A Service Provider was challenged with understanding if the Domain Name Service (DNS) was being negatively affected by components of the DNS service itself (client delay, DNS server performance, load balancer overload, etc.) or if the service was impacted by the transport layer's ability to deliver requests for information. Lacking the proper insight, they were unable to "defend" the innocence of the network.
With IBM SevOne Network Performance Management (NPM), the Service Provider can now determine the cause for a change in DNS response time performance by comparing application response times with synthetic transport latency metrics. This is accomplished from a single dashboard view without having to run manual reports, and has helped the Service Provider reduce churn and ensure near perfect uptime.
How SevOne NPM Helped
- Collected response time metrics from synthetic tests (Exfo/Brix) every five minutes and recorded the data in the SevOne NPM performance management database
- Established performance baselines for both DNS performance and the transport network performance to help "normal" levels
- Set alert conditions to figure when performance levels deviated from historical norms
- Provided a single dashboard view of transport latency alongside response time metrics to identify the cause of performance issues and improve mean time to identify performance events