Dell PowerVault MD3000i Руководство по настройке - Страница 31
Просмотреть онлайн или скачать pdf Руководство по настройке для Хранение Dell PowerVault MD3000i. Dell PowerVault MD3000i 32 страницы. Configuration for vmware esx server software
Также для Dell PowerVault MD3000i: Примечание по поддержке (22 страниц), Руководство по обновлению (28 страниц), Руководство пользователя (46 страниц), Руководство по замене (4 страниц), Руководство по конфигурации (10 страниц)
Печатать страницу
Dell™ PowerVault MD3000 and MD3000i Array Tuning Best Practices
Term
RAID
RAID 0
RAID 1/10
RAID 5
RAID 6
RPA
RMW
RMW2
SAS
SATA
Saturation
SCSI
December 2008 – Revision A01
Definition
directly on the drive in lieu of using an SATA interposer.
Redundant Array of Inexpensive Disks
RAID Level 0; RAID 0 is a striped set with no redundant
information. It is effectively a fully degraded RAID set
with no disk redundancy overhead.
RAID Level 1/10: The RAID 1/10 implementation on the
MD3000/MD3000i follows Berkley RAID 1 standard,
expanding it to a redundant N+N mirrored set.
Functionally, this implementation is set equivalent to a
generic nested RAID 1+0 and operates on as few as 2
physical drives. This allows for any number of drive
failures as long as one of each drive pair is available at
the cost of half the included physical drives.
RAID Level 5; Involves a block-striping algorithm where
n-1 disks per stripe in the raid set contain the data and
a parity disk P contains parity or checksum used to
validate data integrity provide consistency information.
The parity is distributed across all drives in a disk group
to provided further fault tolerance. RAID 5 provides
protection from single drive failure.
RAID Level 6; Involves a block-striping algorithm where
n-2 disks per stripe in the raid set contain the data and
a parity blocks P and Q contains the parity or
checksum used to validate data integrity provide
consistency information. The parity is distributed across
all drives in a disk group to provided further fault
tolerance. RAID 6 provides protection from double drive
failure.
Read Pre-fetch Algorithm: An acronym for the read
ahead cache used on the MD3000\MD3000i.
Read, Modify, Write; The second best algorithm
available in RAID 5 and RAID 6 write operations. A
RMW occurs when a quantity of bits, smaller or equal
to an individual segment are modified.
A Firmware Generation One adaptation of RMW
specifically for Write-Through conditions where write
cache is not enabled or requested to a virtual disk.
Serial Attached SCSI. Protocol maintained by t10.org
Serial Attached Technology Attachment, or Serial ATA.
This is the next stage of the legacy Parallel ATA.
Primarily in this white paper SATA refers to SATA HDD
technology.
See Controller Saturation
Small Computer System Interface, Protocol maintained
Page 31