"Root-cause analysis of Performance anomalies", João Paulo Magalhães

The paper is interesting in the sense of how he detects anomaly in performance, i.e. correlating the workload, the response time of the clients, and the variable monitored from the machine running a DB/WebApp