A Case for Optimistic Coordination in HPC Storage Systems

Conference · Sun Jan 01 00:00:00 EST 2012

OSTI ID:1055192

Carns, Philip ^[1]; Harms, Kevin ^[1]; Kimpe, Dries ^[1]; Wozniak, Justin ^[1]; Ross, Robert ^[1]; Ward, Lee ^[2]; Curry, Matthew ^[2]; Klundt, Ruth ^[2]; Danielson, Geoff ^[2]; Karakoyunlu, Cengiz ^[3]; Chandy, John ^[3]; Gropp, William D ^[4]; Settlemyer, Bradley W ^[5]

---

1. Argonne National Laboratory (ANL)
2. Sandia National Laboratory (SNL)
3. University of Connecticut
4. University of Illinois, Urbana-Champaign
5. ORNL

High-performance computing (HPC) storage systems rely on access coordination to ensure that concurrent updates do not produce incoherent results. HPC storage systems typically employ pessimistic distributed locking to provide this functionality in cases where applications cannot perform their own coordination. This approach, however, introduces significant performance overhead and complicates fault handling. In this work we evaluate the viability of optimistic conditional storage operations as an alternative to distributed locking in HPC storage systems. We investigate design strategies and compare the two approaches in a prototype object storage system using a parallel read/modify/write benchmark. Our prototype illustrates that conditional operations can be easily integrated into distributed object storage systems and can outperform standard coordination primitives for simple update workloads. Our experiments show that conditional updates can achieve over two orders of magnitude higher performance than pessimistic locking for some parallel read/modify/write workloads.

• Research Organization: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: Work for Others (WFO)
• DOE Contract Number: DE-AC05-00OR22725
• OSTI ID: 1055192
• Resource Relation: Conference: 7th Parallel Data Storage Workshop, Salt Lake City, UT, USA, 20121112, 20121112
• Country of Publication: United States
• Language: English