Service call graphs for website performance

摘要: Techniques for generating, storing, and using service call graphs are provided. In one technique, trace data items generated by different services correlated to generate a graph. Trace indicates when certain called their respective latencies as result of client request. A graph may reflect single or multiple traces over particular period time. be analyzed inform administrators web site how application the it relies on performing. used determine whether there sufficient resources support projected increase in traffic application. estimate cost Multiple compared more root causes performance problem.