IBM DS6000 Series manual Failover, DS6800 normal data flow

NVS

for odd

LSSs

Cache

memory for even LSSs

Controller 0

NVS

for even

LSSs

Cache

memory

for odd

LSSs

Controller 1

Figure 3-1 DS6800 normal data flow

Figure 3-1illustrates how the cache memory of controller 0 is used for all logical volumes that are members of the even LSSs. Likewise, the cache memory of controller 1 supports all logical volumes that are members of odd LSSs. But for every write that gets placed into cache, another copy gets placed into the NVS memory located in the opposite controller. So the normal flow of data for a write is:

1.Data is written to cache memory in the owning controller.

2.Data is written to NVS memory of the alternate controller.

3.The write is reported to the attached host as having been completed.

4.The write is destaged from the cache memory to disk.

5.The write is then discarded from the NVS memory of the alternate controller.

Under normal operation, both DS6800 controllers are actively processing I/O requests. This section describes the failover and failback procedures that occur between the DS6800 controllers when an abnormal condition has affected one of them.

Failover

In the example depicted in Figure 3-2 on page 48, controller 0 in the DS6800 has failed. The remaining controller has to take over all of its functions. The host adapters located in controller 0 are now no longer available. All the RAID arrays in the DS6800 will be accessed from the device adapter in controller 1. First, controller 1 has to process the data it is holding in NVS. It then starts operating the entire machine in single controller mode. The steps it takes are:

1.It de-stages the contents of its NVS to disk.

2.The NVS and cache of controller 1 are divided in two, half for the odd LSSs and half for the even LSSs.

3.Controller 1 now begins processing the writes (and reads) for all the LSSs.

Chapter 3. RAS 47