Comments on: Finding your MySQL High-Availability solution – The questions http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/ Percona's MySQL & InnoDB performance and scalability blog Sat, 11 Feb 2012 16:45:54 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: Kmilo http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-860211 Kmilo Mon, 19 Dec 2011 20:29:47 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-860211 I cant' save this url in delicious :( http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%E2%80%93-the-questions/ Looks like and strange character on: --- I cant’ save this url in delicious :(

http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%E2%80%93-the-questions/

Looks like and strange character on: —

]]> By: samana zaidi http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-794830 samana zaidi Fri, 21 Jan 2011 16:20:14 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-794830 average formula is what. average formula is what.

]]> By: Mischelle http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-773919 Mischelle Sat, 11 Sep 2010 11:17:07 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-773919 I hope you guys don't mind. How do you deal with many branches and all accessing the same database and then connection between branches (e.g. lease line and DSL) suddenly becomes offline? Real-time transactions are necessary with our operations. Do you think that any of the mentioned HA solutions would fit our current setup? or is there any other way to achieve the 'Real-time' requirement? Thanks guys! I hope you guys don’t mind. How do you deal with many branches and all accessing the same database and then connection between branches (e.g. lease line and DSL) suddenly becomes offline? Real-time transactions are necessary with our operations. Do you think that any of the mentioned HA solutions would fit our current setup? or is there any other way to achieve the ‘Real-time’ requirement?

Thanks guys!

]]> By: LinuxJedi http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-706968 LinuxJedi Fri, 08 Jan 2010 16:06:01 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-706968 Hi Cédric, A cluster running on good hardware with a good configuration, both tuned to the workload. We have customers who get very good performance and no downtime (ie. no total cluster failure). There are a lot of disk writes with LCPs and redo logs but typically LCPs are quite slow writes (default 10MB/sec). With cluster being real-time the redo data (GCP) does need to be put to disk quickly, so fast disks are important for this aspect, especially on a database with a lot of writes. I would rather a lot of disk IO than the possibility of data loss. Hi Cédric,

A cluster running on good hardware with a good configuration, both tuned to the workload. We have customers who get very good performance and no downtime (ie. no total cluster failure).

There are a lot of disk writes with LCPs and redo logs but typically LCPs are quite slow writes (default 10MB/sec). With cluster being real-time the redo data (GCP) does need to be put to disk quickly, so fast disks are important for this aspect, especially on a database with a lot of writes. I would rather a lot of disk IO than the possibility of data loss.

]]> By: Cédric http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-706897 Cédric Fri, 08 Jan 2010 10:08:30 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-706897 Hi LinuxJedi, i agree but what's a "well configured cluster" ? In this case I think a "well configured cluster" is a cluster with poor performances ! "Poor performances" means a lot of disk write and so, you lose a part of interest of the cluster... Hi LinuxJedi, i agree but what’s a “well configured cluster” ?
In this case I think a “well configured cluster” is a cluster with poor performances !
“Poor performances” means a lot of disk write and so, you lose a part of interest of the cluster…

]]> By: LinuxJedi http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-706711 LinuxJedi Thu, 07 Jan 2010 19:55:49 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-706711 Hi Cédric, Yes, that is possible if all nodes fail simultaneously, which is not something that is likely to happen with a well configured cluster. Hi Cédric,

Yes, that is possible if all nodes fail simultaneously, which is not something that is likely to happen with a well configured cluster.

]]> By: Cédric http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-706690 Cédric Thu, 07 Jan 2010 17:56:54 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-706690 Hi, I just want to indicate that you can lose data with MySQL Cluster (NDB Cluster) ! See description of the TimeBetweenGlobalCheckpoints parameter... Hi, I just want to indicate that you can lose data with MySQL Cluster (NDB Cluster) !
See description of the TimeBetweenGlobalCheckpoints parameter…

]]> By: Baron Schwartz http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-689326 Baron Schwartz Thu, 03 Dec 2009 13:40:55 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-689326 In response to feedback through other channels, I've added a note to the end of the blog post to add some context to the time estimates. These time estimates are for HA projects, not for simple administration tasks. Anyone can follow a WikiHow tutorial and install the tools and technologies mentioned in a relatively short time. Not many people have Yves's level of expertise in architecting a complete HA solution, which is a whole different ballgame. In response to feedback through other channels, I’ve added a note to the end of the blog post to add some context to the time estimates. These time estimates are for HA projects, not for simple administration tasks. Anyone can follow a WikiHow tutorial and install the tools and technologies mentioned in a relatively short time. Not many people have Yves’s level of expertise in architecting a complete HA solution, which is a whole different ballgame.

]]> By: Ewen http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-669629 Ewen Tue, 27 Oct 2009 20:00:25 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-669629 Dba Zed, NDB cluster does not really make sense for Data warehouse applications. There was a discussion on what is good for cluster on the mailing list, check out the thread that starts here: http://lists.mysql.com/cluster/6960. In terms of performance versus Oracle RAC I think they are very different solutions and really answer different problems. Tom Hanlon's post in the thread sums up rather well what scales well with NDB. Dba Zed,

NDB cluster does not really make sense for Data warehouse applications. There was a discussion on what is good for cluster on the mailing list, check out the thread that starts here: http://lists.mysql.com/cluster/6960. In terms of performance versus Oracle RAC I think they are very different solutions and really answer different problems. Tom Hanlon’s post in the thread sums up rather well what scales well with NDB.

]]> By: Dba Zed http://www.mysqlperformanceblog.com/2009/10/16/finding-your-mysql-high-availability-solution-%e2%80%93-the-questions/comment-page-1/#comment-667034 Dba Zed Tue, 20 Oct 2009 23:02:54 +0000 http://www.mysqlperformanceblog.com/?p=1334#comment-667034 Hello Ewen, Is it possible to get the NDB Cluster for Datawarehouse Appliances. In term of performance, is this solution can perform as well as Oracle RAC? What are the limits? Thank you. Best Regards. Hello Ewen,
Is it possible to get the NDB Cluster for Datawarehouse Appliances. In term of performance, is this solution can perform as well as Oracle RAC? What are the limits?
Thank you. Best Regards.

]]>