Authors: Oguz Selvitopi (Lawrence Berkeley National Laboratory (LBNL)); Saliya Ekanayake (Microsoft Corporation); Giulia Guidi (University of California, Berkeley); Muaaz Awan (National Energy Research Scientific Computing Center (NERSC)); Georgios Pavlopoulos (Biomedical Sciences Research Center (BSRC), Greece); Ariful Azad (Indiana University); Nikos Kyrpides (US Department of Energy Joint Genome Institute); Leonid Oliker (Lawrence Berkeley National Laboratory (LBNL)); Katherine Yelick (University of California, Berkeley; Lawrence Berkeley National Laboratory (LBNL)); and Aydin Buluç (Lawrence Berkeley National Laboratory (LBNL); University of California, Berkeley)
Abstract: Similarity search is one of the most fundamental computations that are regularly performed on ever-increasing protein datasets. Scalability is of paramount importance for uncovering novel phenomena that occur at very large scales. We unleash the power of over 20,000 GPUs on the Summit system to perform all-vs-all protein similarity search on one of the largest publicly available datasets with 405 million proteins, in less than 3.5 hours, cutting the time-to-solution for many use cases from weeks. The variability of protein sequence lengths, as well as the sparsity of the space of pairwise comparisons, make this a challenging problem in distributed memory. Due to the need to construct and maintain a data structure holding indices to all other sequences, this application has a huge memory footprint that makes it hard to scale the problem sizes. We overcome this memory limitation by innovative matrix-based blocking techniques, without introducing additional load imbalance.
Back to Technical Papers Archive Listing