发明名称 |
Allocation of replica-sets in a storage cluster |
摘要 |
A data storage system may be configured to allocate replica-sets in a balanced manner and mark some of these balanced replica-sets as being spares. As one or more drives or machines fail, the data storage system may move all copies of an affected replica-set to a marked spare replica-set and mark the affected replica-set as being inactive or invalid. As the failed drives are replaced, the data storage system may reconfigure those inactive replica-sets and use them as new spares. The data storage system may implement a coordinator module that handles the balancing and allocation of spares within a sub-cluster. The coordinator may also reallocate entire replica-sets across sub-clusters to maintain balance at the cluster level. |
申请公布号 |
US9262278(B2) |
申请公布日期 |
2016.02.16 |
申请号 |
US201313946551 |
申请日期 |
2013.07.19 |
申请人 |
eBay Inc. |
发明人 |
Ben Romdhane Sami;El-Charif Rami;Avalani Bhaven |
分类号 |
G06F11/00;G06F11/14;G06F11/20 |
主分类号 |
G06F11/00 |
代理机构 |
Schwegman Lundberg & Woessner, P.A. |
代理人 |
Schwegman Lundberg & Woessner, P.A. |
主权项 |
1. A method comprising:
detecting a failed data volume in a sub-cluster among multiple sub-clusters within a cluster of data storage machines,
the sub-cluster including an affected replica-set with e failed data volume and including a spare replica-set,the affected replica-set being a first group of mirrored data volumes,the spare replica-set being a second group of mirrored data volumes; marking non-failed data volumes of the affected replica-set as inactive in response to the detecting of the failed data volume in the affected replica-set, the marking of the non-failed data volumes being performed by a processor of a machine; determining a spare percentage for the sub-cluster, the spare percentage indicating a proportion of spare replica-sets among multiple replica-sets in the sub-cluster of data storage machines; copying data from the affected replica-set to the proportion of spare replica-set in response to the detecting of the failed data volume; detecting a replacement of the failed data volume with a substitute data volume in the affected replica-set; and marking the non-failed data volumes and the substitute volume as a further spare replica-set in response to the detecting of the replacement of the failed data volume. |
地址 |
San Jose CA US |