I've been playing around with endpoints in SQL Server 2005 looking at how they can be used to enhance security for SQL Servers servicing web servers based in the DMZ. Figure 1 shows a traditional architecture for just such a web server / database combination. The web server exists in the DMZ, outside of the trusted network. The SQL Server exists inside the trusted network and the appropriate ports are opened up in the firewall to allow the web server to connect back to the SQL Server. Best practices would say only open the necessary TCP port SQL Server listens on and no more. Since the port can be specified in the connection string or through the use of an alias, opening up UDP port 1434 isn't necessary for named instances.

Figure 1: Typical Architecture for web server in DMZ

…

MySQLDevelopment.com is no more, All requests to the site will now be served from http://forge.mysql.com/wiki/Category:MySQLDevelopment.

Thanks for all those who visited the site over the last year.

Sun has now published SpecJAppServer2004 benchmark results with MySQL. The results are pretty good as we can see and it is also good to know some room for tuning remains so we can hope getting even better results this time. Jenny Chen published good MySQL Tuning writeup so I would not repeat it. It was great colaboration between Sun and MySQL teams which allowed us to deliver these results.

Jeremy Cole recently posted very nice post about MySQL Replication Performance. There are however few points I should comment on.

• Jeremy Speaks about MyISAM and Innodb in the same terms, in fact they are quite different for Replication. As you probably know MyISAM uses buffered IO so if master crashes you're out of sync anyway - whatever sync_binlog option was set to. Well it is still a bit better with sync_binlog enabled as slave could have consistent data after it catches up but master may have some of its tables out of sync with binary log and it is not easy to find which.
• I'm surprised to read enabling binary log drops update performance. I guess Jeremy speaks about enabling binary log and setting sync_binlog option which is good for safe replication. But it is setting which is off …

Here are some steps involved. Using the current MySQL defacto engine InnoDB. Of course, Falcon, PBXT and others will enable alternative engines to be used.

Convert Table Storage Engine Types

$ mysql -u[user] -p[password] [database] -e "SHOW TABLES" | grep -v "Tables_in" | sed -e "s/^/ALTER TABLE /" | sed -e "s/$/ ENGINE=InnoDB;/" > upgrade.sql
$ mysql -u[user] -p[password] [database] < upgrade.sql

NOTE: This may not work for all tables, for example those with FULLTEXT indexes will fail.

For the introduction of Referential Integrity we need to ensure the following.

• Each Foreign Key column should have an index. Prior to 4.1 I think this was a requirement, however it’s a good general practice regardless for SQL performance.
• The datatype must match between Primary Key and Foreign Keys. The most obvious oversight is normally UNSIGNED, however you also for example have INT …

Following my implementation of UltimateLAMP, read heaps more at this thread, I undertook to provide customisations of a MediaWiki Installation. Here is the first lesson that you can undertake if you wish to beautify the default MediaWiki Installation.

For the purposes of this demonstration, I am going to help out Jay & Colin and propose a restyle the MySQL forge to fit in with the default Home Page. Hopefully you will see it there soon!

For the full lesson Read More Here

Lesson 1 - Updating the default Monobook.css

There are several different ways to make style changes, the simpliest is to customise the system …

There are a few posts i have missed. Lots have happened in the mean time...






Have been to home and to my collegues marriage on 14th of may to bhopal. Had a wonderful day on 15th of may at the famous lake of Bhopal. Had lunch at a four-star, did boating, hung around in odd clothes... In all had great fun. Below are some pics from the same...

The clothes were bought specially for the occassion. We had people enquiring "Where have we come from?". Hope we would have created some waves...

Secondly …

Pro MySQL 5 from Jay Pipes and Mike Kruckenberg is definitely one of the best advanced MySQL books around.

dbazine.com has published a Podcast Episode with an interview with Jay and Mike where they talk about themselves and - of course - MySQL:

http://www.dbazine.com/podcasts/podcast-kruckenberg

Sun just post SpecjAppserver2004 with a very excellent result of 712.87 SPECjAppserver 2004 JOPS@Standard using the latest SJSAS 9.0 Platform Edition on Sun Fire X4100 Cluster with MySQL 5.0.20 yesterday(Also check out Tom Daly's blog, Scott Oak's blog and Robert Lee's blog for more information)

SPECjAppserver is industry standard benchmark for performance testing J2EE application server. As part of important component of SPECjAppserver testing, database's performance and scalability also greatly affect the overall performance result in …

My dual opteron which has gobs (more then 2gigs) of memory will not let a single process spawn more then 12,000 threads at any time.

Well, somewhere between 12,000 and 13,000 threads things go haywire. I had a program I put together yesterday break because it was let loose with thread creation. So last night I wrote a little loop around an application just to see where consistent thread creation could occur (each thread did nothing more then maybe sleep for a second (it had a one in seven chance of the sleep event occurring)).

I am sure there is a system limit I am ignoring, but I wanted to see what the practical numbers where in this case.

Completely random and probably worthless information :)

Its crazy to generate this many threads in the first place. I saw Nusphere once do this with Gemni to prove a point about concurrency, but it didn't prove much of a point for me since I thought …