Power failure management in components of storage area network
Abstract
A storage area network (SAN) is provided with redundancy and recovery mechanism. A primary storage switch performs dynamic address translation between logical storage addresses received from host devices and physical addresses of SAN storage arrays. When power failure in the primary storage switch is detected, metadata associated with the dynamic address translation operation is sent to a secondary storage switch via a network connection to provide the dynamic address translation in lieu of the primary storage switch. A storage array experiencing power failure similarly sends cached data to another storage array via a network connection so that the other storage array can substitute the failed storage array. During the power failure, a data backup module in the primary storage switch or the storage array is powered by a temporary power source.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1 . A method for managing access by a host device to data over a network, comprising:
responsive to failure of a power source of a primary storage switch, receiving mapping information that describes conversions of logical addresses to physical addresses of the data from the primary storage switch at a secondary storage switch; sending an indication to the host device that subsequent requests for the data to the primary storage switch be sent to the secondary storage switch; receiving, by the secondary storage switch, a first request from the host device to read data from or write data to one of the storage devices over a network, the first request identifying a location of the data by a logical address; converting, by the secondary storage switch, the logical address in the first request into a physical address by referencing the received mapping information; and sending, by the secondary storage switch, a second request including the physical address of data to the one of the storage devices.
2 . The method of claim 1 , further comprising:
sending, to the primary storage switch, updates to the mapping information after the secondary storage switch received the mapping information, responsive to restoring of the power source of the primary storage switch.
3 . The method of claim 1 , further comprising:
deploying the received mapping information in memory of the secondary storage switch.
4 . The method of claim 1 , wherein the indication sent to the host device includes a network address of the secondary storage switch that allows the host device to send requests to the secondary storage switch.
5 . The method of claim 1 , further comprising:
storing the mapping information in a data backup module of the secondary storage switch.
6 . The method of claim 5 , further comprising:
detecting, by the secondary storage switch, failure of a power source of the secondary storage switch; operating the data backup module using a second power source that is separate from the power source of the secondary storage switch; determining whether another storage switch distinct from the primary storage switch is available to receive the mapping information; and sending the mapping information stored in the data backup module to the another storage switch responsive to the another storage switch being available to receive the mapping information.
7 . The method of claim 5 , wherein the data backup module is implemented as an expansion card mounted in a slot of the secondary storage switch.
8 . The method of claim 1 , further comprising:
sending updates in the mapping information to another storage switch periodically before detecting failure of a power source of the secondary storage switch; and responsive to detecting failure of the power source of the secondary storage switch, sending incremental changes in the mapping information since previous transmission of the updates to the another storage switch.
9 . The method of claim 1 , wherein the first request is communicated over the network using AT Attachment over Ethernet (AoE) protocol.
10 . The method of claim 1 , wherein the mapping information is received from the primary storage switch via a communication channel separate from the network.
11 . The method of claim 1 , wherein the primary storage switch, secondary storage switch, and the storage device implement a Storage Area Network (SAN).
12 . A storage switch for managing access by a host device to data over a network, the storage switch comprising:
a processor; and a computer-readable storage medium storing executable code, the code when executed by the processor performs steps comprising:
responsive to failure of a power source of another storage switch, receiving mapping information that describes conversions of logical addresses to physical addresses of the data from the another storage switch at the storage switch;
sending an indication to the host device that subsequent requests for data to the primary storage switch be sent to the secondary storage switch;
receiving a first request from the host device to read data from or write data to one of the storage devices over a network, the first request identifying a location of the data by a logical address;
converting the logical address in the first request into a physical address by referencing the received mapping information; and
sending a second request including the physical address of data to the one of the storage devices.
13 . The secondary storage switch of claim 12 , wherein the code when executed by the processor performs further steps comprising:
sending, to the another switch, updates to the mapping information after the storage switch received the mapping information, responsive to restoring of the power source of the primary storage switch.
14 . The secondary storage switch of claim 12 , wherein the indication sent to the host device includes a network address of the storage switch that allows the host device to send requests to the storage switch.
15 . The secondary storage switch of claim 12 , further comprising a data backup module, the data backup module storing the mapping information.
16 . The secondary storage switch of claim 15 , wherein the code when executed by the processor performs further steps comprising:
detecting failure of a power source of the storage switch; operating the data backup module using a second power source that is separate from the power source of the storage switch; determining whether a third storage switch distinct from the another storage switch is available to receive the mapping information; and sending the mapping information stored in the data backup module to the third storage switch responsive to the third storage switch being available to receive the mapping information.
17 . The secondary storage switch of claim 15 , wherein the data backup module is implemented as an expansion card mounted in a slot of the secondary storage switch.
18 . The secondary storage switch of claim 12 , wherein the code when executed by the processor performs further steps comprising:
sending updates in the mapping information to a third storage switch periodically before detecting failure of a power source of the storage switch; and responsive to detecting failure of the power source of the storage switch, sending incremental changes in the mapping information since previous transmission of the updates to the third storage switch.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.