Los Alamos National Laboratory and NGD Systems are partnering through LANL’s Efficient Mission Centric Computing Consortium (EMC3) to explore scalable computational storage offloads for ultrascale high performance computing (HPC) simulation environments.“Computational storage devices become a key source of acceleration when we are able to directly interpret the data within the storage device,” said Brad Settlemyer, senior scientist in Los Alamos’ HPC Design Group. “With that component in place, near-storage analytics unleashes massive speedups via in-device reduction and fewer roundtrips between the device and host processors.”
Computational offloads using both in-network processing and near-storage compute are becoming an important part of both scale-up and scale-out computing, LANL said in its announcement, with future scaling requirements virtually requiring programmable elements along the data path to achieve performance efficiency goals.
“NGD’s versatile computational storage platform makes it easy to try new concepts for offloading functions to near storage,” said Vladimir Alves, chief technology officer at NGD Systems. “By offering an OS-based storage device, with on-board applications processors in our NVMe SSD ASIC solution, we offer partners like Los Alamos the ability to try many different paths to a more complete solution with a simple and seamless programming and device management model.”
In addition to the effort to explore computational storage offloads, Los Alamos has partnered with NGD Systems to focus on building a curriculum for summer internship programs in using NGD’s Newport Computational Storage Drives to accelerate data analytics.
“Los Alamos is happy to see the evolution of computational offloads towards standards-based computational storage technology, and is hopeful explorations into use cases for this technology will bear fruit for the HPC and at-scale computing environments,” said Gary Grider, HPC division leader at Los Alamos. “We look forward to continued partnership with NGD Systems in efficient computing via the Efficient Mission Centric Computing Consortium.”
This collaborative effort is sponsored by Los Alamos’ EMC3, which focuses on applying efficient computing architectures, system components, and environments to the real-world workload mixes that result from mission-centric work. This focus on efficiency seeks to improve application performance, workflows, and code efforts for better exploitation of current and future systems and workflow software and computer platforms, while maintaining a proper balance of compute, memory size, memory bandwidth and latency, performance, and I/O throughout.
EMC3 has nearly 20 members now, including national and international partners.
Founded in 2013, NGD Systems is a maker of computational storage NVMe SSDs designed for near-time processing at the edge where data is generated. The Newport Platform uses a patented in-situ processing solution built to reduce the bandwidth required to analyze mass data sets.
source: LANL
