[Beowulf] Re: MySQL Cluster with SCI interconnect (fwd from tamada at acornnetworks.co.jp)

[Eugen Leitl]

Of tenuous relevance, but it's a cluster database which uses SCI for
signalling fabric.

MySQL 4.1 (just out) actually has cluster functionality built-in (but for Sun
Solaris, which is a release bug, about to be fixed). 

For those of you who're interested in HA Linux, the next release is not far
away, and removes the 2-node limitation: http://linuxha.trick.ca/

----- Forwarded message from Junzo Tamada <tamada at acornnetworks.co.jp> -----

From: "Junzo Tamada" <tamada at acornnetworks.co.jp>
Date: Thu, 28 Oct 2004 16:00:53 +0900
To: "Hugo Kohmann" <hugo at dolphinics.no>
Cc: "Hugo Kohmann" <hugo at dolphinics.no>, <cluster at lists.mysql.com>
Subject: Re: MySQL Cluster with SCI interconnect
Reply-To: "Junzo Tamada" <tamada at acornnetworks.co.jp>
X-Mailer: Microsoft Outlook Express 6.00.2900.2180

Dear Hugo,

Yes, I'm working on it. The latest code from Mikael works well with 
Ethernet.
I built it with SCI yesterday and will test today.
I'll get back to you after verification of MySQL Cluster with SCI.

Thank you very much.
/Junzo
----- Original Message ----- 
From: "Hugo Kohmann" <hugo at dolphinics.no>
To: "Junzo Tamada" <tamada at acornnetworks.co.jp>
Cc: "Hugo Kohmann" <hugo at dolphinics.no>; <cluster at lists.mysql.com>
Sent: Thursday, October 28, 2004 4:02 AM
Subject: Re: MySQL Cluster with SCI interconnect


>
>Dear Junzo,
>
>MySQL cluster has native support for SCI our you can use the SCI Socket 
>software. You will find more information about this and some
>benchmark numbers at 
>http://dev.mysql.com/doc/mysql/en/MySQL_Cluster_Interconnects.html
>
>With SCI Socket, you can configure MySQL cluster for regular networking 
>and just use SCI Socket between the nodes that requires low latency/high 
>bandiwidth connections. SCI Socket has typically 10x lower latency than 
>Gigagit Ethernet and should support any version of MySQL Cluster that runs 
>properly over Ethernet.
>
>You will find the SCI Socket software and more information at 
>http://www.dolphinics.com/products/software/sci_sockets.html
>
>Best regards
>
>Hugo
>
>
>On Tue, 26 Oct 2004, Junzo Tamada wrote:
>
>>Date: Tue, 26 Oct 2004 23:46:49 +0900
>>From: Junzo Tamada <tamada at acornnetworks.co.jp>
>>To: cluster at lists.mysql.com
>>Subject: MySQL Cluster with SCI interconnect
>>
>>Hello
>>
>>Very recently I found sections in the chapter of Cluster, which is 
>>decsribing utilization of high-speed interconnection called SCI.
>>I would like to test it.
>>Currently I have totaly 5 servers and 4 out of 5 are equiped with SCI 
>>network card.
>>Please anyone provide me any suggestions regarding the following cluster 
>>configuration.
>>All server are connected together with Ethernet.
>>Node a1 through a4 are with SCI.
>>I would like to assign mgmd and api(mysqld) to front-end and ndbd for 
>>a[1-4].
>>Is this feasible and reasonable configuration ?
>>
>>I am using MySQL 4.1.7-gamma as of Oct. 26th, 2004.
>>
>>Thank you in advance.
>>/Junzo
>>
>>-- 
>>MySQL Cluster Mailing List
>>For list archives: http://lists.mysql.com/cluster
>>To unsubscribe: 
>>http://lists.mysql.com/cluster?unsub=hugo@dolphinics.com
>>
>
>
>=========================================================================================
>Hugo Kohmann                           |
>Dolphin Interconnect Solutions AS      | E-mail:
>P.O. Box 150 Oppsal                    | hugo at dolphinics.com
>N-0619 Oslo, Norway                    | Web:
>Tel:+47 23 16 71 73                    | http://www.dolphinics.com
>Fax:+47 23 16 71 80                    | Visiting Address: Olaf Helsets 
>vei 6   |
>



-- 
MySQL Cluster Mailing List
For list archives: http://lists.mysql.com/cluster
To unsubscribe:    http://lists.mysql.com/cluster?unsub=eugen@leitl.org

----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://www.beowulf.org/pipermail/beowulf/attachments/20041028/f6fca634/attachment.sig>