Avichai Giat
ABSTRACT
A typical disaster recovery system will have mirrored storage at a site that is geographically separate from the main operational site. In many cases, communication between the local site and the backup repository site is performed over a network which is inherently slow, such as a WAN, or is highly strained, for example due to a whole-site disaster recovery operation.
The goal of this work is to alleviate the performance impact of the network in such a scenario, and to do so using machine learning techniques. We focus on two main areas, prefetching and read-ahead size determination. In both cases we significantly improve the performance of the system.
Our main contributions are as follows: We introduce a theoretical model of the system and the problem we are trying to solve and bound the gain from prefetching techniques. We construct two frequent pattern mining algorithms and use them for prefetching. A framework for controlling and combining multiple prefetch algorithms is presented as well. These algorithms, as well as various simple prefetch algorithms, are compared on a simulation environment. We introduce a novel algorithm for determining the amount of read ahead on such a system that is based on intuition from online competitive analysis and on regression techniques. The significant positive impact of this algorithm is demonstrated on IBM's FastBack system.
Much of our improvements have been applied with little or no modification of the current implementation's internals. We therefore feel confident in stating that the techniques are general and are likely to have applications elsewhere.
- WU Fengguang, XI Hongsheng, and XU Chenfeng. On the design of a new linux readahead framework. SIGOPS Oper. Syst. Rev., 42(5):75--84, 2008. Google Scholar
Digital Library
- Carsten Gerlhof, Carsten A. Gerlhof, and Alfons Kemper. A multi-threaded architecture for prefetching in object bases. In In Proc. of the Int. Conf. on Extending Database Technology, pages 351--364. Springer-Verlag, 1994. Google ScholarDigital Library
- Carsten A. Gerlhof and Alfons Kemper. Prefetch support relations in object bases. In In Proc. of the Sixth Int. Workshop on Persistent Object Systems, pages 115--126. Springer and British Computer Society, 1994. Google ScholarDigital Library
- Binny S. Gill, Luis Angel, and D. Bathen. Amp: Adaptive multi-stream prefetching in a shared cache. In In Proceedings of the Fifth USENIX Symposium on File and Storage Technologies (FAST 07, pages 185--198, 2007. Google ScholarDigital Library
- Binny S. Gill and Dharmendra S. Modha. Sarc: Sequential prefetching in adaptive replacement cache. In In Proceedings of USENIX 2005 Annual Technical Conference, page 293308, 2005. Google ScholarDigital Library
- D. Xin J. Han, H. Cheng and X. Yan. Frequent pattern mining: Current status and future directions. In Data Mining and Knowledge Discovery, 10th Anniversary Issue, pages 55--86, 2007. Google ScholarDigital Library
- Hui Lei and Dan Duchamp. An analytical approach to file prefetching. In In Proceedings of the USENIX 1997 Annual Technical Conference, pages 275--288, 1997. Google ScholarDigital Library
- Zhenmin Li, Zhifeng Chen, Sudarshan M. Srinivasan, and Yuanyuan Zhou. C-miner: Mining block correlations in storage systems. In In Proceedings of the 3rd USENIX Symposium on File and Storage Technologies (FAST 04, pages 173--186, 2004. Google ScholarDigital Library
- Shuang Liang, Song Jiang, and Xiaodong Zhang. Step: Sequentiality and thrashing detection based prefetching to improve performance of networked storage servers. In ICDCS '07: Proceedings of the 27th International Conference on Distributed Computing Systems, page 64, Washington, DC, USA, 2007. IEEE Computer Society. Google ScholarDigital Library
- OLTP traces. Available via http://traces.cs.umass.edu/index.php/Storage/Storage.Google Scholar
- Mark Palmer. Fido: A cache that learns to fetch. In In Proceedings of the 17th International Conference on Very Large Data Bases, pages 255--264, 1991. Google ScholarDigital Library
- R. Hugo Patterson, Garth A. Gibson, Eka Ginting, Daniel Stodolsky, and Jim Zelenka. Informed prefetching and caching. In In Proceedings of the Fifteenth ACM Symposium on Operating Systems Principles, pages 79--95. ACM Press, 1995. Google ScholarDigital Library
- Carl Tait, Hui Lei, and Swamp Acharya. Intelligent file hoarding for mobile computers, 1995.Google Scholar
- A. Inkeri Verkamo. Empirical results on locality in database referencing. SIGMETRICS Perform. Eval. Rev., 13(2):49--58, 1985. Google ScholarDigital Library
- H. Wedekind and George Zoerntlein. Prefetching in realtime database applications. SIGMOD Rec., 15(2):215--226, 1986. Google ScholarDigital Library
Index Terms
- Using machine learning techniques to enhance the performance of an automatic backup and recovery system
Recommendations
Increasing hardware data prefetching performance using the second-level cache
Techniques to reduce or tolerate large memory latencies are critical for achieving high processor performance. Hardware data prefetching is one of the most heavily studied solutions, but it is essentially applied to first-level caches where it can ...
Using Libception to Understand and Improve HTTP Streaming Video Server Throughput
ICPE '17: Proceedings of the 8th ACM/SPEC on International Conference on Performance EngineeringVideo streaming applications generate a large fraction of Internet traffic. Much of this content is delivered over HTTP using standard web servers. Unlike other types of web workloads, HTTP video streaming workloads are typically disk bound, and ...
Improving Performance in Sub-Block Caches with Optimized Replacement Policies
Special Issues on Neuromorphic Computing and Emerging Many-Core Systems for Exascale ComputingRecent advances in computer processor design have led to the introduction of sub-blocking to cache architectures. Sub-block caches reduce the tag area and power overhead in caches without reducing the effective cache size by using fewer tags to index ...
Comments