Storage: NetApp filer, known also as NetApp
Fabric-Attached Storage (FAS).
Is
an enterprise-class storage area network (SAN) as well as a networked storage
appliance (NAS). It can serve storage over a network using file-based protocols
such as NFS, CIFS, FTP, TFTP, and HTTP. And can serve data over block-based
protocols such as FC, RcoE and iSCSi.
The
most common NetAPP configuration consists of a filer (also known as a controller
or head node) and disk enclosures (also known as shelves).
Filers
are connected with two cluster interconnect cables (InfiniBand). This
interconnect is used for HA heartbeat.
Each
Controller is assigned Disks, which can be assigned from Different Disk Shelves
to One Controller. Each Controller should have a HOT SPARE Disk of Each Type.
Both Controllers (FILERS) Keep a SYNC of each other’s configuration to take over
in case of a Filer (Controller) Failure.
The filers run
NetApp's own adapted operating system called Data ONTAP, it is highly tuned for
storage-serving purposes. The Data ONTAP operating system implements a single
proprietary file-system called WAFL (Write Anywhere File
Layout):
WAFL provides
mechanisms that enable a variety of file systems and technologies that want to
access disk blocks.
As the name suggests Write Anywhere File Layout
does not store data or metadata in pre-determined locations on disk in a way
designed to minimize the number of disk operations required to commit data to
stable disk storage using single and dual parity based
RAID.
Disks: SATA, FC, SAS and SSD disks (7.2Krpm up to
15krpm) (100GB SSD up to 2TB SATA)
There are only four
types of disks in Data ONTAP:
|
Data
|
holds data stored
within the RAID group
|
|
Spare
|
Does not hold data
but is available to be added to a RAID group in an aggregate.
|
|
Parity
|
Store data
reconstruction information within the RAID
group
|
|
dParity
|
Stores
double-parity information within the RAID group, if RAID-DP is enabled
|
A disk can only be
in one aggregate. So each aggregate has its own drives. This lets us tune the
performance of the aggregate by adding many spindles.
Raid:
RAID
in NetApp terminology is called RAID
group. NetApp works mostly with RAID 4 and RAID-DP. Where
RAID 4 has one separate disk for parity and RAID-DP has two. Do not think that
it leads to performance degradation. NetApp has very efficient implementation of
these RAID levels.
We
do not build raid groups; they are built behind the scene when you build an
aggregate. Raid groups can be adjusted in size. For FC/SAS they can be anywhere
from 3 to 28 disks. Large raid-group: Better Space utilization, better
performance, more risk (slower rebuilds and ratio of parity to data
disks).
Aggragate: An aggregate is a
raw space. It is made up of one or more raid groups of disks, into a pool of disk space that can be used
to create multiple volumes.
You take a bunch of
individual disks and aggregate them together into aggregates. then layer on
partitions, which in NetApp land are called volumes. The volumes hold the data.
Aggregates
can SPAN Multiple RAID Groups on the SAME Filer. The more disks an
Aggregate has, the better performance it gets
(spindles).
A
RAID Group Cannot SPAN Controllers, so Aggregates Cannot SPAN Controllers
either. Aggregates can SPAN Multiple RAID Groups on the SAME Filer.
Volume:
the volumes hold the data, providing file system. When you share out a volume it
looks like NTFS to a windows box, or it looks like a UNIX filesystem to a unix
box but in the end its just WAFL in the volume. you first make a volume, then
you put a LUN in the volume. the LUN looks like one big file in the
volume.
Qtree: A qtree is similar to a subdirectory. why use
them? to sort data. There are 5 things you can do with a qtree: Oplocks,
security style, Quotas, Snapvault, Qtree SnapMirror.
Lun:
(logical
unit number of block storage arrays)
is
a logical representation of storage. It looks like a hard disk to the client. It
looks like a file inside of a volume. LUNs look like local disks to the OS.
LUN
is
necessary to access data via block-level protocols like FCP and
iSCSI.
In
the end its normally the application that will determine whether you get your
filesystem access through a LUN or a Volume. Some apps will not work across a
network; Microsoft SQL and Exchange are two examples of this. Volumes are access
via NAS protocols, CIFS/NFS. LUNS are accessed via SAN protocols,
iSCSI/FCP/FcoE.
Some more interesting (important)
entries:
· Data ONTAP writes
all data to a storage system in 4-KB
blocks.· Vol0: Root Volume. was created when the storage system was initially setup at the factory . contains special directories and configuration files.
· Igroup: An initiator group specifies which initiators can have access to a LUN. When you map a LUN on a storage system to the initiator group, you grant all the initiators in that group access to that LUN. If a host is not a member of an igroup that is mapped to a LUN, that host does not have access to the LUN.
In the case of
iSCSI clients, hosts are identified in an initiator group by their node names.
In the case of FCP clients, hosts are identified in an initiator group by their
World Wide Port Names (WWPNs).
· NVRAM:
Non-volatile battery-backed
memory (NVRAM) is used for write caching. Before going to hard drives all writes
are cached in NVRAM. NVRAM memory is split in half and each time 50% of NVRAM
gets full, writes are being cached to the second half, while the first half is
being written to disks. When data has been written to disks as part
of Consistency Point (CP), write blocks which were cached in main memory become
the first target to be evicted and replaced by other
data.
·
Plex
is collection of RAID groups and is used for RAID level mirroring. For instance
if you have two disk shelves and SyncMirror license then you can create plex0
from first shelf drives and plex1 from second shelf. This will protect you from
one disk shelf failure.
No comments:
Post a Comment