[opensuse] New Subject - building clusters with OpenSuse

[Michael Folsom]

Thanks!

I checked Science and Productivity and its not there -

However, before we do that - it might be worthwhile to consider what
else do we need for clustering?  It would be really great to see
OpenSuse shipped on clusters of all sizes from the little guys to
biggies but to make that easy the missing parts should be supplied.

Here's my first cut at a list...
- hearbeat (not sure its totally necessary but should be semi-easy)
- OpenPBS   http://www.openpbs.org/
- Oscar  http://oscar.openclustergroup.org/
- OpenMPI  http://www.open-mpi.org/

What else?



Michael

On 6/15/07, Sunny <[EMAIL PROTECTED]> wrote:

On 6/14/07, Michael Folsom <[EMAIL PROTECTED]> wrote:
> Don't really think this is about yast - the rpm for heartbeat isn't
> included in OpenSuse 10.2 or 10.3 Alpha 4 and it is in SLES10.  If
> heartbeat isn't there you really don't need the yast2-heartbeat
> module.
>
> How does one petition the folks developing OpenSuse to include heartbeat?

 

--
Svetoslav Milenov (Sunny)

Even the most advanced equipment in the hands of the ignorant is just
a pile of scrap.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Greg Freemyer]

On 6/15/07, Michael Folsom <[EMAIL PROTECTED]> wrote:

Thanks!

I checked Science and Productivity and its not there -

However, before we do that - it might be worthwhile to consider what
else do we need for clustering?  It would be really great to see
OpenSuse shipped on clusters of all sizes from the little guys to
biggies but to make that easy the missing parts should be supplied.

Here's my first cut at a list...
- hearbeat (not sure its totally necessary but should be semi-easy)
- OpenPBS   http://www.openpbs.org/
- Oscar  http://oscar.openclustergroup.org/
- OpenMPI  http://www.open-mpi.org/

What else?


OCFS2  - (Oracle's Cluster FS) - it is already in the vanilla kernel
and included in SLES10.
http://oss.oracle.com/projects/ocfs2/dist/documentation/ocfs2_faq.html#SLES

drdb - part of SLES10 and integrates well with heartbeat.

Lustre - Don't know status wrt opensuse

OpenSSI - Lots of work since they are still at 2.6.11 kernel, but they
are going to do a kernel upgrade soon, but probably only to 2.6.16 or
so first.

OpenGFS - Don't know status wrt opensuse, but it is a Redhat team that
does most of the r&d.

Greg
--
Greg Freemyer
The Norcross Group
Forensics for the 21st Century
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Thomas Hertweck]

Michael Folsom wrote:
> I checked Science and Productivity and its not there -
> 
> However, before we do that - it might be worthwhile to consider what
> else do we need for clustering?  It would be really great to see
> OpenSuse shipped on clusters of all sizes from the little guys to
> biggies but to make that easy the missing parts should be supplied.

I think building, running, and maintaining an HPC Linux cluster requires
more than just a collection of software tools. Anyway...

> Here's my first cut at a list...
> - hearbeat (not sure its totally necessary but should be semi-easy)
> - OpenPBS   http://www.openpbs.org/
> - Oscar  http://oscar.openclustergroup.org/
> - OpenMPI  http://www.open-mpi.org/

http://ganglia.sourceforge.net/ (might be part of oscar)
http://munin.projects.linpro.no/
http://www.clusterresources.com/pages/products/torque-resource-manager.php
http://www.csm.ornl.gov/torc/C3/
...

-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Michael Folsom]

I absolutely agree -

I was just wondering aloud if the inclusion of a few basic tools may
make it an easier choice for the cluster building crowd.  There is no
way that any distro could do it all -


M-

On 6/15/07, Thomas Hertweck <[EMAIL PROTECTED]> wrote:



Michael Folsom wrote:
> I checked Science and Productivity and its not there -
>
> However, before we do that - it might be worthwhile to consider what
> else do we need for clustering?  It would be really great to see
> OpenSuse shipped on clusters of all sizes from the little guys to
> biggies but to make that easy the missing parts should be supplied.

I think building, running, and maintaining an HPC Linux cluster requires
more than just a collection of software tools. Anyway...

> Here's my first cut at a list...
> - hearbeat (not sure its totally necessary but should be semi-easy)
> - OpenPBS   http://www.openpbs.org/
> - Oscar  http://oscar.openclustergroup.org/
> - OpenMPI  http://www.open-mpi.org/

http://ganglia.sourceforge.net/ (might be part of oscar)
http://munin.projects.linpro.no/
http://www.clusterresources.com/pages/products/torque-resource-manager.php
http://www.csm.ornl.gov/torc/C3/
...

