About difference between log shipping standby and active/passive c

Hi Guys,
What is difference between
log shipping standby and active/passive cluster? is there any difference for
the functionality? Thanks
Hi
In general an active/passive cluster will use SAN technology to replicate
your database from the active cluster to the passive one. The SAN is
replicated between two sites and all the databases etc on the SAN will be
copied. You may have local failover to a second local machine in the cluster
and remote failover to the remote cluster, giving a high degree of
resilience. The cost of this type of solution is usually quite high.
Log shipping is between two databases, not necessarily on different machines
or clusters and the logs will be shipped over the network. As this is a
database level it will need to be configured on all databases that you need
to copy. This solution is usually relatively cheap, but can still give you a
high level of resilience
HTH
John
"Iter" wrote:

> Hi Guys,
> What is difference between
> log shipping standby and active/passive cluster? is there any difference for
> the functionality? Thanks
>
|||An active passive cluster stores the actual data on a shared storage space,
usually a SAN. The active node accesses the data until it fails over at
which point the passive node accesses the data and becomes the active node.
If your shared storage area dies, your data is gone unless you have protected
it some other way. Clustering is more concerned with preventing downtime if
the failure is outside of the storage system.
Log shipping takes transaction log backups of one database and applies them
to a second database, effectively synching this second database to the first.
If the server that the primary database is on fails, you can activate the
secondary database to become the primary. Log shipping is often used to copy
data to a disaster recovery server and is more concerned with duplicating the
actual data.
Just a quick summary, and the emphasis of each method is my own opinion.
I hope that this helps,
Matt
MattMoneyDC @. Hotmail dot com
"Iter" wrote:

> Hi Guys,
> What is difference between
> log shipping standby and active/passive cluster? is there any difference for
> the functionality? Thanks
>
|||Hi
Something I forgot to say, was that you can also use both together, or you
could have two clusters that used log shipping to copy the database changes
between the
active and passive clusters instead of using SAN replication.
John
"John Bell" wrote:
[vbcol=seagreen]
> Hi
> In general an active/passive cluster will use SAN technology to replicate
> your database from the active cluster to the passive one. The SAN is
> replicated between two sites and all the databases etc on the SAN will be
> copied. You may have local failover to a second local machine in the cluster
> and remote failover to the remote cluster, giving a high degree of
> resilience. The cost of this type of solution is usually quite high.
> Log shipping is between two databases, not necessarily on different machines
> or clusters and the logs will be shipped over the network. As this is a
> database level it will need to be configured on all databases that you need
> to copy. This solution is usually relatively cheap, but can still give you a
> high level of resilience
> HTH
> John
>
> "Iter" wrote:
|||"John Bell" <jbellnewsposts@.hotmail.com> wrote in message
news:B23570F5-7A32-4CAF-A788-DE8FE76CA973@.microsoft.com...
> Hi
> In general an active/passive cluster will use SAN technology to replicate
> your database from the active cluster to the passive one. The SAN is
> replicated between two sites and all the databases etc on the SAN will be
> copied. You may have local failover to a second local machine in the
> cluster
> and remote failover to the remote cluster, giving a high degree of
> resilience. The cost of this type of solution is usually quite high.
>
What you describe is a "Geo-Cluster", a special and especially expensive
type of clustering solution. In a "normal" MSCS Failover Cluster the SQL
Server binaries for the instance are installed on multiple computers (called
cluster nodes), but only running on one node at a time. A single copy of
the database resides on shared storage (SAN or SCSI), and in case of failure
of one node, the other node mounts the volumes where the databases live and
starts up the SQL Server Instance.
The cost is still much higher than Log Shipping, both for the hardware, the
setup and the administration. But it supports automatic failover, manual
failover, and rolling upgrades/patches.

> Log shipping is between two databases, not necessarily on different
> machines
> or clusters and the logs will be shipped over the network. As this is a
> database level it will need to be configured on all databases that you
> need
> to copy. This solution is usually relatively cheap, but can still give you
> a
> high level of resilience
>
Also note that SQL Server 2005's Database Mirroring can be thought of as
"Continuous Log Shipping". It's just as easy, has less latency, supports
simple failover and failback.
In general see
Microsoft SQL Server High Availability
http://www.microsoft.com/sql/technologies/highavailability/default.mspx
How to Attain SQL Server High Availability at Minimal Cost
http://www.sql-server-performance.com/sql_server_high_availability.asp
SQL Server 2005 Mission Critical High Availability
http://www.microsoft.com/technet/prodtechnol/sql/themes/high-availability.mspx
David
|||According john's notes, I can use log shipping work together with
active/passive cluster. How about active/active cluster? can it work together
with log shipping? Thanks.
"John Bell" wrote:
[vbcol=seagreen]
> Hi
> Something I forgot to say, was that you can also use both together, or you
> could have two clusters that used log shipping to copy the database changes
> between the
> active and passive clusters instead of using SAN replication.
> John
> "John Bell" wrote:
|||Hi
Active/Active clusters are usually two instances one active on one cluster
with a passive failover on the other and the second instance active on the
other cluster with failover on the cluster where the other instance is
active. Therefore you may want to consider them as two active/passive
clusters thing sharing hardware.
John
"Iter" wrote:
[vbcol=seagreen]
> According john's notes, I can use log shipping work together with
> active/passive cluster. How about active/active cluster? can it work together
> with log shipping? Thanks.
> "John Bell" wrote: