home > news > datacenter outage

Datacenter Outage

John A. Herbert
John A. Herbert

You will have heard the news, a total power outage recently took down the very popular social-networking site Myspace.com last weekend (23 July 2006).

Click here to read news report

datacenter outage, power cut, unplanned outage

So who's to blame when a web site crashes? The site owner? the host? The blame game doesn't seem to have started yet, as Myspace the latest mega-Web site to recovers after an outage.

When was their last time the E&M infrastructure was audited? obviously too long ago

MySpace is one of the largest sites on the net having approximately 80 million users, and estimated to garner 23 billion page views per month suffered a power outage.

The outage at the company's US based datacenter took down the entire MySpace website down - not once but TWICE. Starting on the evening of 23 July 2006, initially for six hours, and then 12 more hours after that. Reports indicate that both the main power, and the standby power systems failed.

datacentre outage, power supply, datacentre auditing

Of course the customers, are furious, venting their anger online through the blogging network. And although the web site is now back online, some glitches remain and the investigation into the failure is unlikely to be public domain.

"It was most probably the classic a single point of failure scenario - for a system with millions of users that is a concern. Myspace is only one of the many datacenter's around the world that suffer unplanned outages, an independent infrastructure audit would have identified the problems long before the lights went out" said Mr John Herbert, an datacenter infrastructure specialist from Kelcroft E&M Limited.

Myspace was not alone

In an unrelated incident, also on 23 July 2006 (Sunday) another datacenter also suffered an unplanned outage, this time in North London - UK.  The Level 3 Communications datacenter suffered a prolonged outage, it is reported that after a utility power outage occurred, the Uninterruptible Power Supply (UPS)  failed causing an unplanned outage that lasted approx. 12 hours. Effected customers included the publisher of IT Week leaving customer websites out of action.

Timely Reminder

These outages are a timely reminder that a datacenter's unlike other facilities evolve continuously reinventing themselves.

These outages have exposed the continuity problems, to ensure the essential services are maintained owners and operators must not overlook the essential E&M infrastructure that supports their operations. Click here to read more about Kelcroft's datacenter E&M infrastructure audits.

Further Information

If you need further information call the datacenter experts, contact Mr John Herbert at our Hong Kong office telephone +(852) 2335 9830 or by fax +(852) 2335 9862.

#datacenter #power #outage