[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-ha-dev
Subject:    Re: [Linux-ha-dev] (no subject)
From:       Joe DiMartino <joe () osdl ! org>
Date:       2002-08-28 22:24:32
[Download RAW message or body]

On Wed, 2002-08-28 at 09:12, jayesh s kitukale wrote:
> 1> why shared disk is more preferred in high availability systems 
> rather than using two separate disks and carrying out lan 
> mirroring .

The important thing in HiAv is redundancy of the data to eliminate
any single point of failure.  Many systems that care about their
data are already running some form of RAID software (like mirroring)
or RAID hardware for protection.  It is only a small step from there
to add a "shared" connection and allow access ownership to fail-over to
another computer.  This eliminates a second point of failure if the
computer breaks before a disk does.  (Note that it's a much bigger
step to provide full concurrent access to the data to make it really
"shared".)

Another way to provide redundant data is replication over a network (as
you suggest).  Replication becomes less likely as the data amount
increases and/or the frequency of updates increase.  For example, you're
more likely to find replicated read-only web pages than a multi-terabyte
frequently updated database.  Replication increases the tolerance to
withstand a site catastrophe but it may also increase the complexity and
cost of the solution (RAID hardware at all replicas, dedicated network
with high bandwidth and low latency, etc).

- Joe DiMartino


_______________________________________________________
Linux-HA-Dev: Linux-HA-Dev@lists.community.tummy.com
http://lists.community.tummy.com/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
[prev in list] [next in list] [prev in thread] [next in thread]