Bloomberg global trading outage a risk to users' confidence, says trading software CEO

A repeat of Bloomberg's outage – when terminals failed, leaving traders unable to conduct business – could leave users looking for plan B

A repeat of Bloomberg's system outage last week – when the company’s terminals failed, leaving traders unable to conduct business – could push customers to look for alternatives, said Hiranda Misra, CEO at trading software developer Algo Technologies.

He said an investigation, followed up by reports to customers, is vital to Bloomberg retaining its users' confidence.

Currently, Bloomberg has the dominant system in the trading sector. Trading companies use Bloomberg's terminals to access information, messaging and trading.

But at about 11am on 17 April 2015, Bloomberg's terminals went down. The company has since reported the problems were “a combination of hardware and software failures in the network, which caused an excessive volume of network traffic”.

Bloomberg said: “This led to customer disconnections as a result of the machines being overwhelmed. We discovered the root cause quickly, isolated the faulty hardware and restarted the software. We are reviewing our multiple redundant systems, which failed to prevent this disruption."

Read more about trading-sector IT

Bloomberg gives no account of outage

Yann L’Huillier, CIO at inter-dealer broker Tradition – itself a Bloomberg customer – said that, even when Bloomberg got the system back up and running, it remained slow.

L’Huillier said Bloomberg had given no explanation about what had happened and said he wanted to know how the company planned to prevent it happening again. “I am surprised I have not been given an explanation, because we are a big customer.”

Misra said Bloomberg's outage was very rare. He said outages do happen, but the supplier's response is critical. “I can’t remember the last time Bloomberg had an outage. Like any supplier, Bloomberg must give the market confidence by doing a 'post mortem' and informing them what went wrong – and how you will avoid the same thing happening in the future.”

Risk of single point of failure

One investment banking IT professional agreed it was unusual for Bloomberg's systems to go down. “The impact shows how reliant the industry globally has become on this service. I think the more serious issue is how most – if not all – the trading firms have become reliant on one supplier, which is a huge risk.”

“All systems will go wrong at some point, it's just a case of how badly, how often and how quickly they get up again. 

"I would say Bloomberg itself could be a global single point of failure – although they will of course explain they don't have any single points of failure with their systems. Maybe there are some vendors behind financial services that should be regulated.”

Traders could cast around for Plan B

When it comes to reducing reliance on single supplier, Misra said a repeat could see users looking at alternatives. “Maybe some firms will think more about contingency plans or backups, but I doubt this will change much. If it happens again though, I expect there will be less tolerance and trading firms will be forced to have a Plan B.”

In the trading sector, where systems complete trades in microseconds, even seemingly insignificant glitches can have huge ramifications.

“When you think that trading is all about speed, this is not good news," said Chris Skinner, chairman at the Financial Services Club.

“Bloomberg is the predominant trading terminal used in the City, so it would have affected most firms.”

Read more on IT for financial services