Betfair's Flywheel outguns LSE Tradelect

While doing a bit of research into high performance computing for our capital markets conference Finexpo, I stumbled across this article in Business Week on the Betfair gambling exchange. It seems that London-based Betfair has developed a new trading engine dubbed Flywheel, which is capable of handling 100,000 transactions per second and five million transactions per day.

As if that's not impressive enough, the Betfair R&D team expects one million trades per second to be possible through Flywheel, which it estimates is the equivalent of the entire combined annual global equity trading volume being processed in a matter of hours. And all of this on a couple of servers costing £25,000!

Now compare this with the London Stock Exchange’s much-vaunted Tradelect platform, introduced with great fanfare in June of this year. Tradelect was hailed by the LSE for doubling capacity five-fold, from 600 messages per second to 3000 per second under the new engine. An upgrade to the system in late-October pushed the capacity up to 4,200 trades per second. This followed events in August, when trading volumes on SETS breached the one million mark for the first time in a single day, and Tradelect was called on to process some 1.27 million trades.

Phew, better run-off another self-congratulatory press release.

Tradelect was designed to deal with the impact of automated trading strategies leading to higher volumes of smaller-value transactions. Flywheel similarly was developed to meet an estimated exponential increase in online gambling and customer trends for high volume, small value bets.

But while the London Stock Exchange spent £40 million developing Tradelect, the R&D team developing Flywheel was given a budget of £1 million.

Now, I realise that's a low blow. We're not exactly comparing like-with-like here. But I'm intrigued all the same. We were thinking about inviting the LSE to come along to Finexpo and tell us all about Tradelect. Maybe we should ask the Betfair mob instead.

So, if anyone at Betfair Labs is reading this, maybe you'd like to get in touch.