Instant recovery in a multi-grained caching framework

A computer-implemented method, comprising:

• receiving indication of failure of a primary cache server at a secondary cache server, the primary and secondary cache servers being configured to manage read requests and write requests for a back-end storage device and a cache storage device;
• setting the secondary cache server to a by-pass mode for read requests directed to portions of the back-end storage device managed by the primary cache server prior to the failure;
• and
• reading an index of cache block descriptors (CBDs) managed by the primary cache server prior to the failure into a memory of the secondary cache server, wherein a complete metadata footprint of the primary cache server is not read into the memory of the secondary cache server,
• wherein CBDs indicated by the index of CBDs managed by the primary cache server prior to the failure provide coarse-grained mapping of data on the back-end storage device, wherein a plurality of fine block descriptors (FBDs) are allocated to each of the CBDs indicated by the index of CBDs managed by the primary cache server prior to the failure, and wherein each FBD of the plurality of FBDs provide fine-grained mapping of data stored to the cache storage device.

A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the embodied program instructions readable/executable by a controller to cause the controller to:

• receive, by the controller, indication of failure of a primary cache server that is configured to manage read requests and write requests for a back-end storage device and a cache storage device;
• set, by the controller, a secondary cache server to a by-pass mode for read requests directed to portions of the back-end storage device managed by the primary cache server prior to the failure, the secondary cache server being configured to manage read requests and write requests for the back-end storage device;
• and
• read, by the controller, an index of cache block descriptors (CBDs) managed by the primary cache server prior to the failure into a memory of the secondary cache server, wherein a complete metadata footprint of the primary cache server is not read into the memory of the secondary cache server,
• wherein CBDs indicated by the index of CBDs managed by the primary cache server prior to the failure provide coarse-grained mapping of data on the back-end storage device,
• wherein a plurality of fine block descriptors (FBDs) are allocated to each of the CBDs indicated by the index of CBDs managed by the primary cache server prior to the failure, and
• wherein each FBD of the plurality of FBDs provide fine-grained mapping of data stored to the cache storage device.