D NAND flash in the platform, resulting in longer life, greater reliability, increased flexibility, and a lower total cost of ownership.

During the Summit, NVXL will display modules powered by FPGAs which can provide compute acceleration, enterprise-grade SSD storage, or both. The core software and controller technologies of the platform have been under development at NVXL for several years at the company's Silicon Valley headquarters.

"NVXL is building on the advantages of our FPGAs" said Erhaan Shaikh, Vice President and General Manager, Business Units, Intel Programmable Solutions Group. "The programmability and features, such as IEEE 754 hard floating-point multipliers and adders in our FPGAs, enable servers and data centers to meet the complex requirements of search, big data, and deep learning with significant benefits in total cost of ownership."

"The idea of combining storage and compute acceleration in an integrated platform is compelling," said Greg Wong, Founder and Principal Analyst, Forward Insights. "NVXL is clearly an early mover in this space and it will be interesting to see how this technology develops."

NXVL's solutions will be introduced at this year's Flash Memory Summit, which runs August 9th-11th. NVXL will be displaying in booth 801-803 and will be privately meeting with analysts, customers, and press.