Please note that JavaScript and style sheet are used in this website,
Due to unadaptability of the style sheet with the browser used in your computer, pages may not look as original.
Even in such a case, however, the contents can be used safely.
All the components such as adapters for host communications (HA), disk drive access adapters (DA), Xbar, cache memory, power supplies, fans, disk drives, maintenance processors, and clocks are duplexed; therefore, failure of a single part will not cause the system to stop. System operations are further secured by duplex fiber channel interfaces on the paths to disk drives.
Now that the disk capacity is 100 times larger than what it was 10 years ago, new technologies are required to handle it. RAID-6 creates two independent parity generation equations based on data information to record individual parities (P and Q). By doing so, it is possible to restore a disk while maintaining redundancy, and furthermore, it is possible to handle the situation even when two disks fail at the same time.
For this reason, RAID-6 can provide reliability that is over 1,000 times higher than RAID-5 and over 100 times higher than RAID-1.
Also, in performance-focused systems such as S2800, different types of RAID such as RAID-6 and RAID-1 can be mixed.
With RAID-6, storage reliability and availability are dramatically improved.
To borrow VALUMO's terms, it can be described as "autonomy."
In the NEC Storage D/E series, complicated calculations for the above purposes are performed by an LSI, and this LSI is patent pending.
When there is a delay in disk responses for some reason, that disk is isolated and its operations are automatically checked. Meanwhile, other disks make responses instead of that isolated disk; therefore, access delays from the server can be minimized. If there are no abnormalities in the isolated disk as a result of self-diagnosis, it is automatically returned to its original configuration.
