In-network block repairing for erasure coding storage systems

摘要

In the erasure coding storage system, it is necessary to extract multiple data blocks from otherremaining storage nodes to a new node when a storage node fails, which repairs the failed datablock satisfactorily. However, this would incur the incast problem at this new node. The existingsolutions for the repair process in the incast problem mainly rely on path planning and resourceallocation. Although these solutions improve the performance of repairing the failed data blocks,they still waste a large amount of storage and bandwidth resources unavoidably. In this paper,we propose the incast problem to be resolved economically via the in-network aggregation.Specifically, we assume that the switches in data centers have certain data processing capabilitiesand can aggregate data flows efficiently. Thereafter, we propose a set of in-network methods torepair a failed data block in the erasure coding storage systems, taking the fat-tree data centeras an example. Thus, the incast problem can be solved effectively during the data transmissionprocess. Compared with the prior methods, our approach effectively avoids the overhead ofextra path computing, as well as significantly reduces the link cost of repairing data blocks, whilepromising similar or faster repair speed.