EMC VMAX 100K, Руководство по эксплуатации продукта

Страница: 44 / 200

array subsystem, allowing the array to provide true end-to-end integrity checking and
protection against hardware or software faults.
The protection metadata is appended to the data stream, and contains information
describing the expected data location as well as CRC representation of the actual data
contents. The expected values to be found in protection metadata are stored
persistently in an area separate from the data stream. The protection metadata is used
to validate the logical correctness of data being moved within the array any time the
data transitions between protocol chips, internal buffers, internal data fabric
endpoints, system cache, and system drives.
Drive monitoring and correction
HYPERMAX OS monitors medium defects by both examining the result of each disk
data transfer and proactively scanning the entire disk during idle time. If a block on the
disk is determined to be bad, the director:
1. Rebuilds the data in the physical storage, if necessary.
2. Remaps the defect block to another area on the drive set aside for this purpose.
3. Rewrites the data from physical storage back to the remapped block on the drive.
4. Rewrites the data in physical storage, if necessary.
The director maps around any bad block(s) detected, thereby avoiding defects in the
media. The director also keeps track of each bad block detected on a drive. If the
number of bad blocks exceeds a predefined threshold, the VMAX array invokes a
sparing operation to replace the defective drive and then automatically alerts EMC
Customer Support to arrange for corrective action, if necessary. With the deferred
service sparing model, often times immediate action is not required.
Physical memory error correction and error verification
HYPERMAX OS corrects single-bit errors and report an error code once the single-bit
errors reach a predefined threshold. In the unlikely event that physical memory
replacement is required, the array notifies EMC support, and a replacement is ordered.
Drive sparing and direct member sparing
When HYPERMAX OS 5977 detects a drive is about to fail or has failed, a direct
member sparing (DMS) process is initiated. Direct member sparing looks for available
spares within the same engine that are of the same block size, capacity and speed,
with the best available spare always used.
With direct member sparing, the invoked spare is added as another member of the
RAID group. During a drive rebuild, the option to directly copy the data from the failing
drive to the invoked spare drive is supported. The failing drive is removed only when
the copy process is finished. Direct member sparing is automatically initiated upon
detection of drive-error conditions.
Direct member sparing provides the following benefits:
l
The array can copy the data from the failing RAID member (if available), removing
the need to read the data from all of the members and doing the rebuild. Copying
to the new RAID member is less CPU intensive.
l
If a failure occurs in another member, the array can still recover the data
automatically from the failing member (if available).
l
More than one spare for a RAID group is supported at the same time.
VMAX3 with HYPERMAX OS
44 Product Guide VMAX 100K, VMAX 200K, VMAX 400K with HYPERMAX OS