发明名称 Scalable concurrent execution of distributed workflows sharing common operations
摘要 Examples perform concurrent execution of distributed workflows sharing common operations by a plurality of nodes, such as execution of recovery plans for disaster recovery of virtual machines operating on and off premises. Concurrent execution of identical operations that were part of a previously executed workflow are prevented, by evaluating the source of the workflow and whether the workflow has previously been initiated by that source. The disclosure is scalable to allow for new nodes to be included.
申请公布号 US9563478(B1) 申请公布日期 2017.02.07
申请号 US201514839887 申请日期 2015.08.28
申请人 VMware, Inc. 发明人 Miller Swaha;Pershin Aleksey;Kolechkin Michael;Golosovker Victor
分类号 G06F9/46;G06F9/50 主分类号 G06F9/46
代理机构 代理人
主权项 1. A computing system comprising: a memory area storing a plurality of recovery plans each including a plurality of workflows, each workflow in the plurality of workflows including a plurality of workflow templates, each workflow template in the plurality of workflow templates being associated with a sequence of operations; a cluster of management nodes that, upon receiving at least one of the plurality of recovery plans and one of the plurality of workflow templates associated therewith, executes at least one of the plurality of workflows included in the at least one of the plurality of recovery plans without duplication of previously performed operations associated with other of the plurality of workflows by: generating an initiator identifier (ID) for each of the plurality of workflows, the initiator ID to be associated with every operation executed as part of respective workflow templates in the plurality of workflow templates;accessing, from the memory area, the at least one of the plurality of workflows and an initiator ID associated with a plurality of operations in the one of the plurality of workflow templates associated with the at least one of the plurality of workflows included in the received at least one of the plurality of recovery plans;performing the plurality of operations in the at least one of the plurality of workflow templates;storing the initiator ID associated with the plurality of operations;if an operation in the plurality of operations fails due to previous execution of the operation: comparing an initiator ID associated with the previous execution of the operation to the stored initiator ID; andbased on determining a match between the initiator ID associated with the previous execution of the operation and the stored initiator ID, performing subsequent operations in the one of the plurality of workflow templates; andbased on determining no match between initiator ID associated with the previous execution of the operation and the stored initiator ID, terminating performance of the one of the plurality of workflow templates.
地址 Palo Alto CA US