MemVerge™, the pioneers of Big Memory software, today announced that TGen, the Translational Genomics Research Institute, an affiliate of City of Hope, has selected MemVerge Memory Machine Big Memory virtualization software to accelerate time to discovery for Idiopathic Pulmonary Fibrosis (IPF), a disease which affects 100,000 people annually in the U.S. Using MemVerge technology, TGen is able to dramatically speed analytical processing by nearly 36% for single-cell RNA sequencing.
As a nonprofit medical research institute, TGen researchers process single-cell RNA sequences to characterize cell transcriptomic profiles. The process can take up to six and a half hours to analyze a matrix of 30,000 genes by 114,000 cells. With consistently growing datasets, this processing time was preventing a desired time to discovery. The data required for analysis was simply too large to retain in traditional memory, and scaling capacity with dynamic random-access memory (DRAM) was too costly.
TGen has instead deployed memory virtualization technology from MemVerge which virtualizes both DRAM and PMem (persistent memory) memory technologies, to increase the memory pool available for processing without requiring more high-cost DRAM. The solution further speeds TGen’s genomics sequencing analysis with Memory Machine ZeroIO in-memory snapshots which capture multi-terabyte data sets at any point for rapid reloads at each stage of processing. The ZeroIO snapshot service is 1,000 times faster than the fastest storage snapshot to SSD and enables TGen to run processing workflows in parallel. This ensures that in the event of a system crash, in-memory snapshots are available to instantly re-start long running jobs without lengthy reloading.
“By utilizing the snapshotting and cloning capabilities of Memory Machine, we were able to parallelize the processing workflow,” said Glen Otero, Ph.D., Vice President of Scientific Computing at TGen. “As a result, we can now save nearly 36% of computational time while also taking advantage of the big memory nodes. This will save a lot of time in downstream analysis.”
“MemVerge Memory Machine has quickly resulted in research value for TGen,” said Jonathan Jiang, COO of MemVerge. “We have removed performance barriers from their research process so that they are able to perform vital, life-saving, research faster than ever possible. Now TGen is expanding the use of Big Memory technology across other research use cases where results and discoveries can produce findings for a healthier tomorrow.”
MemVerge Memory Machine makes 100% use of available memory capacity while providing new operational capabilities to memory-centric workloads. Memory Machine answers the need for a modern in-memory computing model to support emerging applications that require real-time analytics, true in-memory computing, and fault-tolerant memory persistence to speed massive processing workloads.
Source – PR Newswire