Wird geladen...
From Jobs to Work: Scheduling the Right Bioinformatics Tools
A great deal of effort has been expended toward scheduling computationally intensive jobs on Grids(1,2) and other collections of high performance computing resources. Bioinformatics computer jobs are of particular interest as they are often highly computationally intensive. However, the problem has...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
2002
|
| Schlagworte: | |
| Online Zugang: | https://ncbi.nlm.nih.gov/pmc/articles/PMC2244384/ |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Zusammenfassung: | A great deal of effort has been expended toward scheduling computationally intensive jobs on Grids(1,2) and other collections of high performance computing resources. Bioinformatics computer jobs are of particular interest as they are often highly computationally intensive. However, the problem has not been addressed from the viewpoint of the overall work that should be done. Here, we make a distinction between jobs and work. Jobs are specifically bound computational tasks (e.g., a request to run NCBI's BLAST tool or the GCG FASTA program) versus work requests, which are more general (e.g., a request to compare a set of sequences for similarity). We contend that biology researchers often wish to accomplish work rather than run a particular job. With this idea in mind, it is possible to improve resource usage by mapping work to jobs with the goal of choosing appropriate jobs that can best be scheduled at a given time. |
|---|