This is an old archive page of HWZ prior to revamp. Please visit HWZ for the latest reviews and news.


» Articles
IDF Spring 2004 Update Part 3
By CPU-zilla
Category : Events
Published by Jimmy Tang on Friday, 20th February, 2004

Reshaping the Information Age

In the final keynote address today, Intel Senior Vice President and CTO Pat Gelsinger talked about how converged computing and communications industry will need to transform the underlying architectures of its products and applications in order to handle the vast amounts of data created by digital systems in the world today. Citing some examples, Gelsinger said that there are as much as 17 Exabytes worth of information created just in one year’s worth of telephone calls in the whole world today. That is a lot of data, considering that it’s just one single application alone.

Pat Gelsinger, Intel Senior Vice President and CTO.

As people in the world today get accustomed to the digital lifestyle, the vast amounts of digital information generated and transmitted by every individual would require systems to provide some sort of storage, manipulation and retrieval. Gelsinger further points out, “Our society is creating massive amounts of complex data as the world continues to go digital, but it doesn’t have the capability to enjoy the full potential of this rich resource.”

Sharing what is known as the “RMS View” which simply means Recognition, Mining and Synthesis, Intel believes that these three basic classes of workloads will form the core of future applications.

“Recognition” involves systems with the capability to recognize different sets of data patterns and models of interests specific to a user or application, while “mining” involves the retrieval of large amounts of datasets for these patterns from large databases and information sources. Finally, in order that all these have meaningful use, systems must be able to “synthesize” or analyze these large sets of data and present them in a usable manner to the user in ways that we’ve never thought before or not possible at all today.

These new workloads will bring about a new class of algorithms, data structures and new future architectures. Obviously, all these will drive the need for more memory, processing power, bandwidth and storage. Looking at just processing power alone, you’ll need more than TeraFlops of processing capability to digest all that TeraBytes of information. Certainly it is not about driving frequency on processors anymore, but there is a greater need to drive fundamental architectural changes to what is currently available today.

Page 1 of 4 | Next>>