How Many Computers?

In my earlier post On Being The Right Size, I discussed the scenario in which all the computing power in the world is concentrated into a handful of massive servers, with billions of intelligent devices connected into a single network. As John Gage of Sun once quipped, and Sun executives never tire of repeating, The Network is the Computer.

So I find it strange that cyber-crime is to be defined in terms of the number of computers damaged in an attack. [Senate Approves Cyber-Crime Bill eWeek 4 August 2008]. Robin Wilton notes that this decouples 'theft' and 'harm', and wonders how the notion of damage is to be defined. But we also need to count the computers that were damaged. If the network is the computer, that's only one computer. Haven't these guys read Asimov? 

Perhaps this suggests an interesting line of defence for Gary McKinnon, a hacker who cracked the computer systems of the Pentagon and Nasa from his bedroom in north London more than seven years ago, and who is now to be extradited to stand trial in the US. [See analysis by Duncan Campbell via Bruce Schneier.] How many computers were damaged? Really? And how many bruised egos?

Skype Skuppered

It turns out that it was Microsoft that brought down Skype for two days earlier this month. Microsoft's monthly software update (known as Patch Tuesday) triggered millions of computers to reboot at the same time, which always puts an unusual strain on major Internet companies such as Skype.

Source: The Register, August 20th 2007.
See also TechCrunch August 20th 2007.

As Alex from RiskManagement Insight points out, this is equivalent to a form of DDOS (distributed denial of service) attack. From a risk management perspective, it may not matter very much whether an attack is deliberate and malicious, or merely an accidental side-effect of some entirely innocent action.

Although Skype had survived previous Patch Tuesdays without incident, it seems that this month's Patch Tuesday triggered a previously unknown bug in Skype's software. As Alex points out, it is practically impossible to construct a test environment large and complex enough to simulate this scenario.

I haven't seen any figures, but I have little doubt that Skype's competitors (including Microsoft) must have experienced an unusually high level of new registrations during Skype's misfortune. Now that we have become accustomed to free voice calls over the Internet, it seemed outrageous to return to the almost mediaeval practice of paying real money for talking over the telephone, so my colleagues and I signed up to Yahoo Messenger.

It's an ill wind ...

On Being The Right Size

My son brought a reading book home from school yesterday, called My Friend Mr Leakey, which turned out to be by the eminent scientist J.B.S. Haldane. I didn't know he had written any books for children, but I did know he had written accessible books for grown-ups as well as professional scientific papers. One of his most widely-read works is an essay called On Being The Right Size.

According to myth, Thomas Watson of IBM once predicted a global market of perhaps five or six computers. (There is no documentary evidence of this prediction.) Greg Papadopoulos, CTO of Sun Microsystems, now makes a similar prediction: The World Needs Only Five Computers. (See also interview with Greg by Stephen Shankland.) 

We're talking pretty massive computers here. Think Amazon, eBay, Google, Microsoft, Wikipedia and Yahoo, each with thousands of servers linked together into massive server farms. In order to make sense of Greg's prediction, we have to regard the Google grid as a single computer. And there is a plausible scenario in which all the computing power in the world is increasingly consolidated into a handful of massive global grids. 

And of course Sun Microsystems has long been associated with the slogan "The Network is the Computer", coined by John Gage in 1984. So what is the right size for a computer?

Wikipedia: John Gage, J.B.S. Haldane, Thomas Watson sr

Related post: How many computers? (August 2008) 

Update (2021). My list of examples would be slightly different today, but would still include Amazon/AWS, Google and Microsoft/Azure.

 

Werner Vogels on Scalability

Werner Vogels (Amazon CTO) has just posted some notes about Scalability, which he defines thus: 

A service is said to be scalable if when we increase the resources in a system, it results in increased performance in a manner proportional to resources added. 

Scalability is relevant for tools and methods. I have been shown demonstrations of tools that look pretty cool when you've got five web services, but are pretty hopeless when you've got five hundred or five thousand web services. What does a tool do when you have more objects than can reasonably appear on a screen - even with scroll bars? Was the multi-screen, multi-user option conceived from the start, or does it look like an afterthought? 

 

See also Scale and Self-Organization (July 2007)