All the resources you need in one place: compute nodes, GPUs, large memory nodes, blazing fast interconnect, parallel filesystems, and more!
Sherlock provides all the software tools and storage resources you'll need to explore and analyze your research data.
With a whole range of computational tools at your fingertips, scientific breakthroughs will just be a batch job away.
Sherlock is a shared computing cluster available for use by all Stanford Faculty members and their research teams, for sponsored or departmental faculty research. All research teams on Sherlock have access to a base set of managed computing resources, GPU-based servers, and a multi-petabyte, high-performance parallel file system for short-term storage.
Faculty can supplement these shared nodes by purchasing additional servers, and become Sherlock owners. By investing in the cluster, PI groups not only receive exclusive access to the nodes they purchase, but also get access to all of the other owner compute nodes when they're not in use, thus giving them access to the whole breadth of Sherlock resources.
Using Sherlock for your work provides many advantages over individual solutions: hosted in an on-premises, state-of-the-art datacenter dedicated to research computing systems, the Sherlock cluster is powered and cooled by installations that are optimized for scientific computing.
On Sherlock, simulations and workloads benefit from performance levels that only large scale HPC systems can offer: high-performance I/O infrastructure, petabytes of storage, large variety of hardware configurations, GPU accelerators, centralized system administration and management provided by Stanford Research Computing.
Such features are not easily accessible at the departmental level, and often require both significant initial investments and recurring costs. Joining Sherlock allows researchers and Faculty members to avoid those costs and benefit from economies of scale, as well as to access larger, professionally managed computing resources that what would not be available on an individual or even departmental basis.
Sherlock is free to use for anyone doing departmental or sponsored research at Stanford.
Any Faculty member can request access for research purposes, and get an account with a base storage allocation and unlimited compute time on the global, shared pool of resources.
Stanford Research Computing provides faculty with the opportunity to purchase from a catalog a recommended compute node configurations, for the use of their research teams. Using a traditional compute cluster condominium model, participating faculty and their teams get priority access to the resources they purchase. When those resources are idle, other "owners" can use them, until the purchasing owner wants to use them. When this happens, those other owners jobs are re-queued to free up resources. Participating owner PIs also have shared access to the original base Sherlock nodes, along with everyone else.
Quite big! It's actually difficult to give a definitive answer, as Sherlock is constantly evolving and expanding with new hardware additions.
As of October 2024, Sherlock features over 7,600 CPU cores available to all researchers, and more than 48,800 additional CPU cores available to Sherlock owners, faculty who have augmented the cluster with their own purchases. With a computing power over 6.1 Petaflops, Sherlock would have its place in the Top500 list of the 500 most powerful computer systems in the world.
For more details about Sherlock size and technical specifications, please refer to the tech specs section of the documentation. And for even more numbers and figures, see the Sherlock facts page.
You can request an account right now, take a look at the documentation, and drop us an email if you have any questions.
If you're interested in becoming an owner on Sherlock, and benefit from all the advantages associated, please take a look at the catalog of configurations, feel free to use the ordering form to submit your request, and we'll get back to you.