Intel for Hosting: Dynamite Comes in Small Packages

April 18th, 2011 by

As the march towards hardware miniaturisation continues, one tech giant straddles the past and future with regards the direction of web hosting servers: Intel.

Their microprocessors which occupy less space than 1/10th the wavelength of light  (or a strand of hair) are the brainpower which drives UK2’s massive 24 GB DDR3 RAM range of dedicated servers.

At one point Intel owned 92-98% of the server microprocessor market until it started losing ground to rival AMD in 2005.   This sparked pundits, as they had done in the past, to prophesy the death of Moore’s Law, which stated that the numbers of transistors incorporated in a chip will approximately double every 24 months.

This law had historically allowed Intel to reduce costs and increase profits while at the same time extending the remarkable reputation of seeding 100+ million sold computers which all worked when a consumer or web host hit the power switch.

In 2008, when Intel released the successor to the Core micro architecture, code-named Nehalem, many stood back and gasped at chips no taller than 45 nanometers (nm) in height.

This was an incredible achievement if one considers that each chip may contain over 400-million transistors.   Intel’s original 4040 processor released in 1971 held 2300 transistors and cost more than $1.  By 2008 that cost had dropped too less than 1/10,000th of a cent.

The Nehalem line soon sprouted the Xeon array of dedicated servers, which UK2 offers to its customers who seek extra juice, speed and computational power.

Intel then defied the critics of Moore’s Law by releasing Sandy Bridge in January 2011 which shaved 13nm off new chips resulting in faster processing speeds, greater computing capability and more sophisticated applications.

Intel cured a few gremlins in the process, including electrical current leakage, which occurs, as transistors get smaller. It did this using something called metal gates and High-k gate dielectrics to compensate for silicon “shrinkage”.

“The average time to design, test, and fine tune to make a chip ready for fabrication takes hundreds of people working full time for two years,” explains Intel.

“Intel builds chips in batches on wafers in fabrication facilities. We use wafers of silicon, a natural semiconductor. The wafers are sliced from 99.9999% purified silicon ingots and polished to a mirror smooth finish.”

That’s right: Intel Microprocessors originate from something as common as beach sand which is then alchemized into a glass compound and finally shaved into impossibly thin slices that become home to networks of transistors.

A single microprocessor may eventually contain billions of transistors interconnected by fine wires made of copper.

“Each transistor acts as an on/off switch, controlling the flow of electricity through the chip to send, receive and process information in a fraction of a second. Intel’s microprocessors have evolved from single core processors to dual core and quad core processors. Quad core processors deliver four computing “brains” inside a single package.”

But to pull off this amazing feat the lab where technicians (in white coats)  work at require that the air around them be thousands of times cleaner than a typical hospital operating room.

Inside these pristine centers, hundreds of identical microprocessors are created in batches on a single silicon wafer, which are cut using a diamond saw to separate them.

“Each functioning die is assembled into a package that protects the die. This package delivers critical power and electrical connections when placed directly on a computer circuit board, or other devices such as cell phones or personal digital assistants (PDAs).

“The implementation of high-k and metal gate materials marks the biggest change in transistor technology since the introduction of polysilicon MOS transistors in the late 1960s,” said Intel.

The new Xeon line processors, based upon 32 nm chip sets include up to 10 cores with Hyper-Threading, and deliver up to 40 percent greater performance than previous Xeon 7500 series processors.

The new servers will allow web-hosting companies to deploy powerful virtualization software, the technology behind virtual private servers (VPS) or commonly referred to as cloud hosting.

A recent Forbes blog post stated that Intel recently released figures, which suggested Internet traffic exceeded 245 Exabytes in 2010. Intel believes this figure will quadruple by 2015.

Further, it expects 1 billion people will join the Internet within 5 years using 15 billion new connected devices including PCs, Smartphones and tablets, embedded devices and smart TVs.

Intel is hoping all this increased activity will be one of the drivers keeping it at the top of the microprocessor manufacturer stack.

Guest Blogger: Jason Stevens from / Freelance web developer, tech writer and follower of cloud computing trends. Follow him on Twitter @_jason_stevens_

* reserve the right to agree or disagree with our guest bloggers. Call it freedom of speech, but our guest bloggers are entitled to have an opinion. If you wish to agree or  disagree, then feel free to leave a comment. Thanks for visiting our blog! If you wish to become a Guest Blogger for UK2, please contact our marketing department.

  • Share this post

5 Essential Plugins For Your WordPress CMS

Win an iPad 2 and Nintendo 3DS This Easter!


David Precious
# 18th April, 2011

Their microprocessors which occupy less space than 1/10th the wavelength of light (or a strand of hair)

…many stood back and gasped at chips no taller than 45 nanometers (nm) in height.

Um, no. Not even remotely close. You’re confusing the scale of the internal fabrication of the circuits within the chips with the size of the chips themselves.

Jason Stevens
# 18th April, 2011

David, good point: the reference to nm is aimed at the size of the ‘transistors’ as put more eloquently on the Intel site:

“A nanometer (nm) is one-billionth of a meter. The transistors on Intel’s latest processors are just 32 nm wide, while our previous generation processors are built on 45 nm and 65 nm architectures. This allows transistors to be packed more densely, leak less energy, produce less heat and switch faster, which leads to processors that run faster, use less power, and are more energy-efficient. It also means new features can be squeezed onto processors, so computer makers can focus on things like extending battery life, creating ever-sleeker designs and putting new wireless capabilities into devices. Lastly, it means Intel can keep bringing you better performance at relatively the same or even lower prices over time”

Source: (‘What’s this’ under column entitled ‘silicon technology’. Thanks for feedback.

Leave a Response