Title :
SNAP: efficient snapshots for back-in-time execution
Author :
Shrira, Liuba ; Xu, Hao
Author_Institution :
Microsoft Res., USA
Abstract :
SNAP is a novel high-performance snapshot system for object storage systems. The goal is to provide a snapshot service that is efficient enough to permit "back-in-time" read-only activities to run against application-specified snapshots. Such activities are often impossible to run against rapidly evolving current state because of interference or because the required activity is determined in retrospect. A key innovation in SNAP is that it provides snapshots that are transactionally consistent, yet non-disruptive. Unlike earlier systems, we use novel in-memory data structures to ensure that frequent snapshots do not block applications from accessing the storage system, and do not cause unnecessary disk operations. SNAP takes a novel approach to dealing with snapshot meta-data using a new technique that supports both incremental meta-data creation and efficient meta-data reconstruction. We have implemented a SNAP prototype and analyzed its performance. Preliminary results show that providing snapshots for back-in-time activities has low impact on system performance even when snapshots are frequent.
Keywords :
data structures; meta data; object-oriented databases; transaction processing; SNAP high-performance snapshot system; application-specified snapshots; back-in-time execution; in-memory data structures; incremental meta-data creation; object storage systems; snapshot meta-data reconstruction; Analytical models; Computer science; Costs; Data structures; Interference; Performance analysis; Prototypes; Space technology; System performance; Technological innovation;
Conference_Titel :
Data Engineering, 2005. ICDE 2005. Proceedings. 21st International Conference on
Print_ISBN :
0-7695-2285-8
DOI :
10.1109/ICDE.2005.133