Looking at the innards of application acceleration

Insightful analysis by consultants Steve Taylor and Jim Metzler, plus links to the latest WAN news headlines

All too often, today's products for application acceleration as assumed to run on FM (Fudging Magic), and, in fact, resources to determine the fundamental difference between products are too few and far between. However, the title of the paper "The Benefits of Byte-Level WAN Deduplication" belies its fundamental usefulness as an opportunity to look at two different methods for sending "symbols" across the network in order to reduce the total amount of data sent.

As explained in the paper, there are two ways of avoiding resending the same data repeatedly - “token-based” and “instruction-based.” And each has its advantages depending on the type of data that is predominantly being transmitted.

In a token-based method, a particular bit sequence (token) is used to represent a “chunk” of data (to use the technical term). If the chunk of data that is represented by the token is very large, then you get quite significant data reduction. For instance. if the token is a 16-byte token and it’s representing 10,000 bytes, you’re getting a data reduction of 625:1. On the other hand, if the chunk of data is only 100 bytes and you’re still using a 16-byte token, the reduction is only about 6:1.

To continue reading, register here and become an Insider. You'll get free access to premium content from CIO, Computerworld, CSO, InfoWorld, and Network World. See more Insider content or sign in.

All too often, today's products for application acceleration as assumed to run on FM (Fudging Magic), and, in fact, resources to determine the fundamental difference between products are too few and far between. However, the title of the paper "The Benefits of Byte-Level WAN Deduplication" belies its fundamental usefulness as an opportunity to look at two different methods for sending "symbols" across the network in order to reduce the total amount of data sent.

As explained in the paper, there are two ways of avoiding resending the same data repeatedly - “token-based” and “instruction-based.” And each has its advantages depending on the type of data that is predominantly being transmitted.

In a token-based method, a particular bit sequence (token) is used to represent a “chunk” of data (to use the technical term). If the chunk of data that is represented by the token is very large, then you get quite significant data reduction. For instance. if the token is a 16-byte token and it’s representing 10,000 bytes, you’re getting a data reduction of 625:1. On the other hand, if the chunk of data is only 100 bytes and you’re still using a 16-byte token, the reduction is only about 6:1.

By contrast, “instruction-based” systems use tokens to indicate where changes in the data occur as compared with a prior instance. For example, if a file is sent a second time and there are only minor changes, then the capability to send only the changes and not to retransmit large amounts of unchanged data becomes much more efficient.

The bottom line is that when you’re evaluating application acceleration appliances, it’s worth a detailed look under the hood to see exactly how the amount of data to be sent across the network is reduced. Depending on the types of changes that are sent in your most common applications, this can make a tremendous difference in the amount of acceleration that’s realized.

(Check out Network World's Buyer's Guides to compare Application Acceleration and WAN Traffic Optimization products.)

Read more about lans & wans in Network World's LANs & WANs section.

Steve Taylor is president of Distributed Networking Associates and publisher/editor-in-chief of Webtorials. Jim Metzler is vice president of Ashton, Metzler & Associates.