Industry


Ads by TechWords

See your link here


IT Blogwatch's picture
IT Blogwatch

A Daily Digest of IT Blogs from Richi Jennings

EXTRA: London Stock Exchange: blame Microsoft?

Welcome to a special IT Blogwatch EXTRA: watching bloggers' reactions to yesterday's all-day outage at the London Stock Exchange. Not to mention Reese Witherspoon, as you've never seen her before...

Mike Simons reports:

Microsoft adThe London Stock Exchange ... suspended trading as dealers reacted to the dramatic economic news that the U.S. government had taken over control of mortgage groups Freddie Mac and Fannie Mae, in the biggest financial bailout in world history.
...
The breakdown occurred just days after the LSE announced plans to improve services as it competes against a range of new entrants in the market. Among the LSE's initiatives is a move to slash the time it takes to complete a trade from six milliseconds to three
...
Trading was halted at 8:45 a.m. London time. According to Reuters, it was not restored until shortly before close of the trading day, which is 4 p.m. The LSE said the system had been hit by a "connectivity issue" and insisted that the problem did not lie with its flagship TradElect trading platform.more


But Steven J. Vaughan-Nichols knows people who know different:

So what happened? Officially, the LSE ... gave the vague explanation, that "It was software-related, a coincidence, due to two processes we couldn't have foreseen," and not caused by high-volume.
...
I have friends in London and... Well, let me just make the following points about TradElec. First, TradElec runs on ... Windows Server 2003 ... [and] a custom set of C# and .NET programs, which was created by Microsoft and Accenture ... on Microsoft SQL Server 2000. The goal was to maintain sub-ten millisecond response times.
...
The programmers and serious database administrators in the audience can already see where this is going. Sorry, Microsoft, .NET Framework is simply incapable of performing this kind of work, and SQL Server 2000, or any version of SQL Server really, can't possibly handle the world's number three stock exchange's transaction load on a consistent basis.

I'd been hearing from friends who trade on the LSE for ages about how slow the system could get. Now, I know why.more


As confirmed by AreBelongToUs:

The LSE seems to have major problems coping with the order flow traffic whenever there is a major piece of financial news. There have been at least 3 other ****-outs this year where I've spent most of the day looking at their incident status web page.

With PLUS Markets, Chi-X, and the soon to be rolled out Project Turquoise all competing for the LSE's business, I wonder how much longer it will be the preferred execution venue for trading UK equities.more


And Tim says the problem wasn't just in London:

Johannesburg went down, as did the ICE in the USA. They, and the LSE, use the SETS platform.

This looks really bad for [the LSE]. Their rivals ... are more reliable and cheaper. It's only inertia that keeps volumes so high on the LSE. A few more incidents like this and they'll become a backwater, used for trade reporting and listing only.more


Daniel Kottmair launches Linux schadenfreude missiles:

The incident could prove to be particularly embarrassing for Microsoft who at the end of 2006 launched a huge advertising campaign stating that the London Stock Exchange had chosen Windows over Linux because of reliability issues. An opinion obviously not shared by the New York Stock Exchange who has been using Linux and AIX for over a year without any outage at all.

This is not the first time that the new NET, MS-SQL and Windows server based Infolect system has faltered. In September 2007, the London Stock Exchange was hit by connectivity problems when three Infolect gateways couldn't stand the strain.more


And Mike Shade recalls:
Microsoft ad.

Sometime last year, Microsoft released an ad campaign with fake newspaper headlines from a periodical called the “The Highly Reliable Times”. With today’s news about the London Stock Exchange having a 7 hour outage I had to think back and laugh.more


Last word goes to R2.0, who quips:

Looks like someone needs to brush up on their buzzwords, specifically "mission critical" and "services no longer required".more


And finally...

RSS feed icon Like this stuff? Subscribe to the RSS feed.

Richi Jennings is an independent analyst/adviser/consultant, specializing in blogging, email, and spam. A 22 year, cross-functional IT veteran, he is also an analyst at Ferris Research. You can follow him on Twitter, pretend to be Richi's friend on Facebook, or just use boring old email: blogwatch@richi.co.uk.

Previously in IT Blogwatch:

What People Are Saying

Will we ever know the truth?

We already know the truth, it's proving it that will be difficult.

FIXFAST will be an

FIXFAST will be an Armageddon and went head on against all the kbs at Deutsche, CME and LSE. But FIX is here to stay, soon everyone will be using it, even the best exchange in the world.

There was a Tandem non-stop kernel involved ages ago at LSE and it was one of the most reliable exchanges that I ever seen.

Batching is a usually way to reduce latency, but it is not a realistic measure. Those FIXFAST engineers claim similar stupid numbers without understanding that matching is a science in itself and that encoding bits is so 1970.

Chi-X is the Archa copy, and ironically ICE is aiming to be very much like Chi-X.

Whoever the people are writing these comments are, they have little clue.

Accenture did do a mess, but LSE has 100 odd systems and works to integrate them with latest MS tech etc. But Accenture designed Eurex too. And it is a baroque exchange in comparison to LSE.

Chi-X is Java too. Worse junk that .NET.

ROW ROW FIGHT THE POWAH

ROW ROW FIGHT THE POWAH

stu*ed

Why st.people go on windlous? 4rom LINUX???
---
Realy stupied? Sorry, i`m not write by eng.

Will we ever know the truth ?

In such critical mission, 7 hours offline is HUGE.
It shouldn't have happened or should have been quickly fixed.
Someone's reputation will be damaged.

Microsoft can't and won't be blamed.
TradElect can't and won't be blamed (in use in London, Johannesburg, Milan, at least, others coming).

It will be someone else fault, like :
* A network hardware
* A not-Microsoft system
* A not-Microsoft-not-TradElect software
* A worker

I doubt we'll know the truth.

It is does not important, where fault happened

It is does not important, where fault happened. I nmission critical system ALL of these MUST be dublicated and reserved. There is no allowed to downtime! Maximum time - 1 hour (or less) - for switching to reserved server(s) or network hardware or even alternate power supply.

7 hours - it is impossible! It looks like ALL (or a half at least) system was down! And recovered from backups.

PS. Do you remember about some updates from Microsoft, when up to 30% servers was down after updates applied?