--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Alexey Eremenko]

Alternatively, you can write some custom scripts to share hard disk
data between nodes, use "webmin" to manage nodes together, and use
Cisco's NAT with Load Balancing to use both servers.

This should work for basic workloads... such as HTTP server.

I'm not sure how good or bad openSUSE is at "clustering" compared to
Debian and RedHat.

--
-Alexey Eremenko "Technologov"
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Pascal Bleser]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Alexey Eremenko wrote:
> Alternatively, you can write some custom scripts to share hard disk
> data between nodes, use "webmin" to manage nodes together, and use
> Cisco's NAT with Load Balancing to use both servers.

Scripts ? to "share" ?
Seems we've got quite a different definition of "sharing" here ;)
And I think you're really oversimplifying "clustering".

For sharing data between nodes, you definitely need shared storage
(FC-attached, possibly SCSI, or for testing purposes it even works with
Firewire) or a multipathed SAN. Needless to say, that's horrendously
expensive. And with directly shared storage options (i.e. not SAN), you
need a clustering-capable filesystem (GFS or OCFS2) that manages the
dlocks between the nodes.

A cheaper but less performant options is drbd. When possible, it's
advisable to avoiding shared storage completely, e.g. by storing data in
a replicated MySQL (when possible, depends on the application(s)).

Load balancing is the most trivial issue to solve.
High availability ("hot standby", aka "active-passive") isn't that easy
and, depending on the application and its needs, it can be very expensive.
Load balancing _and_ high availability ("active-active") is typically
even a lot more complex than "just" high availability.

If it's about "clustering" as in "HA clustering" (HA=High Availability)
(which has almost nothing to do with "HPC clustering" (HPC=High
Performance Computing), then heartbeat is also a very, very interesting
option.
It's limited to two nodes, but it's pretty easy to configure.
Pity, what it doesn't do is monitoring services (just the nodes). You'll
need another tool for that (e.g. monit or mon) and combine the two
(which, surprisingly, isn't that easy to do nor is there that much
documentation about it).

When it's more about load-balancing, LVS (Linux Virtual Server) is a
good candidate. As it operates the packet rewriting at kernel-level,
it's pretty fast. It doesn't combine all that easily with HA though.

> This should work for basic workloads... such as HTTP server.

That's just load-balancing, which is really trivial to do (with lots and
lots of options on how to implement it) for stateless protocols such as
HTTP.
It becomes more complex already when you have stateful (or
"conversational") protocols (EJB/JRMP/RMI, XMPP/Jabber, ...) or even
HTTP with sessions.
Unless you don't care about users having to start their shopping cart
all over again when a server dies, HTTP with sessions needs session
replication. Session replication can be achieved by persisting sessions
in a replicated database (slow), using e.g. Tomcat and its clustered
in-memory session replication mechanisms (fast), or doing "sticky
load-balancing" with a load-balancer upfront that is smart enough to
understand session URLs and cookies and to redirect traffic to the
cluster node that originated the session (fastest, but only
load-balances the first request and doesn't help all that much wrt HA).

cheers
- --
  -o) Pascal Bleser http://linux01.gwdg.de/~pbleser/
  /\\ <[EMAIL PROTECTED]>   <[EMAIL PROTECTED]>
 _\_v The more things change, the more they stay insane.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org

iD8DBQFGczP3r3NMWliFcXcRAgkRAKCokXIfuDwrH/TSTTqV62P7bJVCsgCfdGaE
DjgWqtFXzvZ4pEAvg9pKkWk=
=eV8u
-END PGP SIGNATURE-
-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Ciro Iriarte]

2007/6/15, Pascal Bleser <[EMAIL PROTECTED]>:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Alexey Eremenko wrote:
> Alternatively, you can write some custom scripts to share hard disk
> data between nodes, use "webmin" to manage nodes together, and use
> Cisco's NAT with Load Balancing to use both servers.

Scripts ? to "share" ?
Seems we've got quite a different definition of "sharing" here ;)
And I think you're really oversimplifying "clustering".

For sharing data between nodes, you definitely need shared storage
(FC-attached, possibly SCSI, or for testing purposes it even works with
Firewire) or a multipathed SAN. Needless to say, that's horrendously
expensive. And with directly shared storage options (i.e. not SAN), you
need a clustering-capable filesystem (GFS or OCFS2) that manages the
dlocks between the nodes.

A cheaper but less performant options is drbd. When possible, it's
advisable to avoiding shared storage completely, e.g. by storing data in
a replicated MySQL (when possible, depends on the application(s)).

Load balancing is the most trivial issue to solve.
High availability ("hot standby", aka "active-passive") isn't that easy
and, depending on the application and its needs, it can be very expensive.
Load balancing _and_ high availability ("active-active") is typically
even a lot more complex than "just" high availability.

