NZX Moves To Oracle On Linux

sn00ker writes "In this story in The New Zealand Herald, we learn that the NZX stock exchange has moved their database systems to Oracle running on RedHat Linux, running on commodity Intel-based hardware. What's really impressive are the performance numbers they're claiming. Quoth the article, "One key query - searching the data on historical trades to identify maximum trade values - has been cut from 36 seconds to 0.03 seconds." An improvement of over 1000 times is spectacular in anybody's books, and is one hell of a boost for the proponents of Linux at the back-end of the financial world."

Comments

Linux?

[Dr Rick]

"An improvement of over 1000 times is spectacular in anybody's books, and is one hell of a boost for the proponents of Linux at the back-end of the financial world."

Oh come on! They consolidated 21 databases and moved to Oracle. That's why it is 1000 times faster. The move to Linux is a footnote as far as the performance issue is concerned -- as stated in the article, the move to Linux was for cost. I'm sure Solaris or god help me, Windows Server 2003 would have given similar performance results. Now if they had moved to MySQL...

Re:Linux?

[killjoe]

The article is a bit short on details. There is no mention of the previous hardware or software for example. Did they move from 21 SQL servers running on windows to 1 oracle running on linux? Maybe they just got rid of a bunch of aging unix machines and build a cluster of intel servers running linux.

Either way though the fact that a major exchange is running linux is big news. Their database is their life and they are trusting it to linux. That says a lot.

Re:Linux?

[Anonymous Coward]

So perhaps it's time we coin a new term here. anti-FUD? I love Linux, but it's been obvious to me from the get go over four years ago now that pro-Linux articles and pro-Linux users sometimes stretch the truth or slightly warp reality to make it seam better than it is. There is no doubt...Linux is better than MS anything in my mind...but come on let's be realists and scientific here...not religious. Linux needs to win on factual merits...not more marketing gimicks that look a lot like those latest MS Office comercials that give it's users joy gasms.

Re:Linux?

[0racle]

You could argue that Linux was a factor if it was a move from 2003 to Linux, though to back that up you'd have to show that Linux handles assloads of threads better then 2003 does. Solaris on the other hand does handle assloads of threads and then some just fine, its one of Solaris's selling points and as far as I know it does handle them better then Linux currently does. So they may have lost performance by choosing Linux over Solaris, but that was more then made up for in the consolidation.

Re:Linux?

[rodgerd]

While you're correct that the article conflates a bunch of work that happened, it's entirely possible that Linux was an enabler for this - for example, the cost per (unit of performance) of Solaris is still (IMO) fucking ridiculous at the lower and middle ranges compared to the cost of the same on Lintel.

And while Windows Server 2k3 can run on the same cheap hardware, can you get a production quality release for AMD64 if you need gobs of RAM? What about the cost of multi-CPU licenses, and any client licenses needed? All money that buys you more power in the Linux world.

It may well have been the case "well, with Linux, we can buy enough CPU, I/O, and DBA tuning time to make this thing sing. With Windows we blow money of software licenses. With Solaris we blow it on licenses and proprietary hardware."

Re:Linux?

[supersnail]

An inprovement of 1000 times might not be that sectacular, it depends what the system was replacing.

Most likely it would be some unix hardware circa 1997. (say 4 x 200 Mhz Solaris, 512MB, with SCSII II disks, or, perhaps even a VAX complete with snails pace IO would be typical for that period in that environment). So a 2 x 3 gHz, 2 GB, with fibre channel ought to be faster. Plus it looks like they rewrote the whole system to take advantage of Oracle 10 features.

What is perhaps more interesting for slashdot readers is that for most people working at the trading end of finacial services this is very much a non news story. The last two sites I worked at had implemented or were implementing Linux cluster server based systems, and, these were both for volume performance critical systems.

1000x performance increases...

[david_reese]

are NOT gotten by changing your OS.

I'm no windows sympathizer, but in the world of enterprise software, only optimizations at the database layer (or reworking badly written networking layer) can yield those kind of results.

Sounds like they data warehoused and redesigned the schema/indexes to better match usage.

Re:1000 times faster?

[mcrbids]

I'm inclined to think that having a request suddenly run 1000 times faster might be due to something a DBA has done, rather than a change of OS.

Yeah. My call would be that they were operating an RAM-starved server. I've seen similar numbers doing basic PC upgrades!

I remember on case (this was a few years ago) where somebody with a customer information database of about 400,000 records came to me because generating a list from a query would often take several minutes.

They were using a Pentium-90 with 32 MB of RAM. I set them up with a (then) top-of-the-line PIII 600 with 256 MB RAM. Query time dropped to 1 second.

No matter what O/S you run, you're going to get JACK for performance if your running your app in swap.

Impressive, but..

[nayigeta]

The improvement is impressive - but I would credit the overall architecture, rather than some single specific factors - like Oracle10g+Redhat or DBA or systems consolidation.

I mean, every part of the architecture has its role.

Some other contributing factors not mentioned, I suspect, would includes - focused performance requirements, specific purpose optimised query framework.

Can someone point to some public material on the architecture? It would be a interesting read.

Re:A pretty telling statement in there...

[Soko]

I doubt they're your garden variety "OMG BillG iz teh debil" Loonix fanbois, friend.

They are a serious enterprise, and there must be a reason something as provocative as " not just because we hated Microsoft" would come out in an interview.

IOW - It's likley that Microsoft's products and/or policies have left a very, very bad impression with these people, and they're glad that they have a compeditor with which to smack Microsoft in the head with.

Soko

Not really

[CaptainZapp]

An improvement of over 1000 times is spectacular in anybody's books, and is one hell of a boost for the proponents of Linux at the back-end of the financial world."

Unless specifics about the query and the physical database model are comparable in both systems this isn't really impressive.

Comparable - not equal - since each database engines optimizer has it's individual quirks and strength.

Assuming that you have large joins on huge tables a couple of good indexes, which make the optimizer happy can reduce execution time from hours to seconds.

Table scans are expensive in database speak.

Re:Some kind of cluster

[killjoe]

I know that. I read the article. The question is what they were using before that. If you are going to say something is 1000 times faster the least you could do is explain both your old setup and your new one.