Data deduplication has been widely applied in storage systems to improve the efficiency of space utilization. In data deduplication systems, the data restore performance is seriously hindered by read amplification since the accessed data chunks are scattered over many containers. A container consisting of hundreds or thousands data chunks is the data unit to be read from or write to the storage. Several schemes such as forward assembly, container-based caching, and chunk-based caching are used to reduce the number of container-reads during the restore process. However, how to effectively use these schemes to get the best restore performance is still unclear. In this paper, we first study the trade-offs of using these schemes in terms of read amplification and computing time. We then propose a combined data chunk caching and forward assembly scheme called ALACC (Adaptive Look-Ahead Chunk Caching) for improving restore performance which can adapt to different deduplication workloads with a fixed total amount of memory. This is accomplished by extending and shrinking the look-ahead window adaptively to cover an appropriate data recipe range and dynamically deciding the memory to be allocated to forward assembly area and chunk-based caching. Our evaluations show the restore throughput of ALACC is higher than that of the optimum case of various configurations using the fixed amount of memory allocated to forward assembly and to chunk-based caching.
|Original language||English (US)|
|Title of host publication||Proceedings of the 16th USENIX Conference on File and Storage Technologies, FAST 2018|
|Number of pages||15|
|State||Published - 2018|
|Event||16th USENIX Conference on File and Storage Technologies, FAST 2018 - Oakland, United States|
Duration: Feb 12 2018 → Feb 15 2018
|Name||Proceedings of the 16th USENIX Conference on File and Storage Technologies, FAST 2018|
|Conference||16th USENIX Conference on File and Storage Technologies, FAST 2018|
|Period||2/12/18 → 2/15/18|
Bibliographical noteFunding Information:
We thank all the members in CRIS group to provide the useful comments to improve our design. We thank Dongchul Park for assistance with the trace exploring and pre-processing, and Baoquan Zhang for the specific reviewing comments. We would like to thank our shepherd, Philip Shilane, for his useful comments and suggestions. This work is partially supported by the following NSF awards: 1305237, 1421913, 1439622 and 1525617.