If it's about "clustering" as in "HA clustering" (HA=High Availability)
(which has almost nothing to do with "HPC clustering" (HPC=High
Performance Computing), then heartbeat is also a very, very interesting
option.
It's limited to two nodes, but it's pretty easy to configure.
Pity, what it doesn't do is monitoring services (just the nodes). You'll
need another tool for that (e.g. monit or mon) and combine the two
(which, surprisingly, isn't that easy to do nor is there that much
documentation about it).

When it's more about load-balancing, LVS (Linux Virtual Server) is a
good candidate. As it operates the packet rewriting at kernel-level,
it's pretty fast. It doesn't combine all that easily with HA though.

> This should work for basic workloads... such as HTTP server.

That's just load-balancing, which is really trivial to do (with lots and
lots of options on how to implement it) for stateless protocols such as
HTTP.
It becomes more complex already when you have stateful (or
"conversational") protocols (EJB/JRMP/RMI, XMPP/Jabber, ...) or even
HTTP with sessions.
Unless you don't care about users having to start their shopping cart
all over again when a server dies, HTTP with sessions needs session
replication. Session replication can be achieved by persisting sessions
in a replicated database (slow), using e.g. Tomcat and its clustered
in-memory session replication mechanisms (fast), or doing "sticky
load-balancing" with a load-balancer upfront that is smart enough to
understand session URLs and cookies and to redirect traffic to the
cluster node that originated the session (fastest, but only
load-balances the first request and doesn't help all that much wrt HA).

cheers
- --
  -o) Pascal Bleser http://linux01.gwdg.de/~pbleser/
  /\\ <[EMAIL PROTECTED]>   <[EMAIL PROTECTED]>
 _\_v The more things change, the more they stay insane.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org

iD8DBQFGczP3r3NMWliFcXcRAgkRAKCokXIfuDwrH/TSTTqV62P7bJVCsgCfdGaE
DjgWqtFXzvZ4pEAvg9pKkWk=
=eV8u
-END PGP SIGNATURE-
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



The new Heartbeat2 with OCFSv2 would be a killer combination in some
future, the only problem with heartbeat2 is the lack of
documentation Maybe adding openSSI we could get something similar
to TruCluster...

Ciro
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Greg Freemyer]

On 6/15/07, Ciro Iriarte <[EMAIL PROTECTED]> wrote:


The new Heartbeat2 with OCFSv2 would be a killer combination in some
future, the only problem with heartbeat2 is the lack of
documentation Maybe adding openSSI we could get something similar
to TruCluster...

Ciro
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




OCFSv2 is part of the vanilla 10.2 distro. and is available via yast2.

And you can find heartbeat 2.0.7 at
http://download.opensuse.org/distribution/10.2/repo/oss/suse/i586/

I'm not sure why I'm not finding those same RPMs via yast2.

Greg
--
Greg Freemyer
The Norcross Group
Forensics for the 21st Century
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [opensuse] New Subject - building clusters with OpenSuse

[Patrick Kirsch]

Greg Freemyer wrote:
> On 6/15/07, Michael Folsom <[EMAIL PROTECTED]> wrote:
>> Thanks!
>>
>> I checked Science and Productivity and its not there -
>>
>> However, before we do that - it might be worthwhile to consider what
>> else do we need for clustering?  It would be really great to see
>> OpenSuse shipped on clusters of all sizes from the little guys to
>> biggies but to make that easy the missing parts should be supplied.
>>
>> Here's my first cut at a list...
>> - hearbeat (not sure its totally necessary but should be semi-easy)
>> - OpenPBS   http://www.openpbs.org/
>> - Oscar  http://oscar.openclustergroup.org/
>> - OpenMPI  http://www.open-mpi.org/
>>
>> What else?
http://openmosix.sourceforge.net/

openMosix is a Linux kernel extension for single-system image clustering
which turns a network of ordinary computers into a supercomputer.

'What is openMosix useful for?' /1/

openMosix allows you to join together multiple computers running the
Linux operating system, and have them appear to the user as one large
multiple-processor computer. For example, suppose you had two computers,
A and B joined in an openMosix cluster. Without openMosix, if you ran
two programs on A they would only get 50% of the CPU time each. With
openMosix, one of the programs could migrate 'automagically' to B, so
both processes would run at 100% CPU. As far as the user is concerned, A
now behaves like a two-CPU SMP computer with twice the CPU power available.



/1/
http://howto.x-tend.be/openMosixWiki/index.php/FAQ#.27What_is_openMosix.3F.27

Regards,
-- 
Patrick Kirsch - Quality Assurance Department
SUSE Linux Products GmbH GF: Markus Rex, HRB 16746 (AG Nuernberg)
-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]