E20-522 CLARiiON Solutions Specialist for Storage Administrator Certification Boot Camp

Demo Question 19.

Answer: B,D

Explanation: When a write request is sent from the host to a clariion array, the SP that owns the LUN accepts the request. Once the primary SP has put that request into write cache, the data is sent over to the secondary SP via the CMI (Clariion messenging interface). Once the information has been written into write cache on the seconday SP it acknowledges to the primary SP that it has the information in cache. Once both SP's have the information in cache, the SP that owns the LUN sends an acknowledge to the host, informing it that the data has been written. At this point the data is only in cache. If the primary SP fails, the secondary SP has the same data in the write cache. The reason for both SP's having the same amount of write cache is for redundancy. All data that is in the write cache of the primary SP is also written to the write cache of the secondary SP before the host receives an ackowledgement. In the event of a complete system failure, only the DPE (which includes the SP's) and the first DAE continue to operate with the Standby power supplies, until all the data in the write cache is written to a vault on the disks. Once the power is restored the data from the vault if read from disk back into write cache, and the data is flushed to disk. This prevents data loss in a complete power failure.