
Data Compression Overview
114061 Rev. A 2-3
Data Compression Performance
The goals in using data compression are to achieve a high compression ratio while
maximizing throughput. Compression ratio is the size of uncompressed data
compared to the size of the same data after it has been compressed. Throughput
refers to the speed of transmitting data.
The compression ratio varies according to the effectiveness of the compression
algorithm, but also according to the characteristics of the data you are
transmitting: data that includes a lot of redundant strings compresses at a high
ratio.
Throughput, or network response time, varies with the number of elements in the
network that the data must traverse, including the compression/decompression
process.
Software Compression
Bay Networks software-based data compression is most effective for sites that
have WAN connections at relatively low speeds such as 56/64 KB, where you
want to achieve data compression at low cost, and with minimal memory
requirements. It supports up to 512 Kb/s on the FRE-040. For networks operating
at faster speeds, you should use hardware compression.
Hardware Compression
Bay Networks hardware-based compression works with Frame Relay and PPP
networks. It best serves sites that support T-1/E-1 lines, which often concentrate
many lower-speed remote connections. Our hardware compression facility
operates at high speeds, and also supports high-density WAN connections. Use
hardware compression when you want to achieve high compression ratios and
throughput, and also want to preserve router memory to perform other functions.
We offer two compression daughterboards for the BN and two compression net
modules for the ASN.
Commentaires sur ces manuels