A remote mirroring architecture with adaptively cooperative pipelining

Authors:
Yongzhi Song;Zhenhai Zhao;Bing Liu;Tingting Qin;Gang Wang;Xiaoguang Liu
Affiliations:
Nankai-Baidu Joint Lab, College of Information Technical Science, Nankai University, Tianjin, China;Nankai-Baidu Joint Lab, College of Information Technical Science, Nankai University, Tianjin, China;Nankai-Baidu Joint Lab, College of Information Technical Science, Nankai University, Tianjin, China;Nankai-Baidu Joint Lab, College of Information Technical Science, Nankai University, Tianjin, China;Nankai-Baidu Joint Lab, College of Information Technical Science, Nankai University, Tianjin, China;Nankai-Baidu Joint Lab, College of Information Technical Science, Nankai University, Tianjin, China
Venue:
ICA3PP'10 Proceedings of the 10th international conference on Algorithms and Architectures for Parallel Processing - Volume Part I
Year:
2010

Citing 6
Cited 0

The Network Block Device

Linux Journal
SnapMirror: File-System-Based Asynchronous Mirroring for Disaster Recovery

FAST '02 Proceedings of the Conference on File and Storage Technologies
Designing for Disasters

FAST '04 Proceedings of the 3rd USENIX Conference on File and Storage Technologies
The case for semantic aware remote replication

Proceedings of the second ACM workshop on Storage security and survivability
A Technique for High-Performance Data Compression

Computer
Smoke and mirrors: reflecting files at a geographically remote location without loss of performance

FAST '09 Proccedings of the 7th conference on File and storage technologies

Quantified Score

Hi-index	0.00

Visualization

Abstract

In recent years, the remote mirroring technology has attracted increasing attention In this paper, we present a novel adaptively cooperative pipelining model for remote mirroring systems Unlike the traditional pipelining model, this new model takes the decentralization of processors into account and adopts an adaptive batching strategy to alleviate imbalanced pipeline stages caused by this property To release the heavy load on CPU exerted by compression, encryption, TCP/IP protocol stack and so on, we design fine-grained pipelining, multi-threaded pipelining and hybrid pipelining We implement a remote mirroring prototype based on Linux LVM2 The experimental results show that, the adaptively cooperative pipelining model balances the primary and the backup sites - the two stages of the pipeline effectively, and fine-grained pipelining, multi-threaded pipelining and hybrid pipelining improve the performance remarkably.