A Data Deduplication Framework of Disk Images with Adaptive Block Skipping
Zhou, Jiangtao, Wen Wen
Abstract
Zhou, Jiangtao, Wen Wen
Abstract
我们描述一个有效、容易适用的数据 deduplication 框架,启发式的预言基于适应的块跳过让象磁盘图象那样的真实世界的数据集保存 deduplication 联系了开销并且改进 deduplication 产量,好 deduplication 效率维持了。在框架下面, deduplication 操作为经由启发式的预言作为可能的非副本决定的数据块被跳过,与为在跳过的块以内的复制鉴定的成功和匹配的延期过程一起,磁滞现象机制基于回锅肉丁索引为 re-encountered 更新回锅肉丁索引的过程跳过块。为性能评估,建议框架集成于存在数据域、实现、稀少索引 deduplication 算法。试验性的结果基于 1.0 幅 TB 磁盘图象的真实世界的数据集证明 deduplication 联系了开销,导致30%~是显著地与适应的块跳过减少了在 deduplication 产量的80%改进当 deduplication 元数据在磁盘上被存储为数据时域,并且25%~什么时候的在 deduplication 产量与15%~保存20%改进的40%内存空间一在里面--内存稀少的索引在稀少的索引被使用。在两个盒子中,减少的相应 deduplication 比率低于 5% 。
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
我们描述一个有效、容易适用的数据 deduplication 框架,启发式的预言基于适应的块跳过让象磁盘图象那样的真实世界的数据集保存 deduplication 联系了开销并且改进 deduplication 产量,好 deduplication 效率维持了。在框架下面, deduplication 操作为经由启发式的预言作为可能的非副本决定的数据块被跳过,与为在跳过的块以内的复制鉴定的成功和匹配的延期过程一起,磁滞现象机制基于回锅肉丁索引为 re-encountered 更新回锅肉丁索引的过程跳过块。为性能评估,建议框架集成于存在数据域、实现、稀少索引 deduplication 算法。试验性的结果基于 1.0 幅 TB 磁盘图象的真实世界的数据集证明 deduplication 联系了开销,导致30%~是显著地与适应的块跳过减少了在 deduplication 产量的80%改进当 deduplication 元数据在磁盘上被存储为数据时域,并且25%~什么时候的在 deduplication 产量与15%~保存20%改进的40%内存空间一在里面--内存稀少的索引在稀少的索引被使用。在两个盒子中,减少的相应 deduplication 比率低于 5% 。
Key concepts: Data deduplication, Computer science, Database