发明名称 Efficiently storing data in a dispersed storage network
摘要 A method includes determining that one or more data blocks of a permanently stored data blocks are to be deleted. In response, the method further includes obtaining a group of partial redundancy data for the permanently stored data blocks. The method further includes identifying a temporarily stored plurality of data blocks for which partial redundancy data does not yet exist. The method further includes creating a new plurality of data blocks from data blocks of the permanently stored plurality of data blocks that are to remain permanently stored and data blocks from the temporarily stored plurality of data blocks that are to be permanently stored. The method further includes permanently storing the new plurality of data blocks. The method further includes generating a new group of partial redundancy data. The method further includes sending the new group of partial redundancy data and the group of partial redundancy data.
申请公布号 US9354972(B2) 申请公布日期 2016.05.31
申请号 US201514953579 申请日期 2015.11.30
申请人 International Business Machines Corporation 发明人 Grube Gary W.;Markison Timothy W.
分类号 G06F12/00;G06F11/10;H04L29/08;G06F3/06 主分类号 G06F12/00
代理机构 Garlick & Markison 代理人 Garlick & Markison ;Markison Timothy W.
主权项 1. A method for execution by a dispersed storage (DS) unit of a dispersed storage network (DSN), the method comprises:determining that one or more data blocks of a permanently stored plurality of data blocks are to be deleted; andin response to the determining that the one or more data blocks are to be deleted: obtaining a group of partial redundancy data for the permanently stored plurality of data blocks; identifying a temporarily stored plurality of data blocks for which partial redundancy data does not yet exist; creating a new plurality of data blocks from data blocks of the permanently stored plurality of data blocks that are to remain permanently stored and data blocks from the temporarily stored plurality of data blocks that are to be permanently stored; permanently storing the new plurality of data blocks; generating a new group of partial redundancy data based on the new plurality of data blocks and in accordance with a dispersed storage error coding function; and sending the new group of partial redundancy data and the group of partial redundancy data such that redundancy data DS units are able to generate redundancy data regarding the new plurality of data blocks.
地址 Armonk NY US