Service management and use cases
Service management was another crucial organization function that was not formalized so far. Now, Cassandra was deployed in a cluster; application and web servers had their own clusters. The backlog of features to have service supportability and application supportability as well as service management needs were huge, but the top priority was to have a central log containing the following contents:
Application logs
Network logs
Performance metrics
Audit data
Monitoring data
All of these were to be classified into different classes such as <APPLICATION>:INFO, <APPLICATION>:WARNING, <APPLICATION>:ERROR, <NETWORK>:INFO, <NETWORK>:WARNING, <NETWORK>:ERROR, <METRIC>:LOW, <METRIC>:25TH, <METRIC>:50TH, <METRIC>:75TH, <METRIC>:90TH, <METRIC>:95TH, <METRIC>:99TH, <METRIC>:99.99TH, AUDIT, MONITORING, and so on, and were to be captured in a central log-like data store. Cassandra was doing...
