SAN - Redundancy and Resiliency Explained

Abstract

SAN terminology - are you redundant or resilient or both?

Contents

Here are four designs that provide redundancy and resiliency in your SAN.

Redundancy and resiliency
When designing the SAN, you should also provide redundancy and resilience in
your design. Redundancy is the duplication of components up to and including
the entire fabric to prevent failure of the total SAN solution. Resiliency is the
capability of a fabric topology to withstand failures. We can group SAN designs
into four types:

1. Single fabric non resilient design
All fabric components are connected together in a single fabric, and there is at
least one single point-of-failure.

If one switch in the single fabric of the SAN fails, we will lose connection from the
top to the bottom of the fabric. So if, for example, we have a server connected to
the top switch which wants to access the data on the storage device connected
to the bottom switch, this will not be possible in this example after the failure.
We have introduced a single point-of-failure in our SAN design.

This image shows a single fabric non resilient  design

2. Single fabric resilient
All fabric components are connected together in a single fabric, but we do not
have any single point-of-failure.

If one of the switches in the single fabric SAN fails, we can still access the storage
devices connected to the bottom tier of switches from the servers connected to the
top tier of switches.
This image shows a single fabric resilient  design

3. Redundant fabric non resilient
The components in the SAN are duplicated into two independent fabrics. But we
still have a single point-of-failure in at least one of them. This type of design can
be used in combination with dual attached servers and storage devices. This will
keep the solution running even if one fabric fails.

Even if one of the switches in the SAN fabric failed, we can still access the
storage device in the bottom tier from the server at the top tier. Even though the
fabric itself is not resilient, the data path availability is ensured through the
redundant fabric.
This image shows a redundant fabric non resilient  design

4. Redundant fabric resilient
The components in the SAN are duplicated into two independent fabrics. There
is no single point-of-failure in either one of them. This type of design can be used
in combination with dual attached servers and storage devices. This will keep the
solution running even if one complete fabric was to fail.

Even if one of the switches in the SAN fabric failed, we can still access the
storage device on the bottom tier from the server at the top tier. With this type of
design we are basically protecting at two levels. First, we are protecting against
switch failure and secondly, we are protecting against a failure of the whole
fabric.
This image shows a redundant fabric resilient  design

Special Notices

This material has not been submitted to any formal IBM test and is published AS IS. It has not been the subject of rigorous review. IBM assumes no responsibility for its accuracy or completeness. The use of this information or the implementation of any of these techniques is a client responsibility and depends upon the client's ability to evaluate and integrate them into the client's operational environment.

Profile

Publish Date
26 August 2002

Last Update
07 August 2003


Rating: Not yet rated


Author(s)

IBM Form Number
TIPS0033