Data center storage availability architecture using rack-level network fabric

1. A system, comprising:

a data center rack comprising:

a plurality of servers operable to provide data storage services via a network;

a plurality of data storage enclosures each comprising one or more data storage devices; and

a rack-level, storage networking fabric coupled to the servers and data storage enclosures, the storage networking fabric providing many-to-many virtual storage connections between the servers and the data storage enclosures, the virtual connections replacing high-availability hard links between the servers and the data storage enclosures;

wherein the system is configured to:

detect a failed of one of the servers;

determine a subset of the data storage enclosures previously connected to the failed server; and

establish a new virtual connection via the storage networking fabric between at least one other of the servers that has not failed and the subset of the data storage enclosures, the at least one other server taking over the data storage services on behalf of the failed server.

2. The system of claim 1 , wherein the at least one other server comprises two or more servers, and wherein the subset of data storage enclosures and associated workload of the failed server are distributed between the two or more servers.

3. The system of claim 1 , wherein the data center rack comprises a first bridge coupled to the storage networking fabric, the system further comprising a second data center rack comprising:

a plurality of second servers operable to provide second data storage services via the network;

a second rack-level, storage networking fabric coupled to the second servers; and

a second bridge coupled to the second rack-level, storage networking fabric and further coupled to the first bridge, wherein the first and second bridge are operable to establish the new virtual connection between at least one other of the second servers and the subset of the data storage enclosures.

4. The system of claim 1 , wherein the plurality of data storage enclosures each comprise two or more arrays each comprising a plurality of the one or more data storage devices, the two or more arrays presented as two or more virtual devices, and wherein the at least one other server comprises two or more servers, the two or more virtual devices being distributed between the two or more servers.

5. The system of claim 1 , wherein the storage networking fabric comprises an NVMeOF fabric.

6. The system of claim 5 , wherein the one or more data storage devices comprise hard disk drives, the system comprising one or more NVMeOF adapters that couple the hard disk drives to the NVMeOF fabric.

7. The system of claim 1 , wherein the storage networking fabric comprises an SAS switch.

8. A system, comprising:

a data center rack comprising:

a plurality of servers operable to provide data storage services via a network;

a plurality of data storage enclosures each comprising one or more data storage devices;

at least one spare data storage device; and

a rack-level, storage networking fabric coupled to the servers and data storage enclosures, the storage networking fabric providing many-to-many virtual storage connections between the servers and the data storage enclosures, the virtual connections replacing high-availability hard links between the servers and the data storage enclosures;

wherein the system is configured to:

detect a failed of one of the data storage devices within one of a selected one of the data storage enclosures, the selected data storage enclosure connected with a selected one of the servers; and

establish a new virtual connection via the storage networking fabric between the selected server and the at least one spare data storage device, wherein the at least one spare data storage device is not located in the selected data storage enclosure.

9. The system of claim 8 , wherein the data center rack comprises a first bridge coupled to the storage networking fabric, the system further comprising a second data center rack comprising:

a second spare data storage device;

a second rack-level, storage networking fabric coupled to the second spare data storage device; and

a second bridge coupled to the second rack-level, storage networking fabric and further coupled to the first bridge, wherein the first and second bridge are operable to establish the new virtual connection between the selected server and the second spare data storage device.

10. The system of claim 8 , wherein the storage networking fabric comprises an NVMeOF fabric.

11. The system of claim 10 , wherein the one or more data storage devices comprise hard disk drives, the system comprising one or more NVMeOF adapters that couple the hard disk drives to the NVMeOF fabric.

12. The system of claim 8 , wherein the storage networking fabric comprises an SAS switch.

13. A method comprising:

monitoring operations involving a plurality of servers coupled to a plurality of data storage enclosures via a rack-level, storage networking fabric, the storage networking fabric providing many-to-many virtual storage connections between the servers and the data storage enclosures, the virtual connections replacing high-availability hard links between the servers and the data storage enclosures, the servers operable to provide data storage services utilizing the data storage enclosures via a network, the data storage enclosures each comprising one or more data storage devices, the servers and the data storage enclosures mounted within a data center rack;

determining a failed unit comprising a failed one of the servers or a failed one of the data storage devices within a selected one of the data storage enclosures;

determining a replacement for the failed unit within the data center rack; and

establishing a replacement for the failed unit within the data center rack, the establishment of the replacement comprising establishing a new virtual connection via the storage networking fabric.

14. The method of claim 13 , wherein the failed unit comprises the failed server, the establishing of the replacement comprises establishing multiple new virtual connections via the storage networking fabric between two or more the servers that has not failed and a subset of the data storage enclosures associated with the failed server, and wherein the subset of data storage enclosures and associated workload of the failed server are distributed between the two or more servers.

15. The method of claim 13 , wherein the failed unit comprises the failed data storage device, the establishing of the replacement comprises establishing the new virtual connection via the storage networking fabric between a selected server coupled to the selected data storage enclosure and a spare data storage device of the data center rack.

16. The method of claim 13 , wherein the storage networking fabric comprises an NVMeOF fabric.

17. The method of claim 16 , wherein the one or more data storage devices comprise hard disk drives that are coupled to the NVMeOF fabric via one or more NVMeOF adapters.

18. The method of claim 13 , wherein the storage networking fabric comprises an SAS switch.

19. The method of claim 13 , wherein the data center rack comprises a first bridge coupled to the storage networking fabric, the system further comprising a second data center rack comprising:

a spare data storage device;

a second rack-level, storage networking fabric coupled to the spare data storage device; and

a second bridge coupled to the second rack-level, storage networking fabric and further coupled to the first bridge, wherein the new connection is between a server of the data storage rack and the spare data storage device.

20. The system of claim 1 , wherein the system is configured to detect a failed of one of the data storage devices within one of a selected one of the data storage enclosures, wherein the data center rack comprises a first bridge coupled to the storage networking fabric, the system further comprising a second data center rack comprising:

a spare data storage device;

a second rack-level, storage networking fabric coupled to the spare data storage device; and

a second bridge coupled to the second rack-level, storage networking fabric and further coupled to the first bridge, wherein the first and second bridge are operable to establish a connection between one of the servers of the data center rack and the spare data storage device.