Application acceleration: Compression and centralisation

Application acceleration has made it possible to move to a centralised architecture while optimising the WAN at the same time. This is due to a key facet of application acceleration: compression. In this tip, Robbie Harrell explains how compression techniques work and provides advice for considering application acceleration and compression for WAN optimisation.

Regulatory and competitive pressures are driving enterprises to centralise and consolidate application services and the resources supporting application services. Centralised management provides a mechanism for ensuring data integrity and audit trails for federal regulations such as Sarbanes-Oxley and HIPAA. Consolidations drive down infrastructure costs while supporting an operational model that can be maintained with fewer resources. It's not surprising that multiple technologies have come along that support this requirement within the enterprise IT space.

The advent of application acceleration technologies and services is providing a wealth of benefits to enterprise organisations looking to move toward a centralised and consolidated model. Prior to application acceleration, enterprises could move toward a centralised model, but this was followed by an increase in bandwidth over the wide area network (WAN). Enterprises had to shuffle the centralised data all over the WAN and incur the additional costs of the bandwidth and circuits.

Application acceleration has made it possible to move to a centralised architecture while optimising the WAN at the same time. This is due to a key facet of application acceleration: compression. Compression techniques have been around for quite some time and can commonly be found on routers as well as outboard appliances.

Here's a little background on how this technology works. Compression techniques rely on the fact that data in its binary form is redundant and somewhat predictable. Data text is generated with some rules, and compression techniques rely on predicting the data which will be transmitted. If the data can be predicted, compression techniques can use this information to reduce the amount of data used to represent text, graphics and sound. Compression isn't making the data any smaller; it's just encoding the data in a more efficient way. Binary data has the same bits represented over and over. Compression techniques use algorithms (several are mentioned below) to predict the redundant data accurately so that it doesn't have to be transmitted over the network every time.

Cisco has compression capabilities within the Cisco IOS that allow for the configuration of the Cisco router to compress and decompress traffic at each end of the link. Cisco can support the following compression techniques:

  • Van Jacobson Header Compression for TCP/IP, which conforms to RFC 1144.
  • Per-interface compression (also called link compression), which has one set of dictionaries per hardware link.
  • Per-virtual circuit compression (also called payload compression), which has one set of dictionaries per compressed virtual circuit.
  • Microsoft Point to Point Compression (MPPC), which is defined in RFC 2118 and is a per-PPP connection compression algorithm.
  • Frame relay payload compression (FRF.9) is per-DLCI.

These are just a few examples of compression techniques. Compression can be accomplished with many different vendors, but advances in compression techniques and the combination of compression with caching, application acceleration and wide area file services (WAFS) has provided organisations with a complete solution for optimising the WAN while providing centralised services.

The point is that compression is only a part of the picture. Most vendors in the application acceleration space have built their application acceleration techniques to focus on common applications that are found enterprise-wide. ERP, CRM, Exchange and database applications are known to be bandwidth hogs. Centralisation of these applications can increase bandwidth significantly, and compression will help only for so long.

Centralisation and consolidation make good sense from a cost perspective, but the lost productivity from poorly performing applications and increased bandwidth is hard to discount. If your organisation is looking into centralisation, and compression is a major consideration, be sure to include application acceleration and WAFS as a part of the analysis.

About the author: Robbie Harrell (CCIE#3873) is the National Practice Lead for Advanced Infrastructure Solutions for SBC Communications. He has more than 10 years of experience providing strategic, business and technical consulting services. Robbie lives in Atlanta and is a graduate of Clemson University. His background includes positions as a principal architect at International Network Services, Lucent, Frontway and Callisma.

Read more on Network monitoring and analysis