Intel, along with engineering from the Broad Institute and Google, has contributed to the advancement of biomedical research by presenting a new level of research method through workflow optimization.

The Broad Institute, a joint research institute of MIT and Harvard University, announced on the 13th that it has optimized its workflow for fast and cost-effective Google N1 and N2 instances, building on its ongoing collaboration with Intel and Google Cloud to accelerate genomics research. The research team said data processing costs have been reduced by 85% compared to when they first deployed workloads on Google Cloud.

“The Broad Institute research team knew that the cloud would enable new levels of collaboration across data,” said Geraldine Van der Auwera, director of the Broad Institute’s Data Science Platform, Public Affairs and Communications. “We could build a cloud-based data ecosystem that would allow researchers to combine their workloads with many more datasets than just their own to perform richer and more powerful computational experiments.”

The Broad Institute has migrated its workloads to Google Cloud N2 instances to meet the rapidly growing demand for genomics data generation and computational research. Based on workload requirements, we modularized and optimized cloud instances for pipeline workloads and right-sized them for Intel® Xeon® Scalable processors. Users can run genomic workflows on Google Cloud approximately 25% faster and 34% cheaper by deploying N2 instances powered by Intel Xeon Scalable processors.

Intel has been partnering with Broad Institute since 2017, supporting Broad Institute's pipeline and Genome Analysis Toolkit (GATK) optimizations with Intel libraries, including the Intel® Genomics Kernel Library. Intel and Broad Institute jointly manage the Intel-Broad Genome Data Engineering Center, which enables researchers and software engineers around the world to collaborate and build, optimize, and share new tools and infrastructure to integrate and process genomic data.