SK hynix partnered with the US Los Alamos National Laboratory (LANL) to develop a Key Value Store Computational Storage Device (KV-CSD). The device is capable of quickly analyzing large amounts of data.
KV-CSD uses computational storage, a technology that processes large amounts of data close to the source. There’s little to no time lost when moving data from a storage to a computational device. Samsung, for instance, applies the technology to SSDs. Processing power is built directly into the drive controller, allowing direct access to the data stored.
Sk hynix recently built a Key Value Store Computational Storage Device (KV-CSD). The device speeds up the analysis of large datasets. The technology was developed in cooperation with the Los Alamos National Laboratory (LANL), a US physics research institute.
Research data analysis
According to researchers, the computational storage device speeds up some of LANL’s analytics applications by a factor of a thousand. The device features indexing capabilities to allow indexing while data is written. This saves an enormous amount of time whenretrieving data for analysis.
The researchers further note that the technology assists in the transition from file-based I/O to record-column-indexed I/O, which LANL recently initiated. The transition should provide more processing speed for simulation data analysis.
Both parties signed a Memorandum of Understanding (MoU) on the design, implementation and evaluation of the KV-CSD. This document underlines that the technology is a joint effort. Both parties have a say in future applications.