I/O groups and uninterruptible power supply Planning guidelines for using your svc in a SAN environment Uninterruptible power supplyInternal battery backup and power domains

Each pair of nodes is known as an I/O group. Each node can only be in one I/O group. The I/O groups are connected to the SAN so that all backend storage and all application servers are visible to all of the I/O groups. Each pair has the responsibility to serve I/O on a particular virtual disk.

Virtual disks are logical disks that are presented to the SAN by SAN Volume Controller nodes. Virtual disks are also associated with an I/O group. The SAN Volume Controller does not contain any internal battery backup units and therefore must be connected to an uninterruptible power supply to provide data integrity in the event of a cluster wide power failure.

When an application server performs I/O to a virtual disk, it has the choice of accessing the virtual disk via either of the nodes in the I/O group. A virtual disk can have a preferred node specified when it is created. This is specified once the virtual disk is created. This is the node through which a virtual disk should normally be accessed. As each I/O group only has two nodes, the distributed cache in the SAN Volume Controller need only be two-way. When I/O is performed to a virtual disk, the node that processes the I/O duplicates the data onto the partner node that is in the I/O group.

I/O traffic for a particular virtual disk is, at any one time, handled exclusively by the nodes in a single I/O group. Thus, although a cluster may have many nodes within it, the nodes handle I/O in independent pairs. This means that the I/O capability of the SAN Volume Controller scales well, since additional throughput can be obtained by adding additional I/O groups.

The figure below shows an example I/O group. A write operation from a host is shown (item 1), that is targeted for virtual disk A. This write is targeted at the preferred node, Node 1 (item 2). The write is cached and a copy of the data is made in the partner node, Node 2's cache (item 3). The write is now complete so far as the host is concerned. At some time later the data is written, or destaged, to storage (item 4). The figure also shows two uninterruptible power supply units (1 and 2) correctly configured so that each node is in a different power domain.

Figure 32. I/O group and uninterruptible power supply
san7

When a node fails within an I/O group, the other node in the I/O group will take over the I/O responsibilities of the failed node. Data loss during a node failure is prevented by mirroring the I/O read/write data cache between the two nodes in an I/O group.

If only one node is assigned to an I/O group, or a node has failed in an I/O group, the cache goes into write-through mode. Therefore, any writes for the virtual disks that are assigned to this I/O group is not cached, it is sent directly to the storage device. If both nodes in an I/O group go offline, the virtual disks that are assigned to the I/O group cannot be accessed.

When a virtual disk is created, the I/O group that will provide access to the virtual disk must be specified. However, virtual disks can be created and added to I/O groups that contain offline nodes. I/O access will not be possible until at least one of the nodes in the I/O group is online.

The cluster also provides a recovery I/O group. This is used when both nodes in the I/O group have suffered multiple failures. This allows you to move the virtual disks to the recovery I/O group and then into a working I/O group. I/O access is not possible when virtual disks are assigned to the recovery I/O group.

(C) Copyright International Business Machines Corporation 2003, 2004