Information Technology Reference
In-Depth Information
Storage Processors
Different vendors have different names for storage processors, which
are considered the brains of the array. They handle the I/O and run the array software. In
most modern arrays, the storage processors are not purpose-built application-specii c inte-
grated circuits (ASICs) but instead are general-purpose CPUs. Some arrays use PowerPC,
some use specii c ASICs, and some use custom ASICs for specii c purposes. But in general, if
you cracked open an array, you would most likely i nd an Intel or AMD CPU.
Array Software
Although hardware specii cations are important and can dei ne the scaling
limits of the array, just as important are the functional capabilities the array software pro-
vides. The capabilities of modern storage arrays are vast—similar in scope to vSphere itself—
and vary wildly among vendors. At a high level, the following list includes some examples of
these array capabilities and key functions:
◆
Remote storage replication for disaster recovery. These technologies come in many
l avors with features that deliver varying capabilities. These include varying recovery
point objectives (RPOs)—which rel ect how current the remote replica is at any time,
ranging from synchronous to asynchronous and continuous. Asynchronous RPOs
can range from less than minutes to more than hours, and continuous is a constant
remote journal that can recover to varying RPOs. Other examples of remote replica-
tion technologies are technologies that drive synchronicity across storage objects (or
“consistency technology”), compression, and many other attributes, such as integra-
tion with VMware vCenter Site Recovery Manager.
Snapshot and clone capabilities for instant point-in-time local copies for test and
development and local recovery. These also share some of the ideas of the remote
replication technologies like “consistency technology,” and some variations of point-
in-time protection and replicas also have TiVo-like continuous journaling locally and
remotely where you can recover/copy any point in time.
◆
Capacity-reduction techniques such as archiving and deduplication.
◆
Automated data movement between performance/cost storage tiers at varying levels
of granularity.
◆
LUN/i le system expansion and mobility, which means reconi guring storage proper-
ties dynamically and nondisruptively to add capacity or performance as needed.
◆
Thin provisioning, which typically involves allocating storage on demand as applica-
tions and workloads require it.
◆
Storage quality of service (QoS), which means prioritizing I/O to deliver a given
MBps, IOPS, or latency.
The array software dei nes the “persona” of the array, which in turn impacts core concepts
and behavior. Arrays generally have a “i le server” persona (sometimes with the ability to
do some block storage by presenting a i le as a LUN) or a “block” persona (generally with no
ability to act as a i le server). In some cases, arrays are combinations of i le servers and block
devices.
Cache Memory
Every array differs as to how cache memory is implemented, but all have
some degree of nonvolatile memory used for various caching functions—delivering lower
latency and higher IOPS throughput by buffering I/O using write caches and storing com-
monly read data to deliver a faster response time using read caches. Nonvolatility (meaning
◆
