Feature

Will super compression slow PCs?

As reported in Computer Weekly last week, dramatically high file compression ratios are being promised by US research group ZeoSync, but the company has yet to provide proof of concept and will need to explain more clearly how the system works, writes Eric Doyle.

As the company admits, its Relational Differential Encoding (RDE) technology is still about 18 months away from becoming a product that could revolutionise the data communications and storage industries. On offer is a compression algorithm that will reduce files to one hundredth of their size - current methods, at best, only offer 10 times compression.

The breakthrough is the ability to compress "singularities". When a file is compressed, repeated strings of binary code are sought out, replaced by a short marker, and stored only once. When the file is decompressed, the markers are replaced by the original string. After one or two passes, the code that results comprises unique strings, known as singularities.

According to Peter St George, ZeoSync's chief executive, the company has devised an algorithm that can even compress the singularities to collapse code several hundred times its original length.

The main problem facing the company is to show that encoding and decoding can be done quickly. When PC data compression first appeared, specialist Stack, for example, provided a dedicated PC board for its compression system to take the load off the PC's central processor. In later years, more powerful Intel processors appeared that could soak up the overhead for encryption. For widespread acceptance, RDE has to prove that it can increase compression without noticeably slowing down the computer.

David Hill, research director for storage and storage management at analyst firm the Aberdeen Group, said ZeoSync's proposed system will have a profound effect if it does come to market. "There are at least two challenges that ZeoSync will face," he said. "First, will the technology be able to scale to encode and decode large bit streams in a reasonable time without a huge investment in processing power?

"Second, will the software be able to be 'frozen' as firmware in hardware such as modems or satellite transmitter/receivers, given that the algorithm to be frozen has to fit the silicon's parameters, such as size?"

Email Alerts

Register now to receive ComputerWeekly.com IT-related news, guides and more, delivered to your inbox.
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

This was first published in January 2002

 

COMMENTS powered by Disqus  //  Commenting policy