• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

MS 2k Server w/ MS SQL vs. FreeBSD with mySQL/Postgre

A client has asked me to look into porting their MS Server/MS SQL database to a NIX solution.

I'm not so sure that the performance gain will justify the riskiness of the transition.

Any takers on this subject?

Thanks for any input!
 
Like in the other thread, you need to tell a lot more about the MS SQL setup before it can even be decided if it's possible at all.
 
well, assuming they already have the licenses of win2k and mssql, wouldn't it be pissing money down a tube to pay a developer to port it?

unless there's some new feature/functionality you're looking for. of course, mysql ain't gonna have it, so that leaves postgres or sapdb.

sounds pointless. unless there's something else, like you're trying to standardize servers a bit, or ... ?

bart
 
A portion of this is my own curiousity. I will probably tell them to port their static pages to a smaller server running freebsd, and upgrade the main cow, sticking with the MS products. Porting could be a nightmare, and unless the nix road proves a 150%-250% performance increase, I don't think i will recommend the switch.
 
Mysql is design with web speed in mind therefor it doesn't have much features, but can be as much as 2.5X faster than other db at delivering certain web funtions .

Postgre & Interbase/Firebird are much better with more features (better primary & foreign key features) than Mysql, and is compleately free if you want something more than a web server.

Sysbase is quite good & is an inexpensive solution that you might want to look at.

And, it shouldn't be much of a problem if you could convert the data into XML format, because all above database support XML.

In the long run MS-sql solution can be expensive because of the licensing scheme, upgrade & the additional 16-18% user tax every year can hurt.

You have to wean yourself soon or later from the over price MS pacifier.
 
You have to wean yourself soon or latter from the over price MS pacifier.

Not if you get MS software for free!


Anyway, I've found MSSQL2k to be very reliable and rather robust. However, porting anything to or from it is a nightmare. I tried to move a PHPNuke on a mySQL database to PHPNuke on MSSQL ... talk about 3 days of my life being flushed down the john.

If it works, leave it -- find ways to optimize IIS, because that will be *WAY* more cost-efficient and won't take months/years to do.

cheers,
randal
😀
 
the mysql-speed argument falls down fast when you get into any real business application (an inherently multiple related tables), and especially anything insert-heavy (as in more than 8% insert 92% select).

bart
 
Not if you get MS software for free!

If you want to run a business legally that doesn't apply to you.

the mysql-speed argument falls down fast when you get into any real business application (an inherently multiple related tables), and especially anything insert-heavy (as in more than 8% insert 92% select).

That's why the speed argument is usually immediately followed or preceded by a statement recommending it for most web type services, because >90% of their use is select based.

MySQL has name recognition, which is why it's usually mentioned instead of PostgreSQL (which has a name noone's sure how to even pronounce) but that doesn't mean it's a bad choice. You just need to figure out which one fits your needs, hell switching from MS SQL on Win2K to Oracle on Tru64 might give a bigger performance jump than switching to MySQL on FreeBSD but I doubt many would say it's worth the cost.

And on a side note, I would recommend Linux over FreeBSD just for support purposes. Again it's a name recognition thing, but big companies like Dell and Compaq officially support Linux on a lot of hardware configurations and they don't even recognize that FreeBSD exists.
 
The guy next to me at work pronounces it "post-gray-sql".
Ok I suck at explaining sounds, excuuuse me 😛
 
Part of the database is for a small online auction site they are building. So there will definately be a lot of transactions, new records, updated records, deleting records, etc. It sounds like Post-GreS-ql will be a better fit for the database, based on what I have read.
 
have you done the uptime/availability math? like figured out the cost/min of downtime?

Last I checked, PostgreSQL doesnt have any replication. This means keeping a hot-standby or loadbalancing selects wont be possible. Which means you'll be relying on how quality your hardware and datacenter is, and how fast you can get replacement parts for your cost-of-downtime equation.

I'm not sure if MS SQL can do regular replication either. I've only ever seen examples of clustering with some kind of shared-storage equipment.

This is one thing MySQL does very well. Using their binary log replication (also good walkthrough/tutorial here), and heartbeat, you can failover from your primary to a hot-backup server in about 4 seconds, all automatically.

[edit] Plus if you're really tricky, you can seperate your selects from your insterts at the application level, and use LVS to load balance your selects. This scales extremely well, as you can practically slap any old hardware in as an extra "select" db server, and heavily tweak your "insert" server for that purpose. Anyway, this is probably getting beyond what you're interested, and the biggest caveat of all for mysql is that if you need this kind of database availability, you usualy also need features like transaction support and others that mysql doesnt support (dont give me that InnoDB or MySQL4 cruft).[/edit]

bart
 
I can only find this
New /contrib/rserv replication toolkit (Vadim)
in that link lowtech.

I looked around and couldn't find any documentation or articles in actually using it. Plus I found this website:
http://www.erserver.com/

Which from its very existance leads me to believe replication in PostgreSQL is no simple built in feature. They also say
RServ v0.1 was released for Beta on December 24, 2000 as the Perl based open-source version of our commercial enterprise replication server, eRServer. To download a copy of the RServ Beta, please go to the PostgreSQL, Inc. site and click on the 'Downloads' option from the right menu box.
And if you follow their instructions you get led to download that same version (0.1) for older versions (7.0 and 7.1) of postgres.

bart
 
Originally posted by: Buddha Bart
I can only find this
New /contrib/rserv replication toolkit (Vadim)
in that link lowtech.

I looked around and couldn't find any documentation or articles in actually using it. Plus I found this website:
http://www.erserver.com/

Which from its very existance leads me to believe replication in PostgreSQL is no simple built in feature. They also say
RServ v0.1 was released for Beta on December 24, 2000 as the Perl based open-source version of our commercial enterprise replication server, eRServer. To download a copy of the RServ Beta, please go to the PostgreSQL, Inc. site and click on the 'Downloads' option from the right menu box.
And if you follow their instructions you get led to download that same version (0.1) for older versions (7.0 and 7.1) of postgres.

bart
Try here: PostgreSQL Replicator, Rserv, and Usogres - You have to install perl & compile for rserv, but pgReplication is another program that is precompile in tar ball format & also available on RedHat site as RPM (RedHat Database)

Usogres replication test report

Rserv source code

pgReplicator RedHat_postgresql_contrib.RPM

pgReplicator tar ball
 
Back
Top