This week we're testing some WAN acceleration gear in the lab and since its on my mind now and has been for a few weeks I thought I'd spend some time talking about WAN performance, optimization, and acceleration.
Management of WAN connections can be a real bugger. Many of us "cut our teeth" as network engineers troubleshooting WAN circuit problems, arguing with our service providers about link states, and trying to figure out DLCI problems on frame-relay PVCs. Seems like nowadays the problems we see on WAN circuits are different. It's not so much that the links are unreliable from an availability perspective, but that they are unreliable from a performance perspective. Part of these problems are self-induced I suppose. When a companies decide to rollout new applications, we're sometimes the last ones to hear about it irrespective of the fact that the application may have significant requirements in the areas of WAN latency and bandwidth (and they wonder why they don't work right). These applications combined with VoIP, Video over IP, SaaS, and a general increase in internet usage from the workplace have caused a surge in the usage and expectations for our WANs (nobody was watching YouTube 3 years ago).
Anyhow, I'm going to dive into some of the key components of WAN optimization in later posts this week. Drop me a line if you have a particular interest and I'll try to add some content in that area.
p.s. I'm taking a trip out West to meet with a few of the leading vendors that provide WAN optimization and acceleration technologies. Let me know if there's any information I can gather for you while out there.