Hardware Overview
LLNL's Jonathan Allen, a bioinformatics scientist, uses the Catalyst supercomputer to develop and refine computational methods for identifying pathogens.
Livermore Computing's (LC's) advanced hardware makes possible the fast and efficient computing that is integral to fulfilling LLNL's missions. Jonathan Allen, a bioinformatics scientist at the Lab, has been working on new methods to rapidly detect and characterize pathogenic organisms such as viruses, bacteria, or fungi in a biological sample. However, traditional technologies and storage limitations made it challenging to rapidly search a database of reference genomes as more organisms were sequenced and more variants in the population of an organism were included—until he started using Catalyst, a Cray supercomputer customized for data-intensive computing tasks such as this. With Catalyst’s unique architecture, Allen and his team can store very large reference databases of genomes in memory and execute expansive analyses with higher resolution than on other systems.
LC helps furnish Livermore researchers and collaborators with an array of well-managed, high-performing, parallel computing systems, backed by massive parallel file systems and storage archives and knowledgeable support staff. Some of these systems are general computational workhorses, while others, like Catalyst, perform certain in-demand tasks such as parallel numerical simulation, visualization, and data analytics particularly well. By working closely with researchers and research groups at the Lab, LC keeps close tabs on research trends and research needs and helps to ensure that scientists and engineers have access to the computing resources they need to complete cutting-edge simulations and calculations in service of the nation. In some instances, LC even works with hardware developers to help shape the design for new computer systems based on the unique and diverse needs of LLNL users.
Major OCF (unclassified) Systems
Vendor
Total Nodes
Total Cores
Total Memory (GB)
Peak TFLOPS (CPUs+GPUs)
Quartz
Penguin
3,018
108,648
344,064
3251.40
Ruby
Supermicro
1,512
84,672
290,304
5959.20
Corona
Penguin
291
14,064
127,488
11335.00
Lassen
IBM
795
34,848
253,440
23047.20
For more a detailed systems summary and information on individual systems, visit our User Portal platforms page.