Fibre-Fibre RAID
Controllers
Overview
The NS-RC 2000 is the performance champion among
RAID controllers. Powered by a high-performance 64-bit
RISC PowerPC CPU and dual Infortrend sophisticated RAID
ASICs, I/Os are rapidly processed and distributed to array
hard drives. Data is transferred between host and drive
through a high-speed 64-bit data path at a burst rate
of 533MB/sec. The result is a controller which excels
in applications that require high sustained data rates
and I/O turnaround.
High Flexibility
All Fibre channels are configurable as either host or
drive in single or dual loop configuration. The active-active
configurations via single or multiple host ports allow
the optimal connectivity in a multi-host environment.
The Workload over dual-loop paths can be automatically
adjusted through intelligent optimization algorithms.
Firmware upgrade does not affect the redundant configuration
of controllers. Controllers can shift I/Os temporarily
to allow both firmware to be updated. In the event of
the unlikely controller failure, a failed controller
can be replaced with one running later version of firmware.
Sync. Cache over Fibre Channels
Besides SCSI channels, the intercontroller communications
can be NS-RC 2000R configured as using one or two
dedicated Fibre channels or sharing with all drive 6
Fibre Redundant loops together. The communications Channels
traffic can be automatically balanced between or among
loops. Should one communications path fail, the existing
channel(s) will continue its work.
SAN Application
The NS-RC 2000 has all the necessary protocols for SAN
applications. This includes the capabilities for operation
in a multi-host and multi-platform environment such
as LUN filtering (also known as SAN mapping or masking),
active-active failover /failback, and partitioning of
multiple-level RAID array. 64 array partitions and 128
filter entries per controller are supported.
High Data Availability
With a multitude of RAID functions, the user gets the
highest data availability and security. The NS-RC 2000’s
capabilities include automatic drive failure detection,
concurrent background rebuilding, and bad sector reassignment;
all without affecting host access.
Active-Active Redundancy
The NS-RC 2000 supports active-active redundancy with a
synchronized writeback cache. Integrators can choose
the two dedicated SCSI channels or Fibre channel(s)
as the cache synchronization paths. In redundant mode,
a failed controller automatically shifts I/O functions
to the existing controller, and the controller failover/failback
process is totally transparent to the host. The controller
itself is hot-swappable for full non-stop redundant
operation. A faulty controller can easily be ejected
from the front.
The optional battery backup protects any data cached
in memory from sudden power interruptions.
Board-Only Solutions for OEMs
The NS-RC 2000 controller is an ideal solution for OEM
customers who design their own backplanes. The controller
board can be readily integrated with a cable-less and
hot-swappable enclosure design.
Redundant Chassis (2000R)
The 2000R series consists of two controllers in a 5.25"
full-height chassis. All signal paths, including two
dedicated communications channels, are routed on the
back panel of the solution for path redundancy and cache
synchronization. Bypass circuits are included to maintain
loop integrity providing fabric connectivity and also
the freedom of controller hot-replacement. The cableless
design makes system integration effortless.
|
