Lachesis: Robust Database Storage Management Based on Device-specific Performance Characteristics Jiri Schindler, Anastassia Ailamaki, and Gregory R. Ganger   View Paper (PDF)   Return to Storage Management (Session B9) Abstract Database systems work hard to tune I/O performance, but do not always achieve the full performance potential of mod- ern disk systems. Their abstracted view of storage compo- nents hides useful device-specific characteristics, such as disk track boundaries and advanced built-in firmware algorithms. This paper presents a new storage manager architecture, called Lachesis, that exploits and adapts to observable device-specific characteristics in order to achieve and sustain high perfor- mance. For DSS queries, Lachesis achieves I/O efficiency nearly equivalent to sequential streaming even in the presence of competing random I/O traffic. In addition, Lachesis sim- plifies manual configuration and restores the optimizer's as- sumptions about the relative costs of different access patterns expressed in query plans. Experiments using IBM DB2 I/O traces as well as a prototype implementation show that Lach- esis improves standalone DSS performance by 10% on average. More importantly, when running concurrently with an on-line transaction processing (OLTP) workload, Lachesis improves DSS performance by up to 3x, while OLTP also exhibits a 7% speedup. ©2004 Association for Computing Machinery