// project
PATh
CHTC and OSG partnership expanding access to distributed high-throughput computing and the OSDF.
PATh (Partnership to Advance Throughput Computing) connects the Center for High Throughput Computing and the Open Science Grid to broaden access to distributed high-throughput computing for research communities.
My contribution is on the data side: operating the Open Science Data Federation (OSDF) caches and origins that PATh workloads read from, and the monitoring that shows whether that data movement is healthy.
The recurring problem is not peak throughput but predictability — a cache that quietly stops serving, or a dataset that is slow from one site and fast from another, costs researchers more time than raw bandwidth ever wins back.
Technologies
- HTCondor — HTCondor high-throughput computing scheduler
- OSDF — Open Science Data Federation
- XRootD — XRootD distributed data access framework
- CVMFS — CernVM File System
- SciTokens — SciTokens capability-based authorization
- Python