Today NGD Systems was given SBIR Phase IIB award from the National Science Foundation (NSF) to continue its work on in-situ processing. The award is a continuation of the SBIR Phase I award that we received in 2016. For a venture capital funded company, the value of this award is the vote of confidence that it represents in computational storage and in the ability of NGD Systems to commercialize this technology (though the money is also nice !).
The greatest problem with petabyte-scale problems is the huge amount of data that needs to be moved from storage to RAM so that the CPU can process it. This can take tens of minutes even over a 32-lane PCIe bus, making “real-time” analytics difficult. Tim Stammers of 451 Research said it very succinctly: “Because even small amounts of data have gravity, conventional architectures need time to move data from storage devices into the processors that need to access it.” Our architecture solves this problem, not by making data “weightless”, but by moving computing to the data. By doing so, we get rid of the largest source of latency in petabyte-scale problem – moving the data. The NSF SBIR Phase IIB program will help NGD Systems apply computational storage to more commercial applications. You can read more about it in our in-situ processing white paper.