A peer-to-peer IO buffering service based on RAM-grid

Authors:
Yiming Zhang;Rui Chu;Dongsheng Li
Affiliations:
National Laboratory for Parallel and Distributed Processing, National University of Defense Technology, Changsha 410073, China.;National Laboratory for Parallel and Distributed Processing, National University of Defense Technology, Changsha 410073, China.;National Laboratory for Parallel and Distributed Processing, National University of Defense Technology, Changsha 410073, China
Venue:
International Journal of Autonomous and Adaptive Communications Systems
Year:
2009

Citing 17
Cited 0

Implementing global memory management in a workstation cluster

SOSP '95 Proceedings of the fifteenth ACM symposium on Operating systems principles
Managing server load in global memory systems

SIGMETRICS '97 Proceedings of the 1997 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
LIRS: an efficient low inter-reference recency set replacement policy to improve buffer cache performance

SIGMETRICS '02 Proceedings of the 2002 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
WSCLOCK—a simple and effective algorithm for virtual memory management

SOSP '81 Proceedings of the eighth ACM symposium on Operating systems principles
BRITE: An Approach to Universal Topology Generation

MASCOTS '01 Proceedings of the Ninth International Symposium in Modeling, Analysis and Simulation of Computer and Telecommunication Systems
The Utility of Exploiting Idle Memory for Data-Intensive Computations

The Utility of Exploiting Idle Memory for Data-Intensive Computations
A scalable distributed information management system

Proceedings of the 2004 conference on Applications, technologies, architectures, and protocols for computer communications
Parallel Network RAM: Effectively Utilizing Global Cluster Memory for Large Data-Intensive Parallel Programs

ICPP '04 Proceedings of the 2004 International Conference on Parallel Processing
Latency lags bandwith

Communications of the ACM - Voting systems
Delay-Bounded Range Queries in DHT-based Peer-to-Peer Systems

ICDCS '06 Proceedings of the 26th IEEE International Conference on Distributed Computing Systems
A low-overhead high-performance unified buffer management scheme that exploits sequential and looping references

OSDI'00 Proceedings of the 4th conference on Symposium on Operating System Design & Implementation - Volume 4
Life or death at block-level

OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
Program-counter-based pattern classification in buffer caching

OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
Implementation of a reliable remote memory pager

ATEC '96 Proceedings of the 1996 annual conference on USENIX Annual Technical Conference
An implementation study of a detection-based adaptive block replacement scheme

ATEC '99 Proceedings of the annual conference on USENIX Annual Technical Conference
A push-based prefetching for cooperative caching RAM Grid

ICPADS '07 Proceedings of the 13th International Conference on Parallel and Distributed Systems - Volume 01
A distributed paging RAM grid system for wide-area memory sharing

IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The performance of IO-intensive applications is determined by the hit ratio of local disk cache and the IO latency of missed disk accesses. To improve the IO performance, in this paper we propose PIB, a peer-to-peer IO buffering service based on RAM-grid (a new grid system aiming at memory resource sharing in WAN). The PIB service acts as a two-level disk cache, which buffers obsolete blocks in idle nodes on the internet for IO-intensive applications. It reduces the IO latency of missed disk accesses by means of the speed advantage of network over disks, and improves the hit ratio of local cache based on accurate identification of IO patterns. The effectiveness of our proposals is demonstrated through the trace driven simulation studies.