Especially challenging is traffic monitoring of hybrid networks that are both in the cloud and on premises at a price that makes attaining network excellence cost-effective.
Enterprises must keep tabs on their networks in real time to find and fix glitches before they affect their customers. Quickly finding and resolving problems like traffic congestion, jitter, packet loss, faulty hardware or software or misbehaving apps is the key to customer satisfaction excellence.
Cirries’ Data Analysis in Real Time (DART) solution includes a packet sensor that can be deployed in public and private Virtual Machines, containers, or Kubernetes clusters. This packet sensor is a software-based, intelligent network probe that processes packet streams into highly analytic metadata summaries for transmission to DART. The reduction of raw packets to analytic metadata reduces public cloud data transmission costs significantly. The packet sensor has the flexibility to collect both north-south and east-west workload packet streams.
The packet sensor high-performance packet collection and processing engine incorporates both an input filter and analytic metadata generator. The input filter discards duplicate and non-IP packets and can optionally select IP source and destination addresses and subnets of interest.
The metadata generator in addition to creating flow metrics, can assign application names to destination IP addresses by either manual assignment or by monitoring DNS lookups.