Los Alamos National Laboratory and SK hynix have developed the first ordered Key Value Store Computational Storage Device (KV-CSD), and plan to demonstrate it at the Flash Memory Summit this week.
The computational storage device will help accelerate indexing capabilities, speeding up the analysis of large volumes of data.
Computational storage processes the data as close as possible to where it is stored, instead of wasting time and energy shuttling it back and forth.
In the KV-CSD prototype, taking over the indexing capability speeded up some LANL’s applications for analyzing simulation data by a thousand times, the government research body said.
“Moving our large-scale physics simulations from file-based I/O to record- and columnar-indexed I/O has shown incredible speedups for analysis of simulation output,” said Gary Grider, High Performance Computing division leader at Los Alamos. “Demonstrations like this show it is possible to build an ordered KV-CSD that moves the ordering and indexing of data as close to the storage device as possible, maximizing the wins on retrieval from on-the-fly indexing as data is written to the storage. The ordering capability enables range queries that are particularly useful in computational science applications as well as point queries that key value storage is known for.”
Los Alamos National Laboratory and SK hynix have a memorandum of understanding around the design, implementation, and evaluation of the KV-CSD.
Charles Ahn, head of Solution Development at SK hynix. “We are very excited about continuing our research partnership with Los Alamos on this high-performance innovation, we consider the partnership with Los Alamos an important steppingstone toward our commitment to invest and find innovative technologies that tackle memory and storage bottlenecks in traditional HPC infrastructure.”
The announcement comes a week after Intel admitted it was crashing out of the Optane Persistent Memory business, costing it hundreds of millions of dollars.