Knowledge Base

Article ID: 249 - Last Modified: December 4, 2010

Does Glide scale well on multiple cores/cpus? Are there any particular recommendations for running parallel Glide jobs?

Glide is an 'embarrassingly parallel' distributed program, rather than a 'true parallel' job requiring interprocess communication; that is, the ligands to be docked are split into multiple subsets that are docked in independent subjobs, with the results merged together at the end. Therefore, Glide scales very well to many processors, as long as the number of structures docked in a subjob doesn't get too small. Our hardware and software requirements are listed in Chapter 2 of the Installation Guide; one particular recommendation that is relevant when considering multiple-core machines is that you have at least 1 GB of memory per core. If you'll be docking large ligand libraries, we recommend keeping the subjob size small (say, less than 20K structures each). You can split the job into many more subjobs than processors/licenses you'll be using (these are separate options in the Start dialog when launching the job from Maestro). This keeps the subjob execution time reasonable, which helps with load-balancing and restartability.

Keywords: docking, parallel, distributed, multiple, cores, cpus, cluster, scale

Back to Search Results

Was this information helpful?

What can we do to improve this information?


To ask a question or get help, please submit a support ticket or email us at help@schrodinger.com.
Knowledge Base Search

Type the words or phrases on which you would like to search, or click here to view a list of all
Knowledge Base articles