Case Study: MIT achieves record-scale computing and faster research with Google for Education Google Cloud Platform

MIT professor pushes computing limits with the largest cluster ever built in the public cloud

Massachusetts Institute of Technology (MIT) used Google for Education’s Google Cloud Platform to support the L-Functions and Modular Forms Database (LMFDB), a massive research project that stores and serves highly complex mathematical objects. The team needed a cloud environment that could handle rapidly growing storage needs, scale to support constant public searches, and let collaborators across countries administer the system easily.

Google for Education provided Google Compute Engine, Persistent Disk, Cloud Load Balancing, Stackdriver, and Cloud Console to power the LMFDB infrastructure. With Google Compute Engine, MIT researcher Andrew V. Sutherland was able to run 580,000 cores with preemptible VMs—the largest known high-performance cluster ever run in the public cloud—producing 70,000 new curves while keeping costs low, with preemptible instances reducing compute costs by up to 80 percent.

Massachusetts Institute of Technology

Andrew V. Sutherland

Computational Number Theorist and Principal Research Scientist