Computer Sciences Corporation (CSC), defence contractor BAE
Systems, Australian Mutual Providence and other financial
institutions have all been hit by the failure of a server in a
Cable & Wireless data centre.
CSC is one of the largest outsourcing companies in the world and it
is believed that the affected server was responsible for a wide
range of tasks, including Internet traffic, extranet applications
and database access for the CSC and some of its customers.
Cable & Wireless (C&W) told CW360 that the server's
downtime lasted 36 hours. Problems began late on 19 September,
after what C&W described as a "serious hardware failure".
A new server in C&W's Swindon data centre was brought online in
the morning of 21 September.
There is, however, some confusion about the scale of the problem
and the location of the affected server and data centre.
One customer's emergency response team was reportedly told the
problem lay in hardware in one of C&W's data centres in the
US.
C&W UK could not confirm this: "The only affected server I'm
aware of is a small server in the UK which is not within in our
main facility in Swindon," said a spokesman.
C&W would not say what the fault was, if any other customers
were affected, or why it took 36 hours to rectify.
A customer who asked not to be named said, "[we] will be
considering our SLA agreement to see if any financial compensation
can be demanded from C&W."
C&W has 17 primary hosting centres around the world, with over
500,000 sq ft of hosting space. It also claims that its largest US
hosting centre is able to process 700 million Web visitors in a
single day.
C&W would not say which vendor supplied the affected hardware.
However, C&W has a £350m, five-year deal with Compaq for joint
hosting and application services for large corporations.