PDF of this tutorialFrame relay–service providers offering SLAs must supply some type of report (hard copy or electronic) on a periodic basis, which covers the areas guaranteed in the frame-relay contract (e.g., availability, PVC delay, throughput, etc.). Historically, these reports have inaccurately reported on the key SLA metrics or have presented the data in a confusing and almost unintelligible way.
Some carriers have delivered reports based on data derived by the carrier–frame relay switch. These reports have not met subscriber–SLA requirements because they are carrier-network centric rather than end-to-end–application centric. Additionally, carriers have not supplied the following:
- interpretation of the report and its statistics
- any gaps in data gathering
- process for gathering the data
- suggestions for optimization of the network to limit the following:
- capital investment (routers, FRADs, etc.)
- recurring transmission costs (by identifying available bandwidth without increasing CIR through determination of unnecessary WAN traffic caused by spurious broadcasts, over-polling, and incorrect window-sizing)
- bandwidth growth by identifying rogue users or applications that are monopolizing WAN bandwidth inappropriately
- warning indication of degradation before users are affected
It is clear that even if hours are spent scouring over carrier reports, there still might not be sufficient data to determine whether or not the carrier is meeting the service-level guarantee. Additionally, key areas affecting the performance and reliability—as well as operation costs—of the network have been overlooked. Therefore, is it possible to optimize network investment and assure that contracted service-level guarantees are met? Yes, but only if network-management tools designed specifically to monitor and manage service levels on carrier–frame relay networks are utilized.
