Grid storage system and method of operating thereof
The is provided a storage system comprising a plurality of disk units adapted to store data at respective ranges of logical block addresses (LBAs), said addresses constituting an entire address space, and a storage control grid operatively connected to the plurality of disk units and comprising a plurality of data servers. The method of operating the storage system comprises: a) configuring a first data server to have a primary responsibility over a certain range of LBAs, b) configuring a second data server to have a secondary responsibility over said certain range of LBAs and to overtake the primary responsibility over said certain range of LBAs if the first server fails; c) responsive to a request to one or more LBAs within said certain range of LBAs, generating a primary cache object in the first data server and generating a redundancy object in the second data server; d) deleting said redundancy object upon successful permanent storing of data and/or metadata corresponding to said request, said storing provided by the first data server, e) if operation d) fails, using said redundancy object by the second server for providing permanent storing said data and/or metadata when overtaking primary responsibility, wherein said redundancy object is derivation of the primary cache object and need to be reconstructed in order to enable overtaking the responsibility.
1. A storage system comprising:
a) a plurality of disk units adapted to store data at respective ranges of logical block addresses (LBAs), said addresses constituting an entire address space,
b) a storage control grid operatively connected to the plurality of disk units and comprising a plurality of data servers;
wherein each certain LBA is assigned to at least two data servers: a first data server configured to have a primary responsibility for handling requests directed to at least said certain LBA and to generate a primary cache object in response to a request directed to at least saidcertain LBA, and a second data server configured to have a secondary responsibility for handling requests directed to at least said certain LBAand to overtake the primary responsibility for handling requests directed to at least said certain LBA if the primary server fails; and
wherein the storage system is further configured, in response to said request directed to at least said certain LBA, to generate a redundancy object in the second data server, said redundancy object to be deleted upon successful permanent storing by the first data server of data and/or metadata corresponding to said request, and to be used by the second data server for permanent storing said data and/or metadata when overtaking respective responsibility,
wherein said redundancy object is further configured to be handled in a searchable data structure with no need to be incorporated into cache-object control procedures and/or dirty data destaging procedures of the second server before overtaking the primary responsibility, and
wherein said redundancy object is derivation of the primary cache object and need to be reconstructed in order to enable overtaking the responsibility.
2. The system of claim 1 wherein the request directed to one or more LBAs is selected from a group comprising respectively directed I/O requests and configuration-related requests.
3. The system of claim 1 wherein the primary cache object comprises data selected from a group comprising data to be destaged, metadata to be destaged and temporary data to be used in further processing, metadata to be used in further processing, and a combination thereof.
4. The system of claim 1 wherein the redundancy object is derived from the primary cache object as a serialized version of the primary cache object.
5. The system of claim 1 wherein the redundancy object is derived from the primary cache object as a compressed version of the primary cache object.
6. The storage system of claim 1 operable in accordance with at least one protocol selected from a group comprising file-access storage protocols, block-access storage protocols and object-access storage protocols.
7. The method of claim 1 wherein responsive to a failure of the first data server, the redundancy object is configured to be incorporated into the cache-object control procedures and the dirty data destaging procedures of the data second server.
8. A method of operating a storage system comprising a plurality of disk units adapted to store data at respective ranges of logical block addresses (LBAs), said addresses constituting an entire address space, and a storage control grid operatively connected to the plurality of disk units and comprising a plurality of data servers, the method comprising:
a) configuring a first data server to have a primary responsibility over a certain range of LBAs,
b) configuring a second data server to have a secondary responsibility over said certain range of LBAs and to overtake the primary responsibility over said certain range of LBAs if the first server fails;
c) responsive to a request to one or more LBAs within said certain range of LBAs, generating a primary cache object in the first data server and generating a redundancy object in the second data server, wherein said redundancy object is further configured to be handled in a searchable data structure, and wherein, responsive to a failure of the first server, the redundancy object is configured to be incorporated into cache-object control procedures and dirty data destaging procedures of the second server;
d) deleting said redundancy object upon successful permanent storing of data and/or metadata corresponding to said request, said storing providedby the first data server,
e) if operation d) fails, using said redundancy object by the second server for providing permanent storing said data and/or metadata when overtaking primary responsibility,
wherein said redundancy object is derivation of the primary cache object and need to be reconstructed in order to enable overtaking the responsibility.
9. The method of claim 8 wherein the request directed to one or more LBAs is selected from a group comprising respectively directed I/O requests and configuration-related requests.
10. The method of claim 8 wherein the primary cache object comprises data selected from a group comprising data to be destaged, metadata to be destaged and temporary data to be used in further processing, metadata to be used in further processing, and a combination thereof.
11. The method of claim 8 wherein operations of generating a primary cache object in the first data server, incorporating the primary cache object into cache-object control procedures and dirty data destaging procedures of the first server, generating a redundancy object in the second data server and adding the redundancy object to a searchable data structure in the second server are configured as an atomic operation with regard to an operation system running on the servers.
12. The method of claim 8 further comprising return acknowledgement to the request responsive to completion the following operations:
a) generating a primary cache object in the first data server,
b) incorporating the primary cache object into cache-object control procedures and dirty data destaging procedures of the first server,
c) generating a redundancy object in the second data server, and
d) adding the redundancy object to a searchable data structure in the second server.
13. The method of claim 8 wherein the redundancy object is derived from the primary cache object as a serialized version of the primary cache object.
14. The method of claim 8 wherein the redundancy object is derived from the primary cache object as a compressed version of the primary cache object.
15. The method of claim 8 wherein, responsive to a failure of the first server, the second server:
a) reconstructs the redundancy object into a copy of the primary object;
b) accommodates the reconstructed redundancy object within the cache; and
c) incorporates the reconstructed redundancy object into cache-object control procedures and dirty data destaging procedures.
16. A computer program embodied on a non transient computer readable medium comprising computer program code means for performing the method of claim 8 when said program is run on a computer.
17. The method of claim 8 wherein the redundancy object is handled in the searchable data structure with no need to be incorporated into the cache-object control procedures and/or the dirty data destaging procedures of the second data server, before overtaking the primary responsibility.