An efficient buffer management scheme for multimedia file system

Authors
Citation
J. Nang et S. Heo, An efficient buffer management scheme for multimedia file system, IEICE T INF, E83D(6), 2000, pp. 1225-1236
Citations number
16
Categorie Soggetti
Information Tecnology & Communication Systems
Journal title
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS
ISSN journal
09168532 → ACNP
Volume
E83D
Issue
6
Year of publication
2000
Pages
1225 - 1236
Database
ISI
SICI code
0916-8532(200006)E83D:6<1225:AEBMSF>2.0.ZU;2-I
Abstract
File system buffers provide memory space for data being transferred to and from disk and act as caches for the recently used blocks, and the buffer ma nager usually reads ahead data blocks tu minimize the number of disk access es. However, if several multimedia files with different consumption rates a re accessed simultaneously from the file system in which LRU buffer replace ment strategy is used, the read-ahead blocks of the low rate file are unloa ded from memory to be used for loading a data block of a high data rate fil e, therefore they should be reloaded again into memory from disk when these blocks are actually referenced. This paper proposes and implements a new b uffer cache management scheme for a multimedia file system and analyzes the performance of the proposal scheme by modifying the file system kernel of FreeBSD. In this proposed scheme, initially, some buffers are allocated to each opened multimedia file, privately, then these buffers are reused for o ther data blocks of that file when they are loaded from the disk. Moreover, the number of private buffers allocated for the file is dynamically adjust ed according to its data rate. An admission control scheme is also proposed to prevent opening of a new file which may cause overloads in the file sys tem. Experimental results comparing proposed scheme with the original FreeB SD and a simple CTL-based model show that the proposed buffer management sc heme could support the realtime play back of several multimedia files with various data rates concurrently without helps of a realtime CPU and disk sc heduling.