|
4th USENIX Conference on File and Storage Technologies—Abstract
Pp. 253–266 of the Proceedings
Managing Prefetch Memory for Data-Intensive Online Servers
Chuanpeng Li and Kai Shen, University of Rochester
Abstract
Data-intensive online servers may contain a significant amount of prefetched data in memory due to largegranularity
I/O prefetching and high execution concurrency.
Using a traditional access recency or frequencybased
page reclamation policy, memory contention can
cause a substantial number of prefetched pages to be
prematurely evicted before being accessed. This paper
presents a new memory management framework that
handles prefetched (but not-yet-accessed) pages separately
from the rest of the memory buffer cache. We examine
three new heuristic policies when a victim page
(among the prefetched pages) needs to be identified for
eviction: 1) evict the last page of the longest prefetch
stream; 2) evict the last page of the least recently accessed
prefetch stream; and 3) evict the last page of the
prefetch stream whose owner process has consumed the
most amount of CPU since it last accessed the prefetch
stream. These policies require no application changes or
hints on their data access patterns.
We have implemented the proposed techniques in the
Linux 2.6.10 kernel and conducted experiments based
on microbenchmarks and two real application workloads
(a trace-driven index searching server and the Apache
Web server hosting media clips). Compared with access
history-based policies, our memory management
scheme can improve the server throughput of real workloads
by 11–64% at high concurrency levels. Further, the
proposed approach is 10–32% below an approximated
optimal page reclamation policy that uses applicationprovided
I/O access hints. The space overhead of our
implementation is about 0.4% of the physical memory
size.
- View the full text of this paper in HTML and PDF.
Until December 2006, you will need your USENIX membership identification in order to access the full papers. The Proceedings are published as a collective work, © 2005 by the USENIX Association. All Rights Reserved. Rights to individual papers remain with the author or the author's employer. Permission is granted for the noncommercial reproduction of the complete work for educational or research purposes. USENIX acknowledges all trademarks within this paper.
- If you need the latest Adobe Acrobat Reader, you can download it from Adobe's site.
|
Need help? Use our Contacts page.
