Wednesday, August 22, 2007

Routers, Switches, and Firewalls: Marketing Benchmark Numbers versus Reality

A consistent problem when reading datasheets for networking devices (routers, switches, firewalls) is that the throughput numbers offered by the vendor are not useful without context and are coming from a bias source. Today I happen to be reading through a SonicWALL firewall datasheet and I notice a reference to RFC 2544 in the fine print:

**Testing Methodologies: Maximum performance based on RFC 2544 (for firewall). Actual performance may vary depending on network conditions and activated services
***VPN throughput measured using UDP traffic at 1280 byte packet size adhering to RFC 2544
****Throughput measured using HTTP throughput test
Well, RFC 2544 has apparently been around since 1999. It suggests a framework for a standardized methodology for benchmarking networking devices. I'm not sure why I've never come across it before. (Note: I haven't read through it yet so I'm not endorsing it).

No comments: