Active/active and active/passive

To:

 

Q: What is the difference between active/active and active/passive arrays? Which technology is better?

A:

 

Storage arrays (SCSI, FC, iSCSI, and NAS) with two or more controllers can be configured in active/active mode or active/passive mode.

Active/passive means that one controller actively processes I/O requests, and the other is idle for taking over the work when the master controller fails or is offline. Active/active configurations enable both controller nodes to process I/O requests and provide redundant performance for each other.

Normally, the active/active storage system contains an image cache supported by the battery. The cached content in the controller is completely mirrored to another controller and can ensure its availability.

For example, if the active/active controllers share a 4 GB cache (2 GB per Controller), the number of images that can be used is 50%, that is, 2 GB.

Generally, a specific controller is used to access the lun to maintain cache consistency. Some controllers can also enable any controller to respond to the server's I/O requests to the Lun.

However, the actual I/O is usually done through a specified controller.

Which method is better? If you can provide an idle controller, the active/passive mode configuration will provide an expected performance advantage for failover.
Many NAS vendors use this model, which is generally called a NAS cluster. Active/active configurations enable both controllers in active mode to improve the performance of conventional operations. However, in case of failover, this mode may cause performance degradation.