Third Party Benchmark Test Shows 54% Improvement in Netlist, Inc. (NLST) HyperCloud Modules
8/3/2012 1:50:38 PM
Yesterday, Netlist announced that its HyperCloud HCDIMM memory modules, recently run through a benchmark test by third party tester Deopli, achieved a 54% improvement in memory bandwidth. The HyperCloud memory is intended for use with electronic design automation tools.
Located in Irvine, CA, Netlist designs and manufactures logic-based memory subsystems for server and storage applications for use in cloud computing. HyperCloud is Netlist’s flagship product. The company also distributes the NuVault and EXPRESSvault product families. Netlist clients include OEMs that design and build tower, rack-mounted, and blade servers, high-performance computing clusters, engineering workstations, and telecommunications equipment. The company has facilities in Suzhou, China, and Silicon Valley.
HyperCloud uses a distributed buffer architecture to reduce latency, while also incorporating rank multiplication and load reduction technologies; rank multiplication allows the for higher DRAM capacity, whereas load reduction shortens loading to the memory interface. Together, these two technologies allow HCDIMM memory to run at faster speeds at maximum capacity. Deopli’s testing showed that the new HyperCloud modules have a HCDIMMs showed a 54 percent memory bandwidth improvement and a 25 percent EDA simulation runtime improvement over LRDIMM (Load Reduced DIMMs) memory.
Chuck Hong, CEO of Netlist, said, “Today’s memory performance in servers is not keeping pace with the evolution of processor technology, creating a high density memory cliff which HyperCloud technology addresses. Deopli’s benchmark results show that memory-hungry EDA tools run faster with HyperCloud. This creates opportunities to either use fewer software licenses or allow more workloads to run at the same time, allowing customers to deliver product to market faster and at lower cost.”
Scott Clark, CEO of Deopli, added, “The specific facets of EDA design flow that would benefit from a larger memory footprint and faster memory access speeds include select parts of block and top-level design, verification and implementation tools which are extremely expensive. By utilizing HyperCloud, our benchmark test results revealed that HCDIMMs can significantly reduce costs and drive productivity in an EDA environment.”
MissionIR is committed to connecting the investment community with companies that have great potential and a strong dedication to building shareholder value. We know our reputation is based on the integrity of our clients and go to great lengths to ensure the companies represented adhere to sound business practices.
Please see disclaimer on the MissionIR website http://www.missionir.com/disclaimer.html