OpenCensus vs TraceView

Track every machine involved in a transaction and identify bottlenecks in a single click.;Isolate interesting calls and drill down to the line of code and machine it ran on.;Tie together code and infrastructure metrics with database, service, and cache calls, all in the context of a single transaction.;Smart Tracing means you can run in production, with negligible overhead (< 1%).;Understand spikes in the graphs, even if the cause is only a single outlying request.;Visualize request patterns in one place to separate the noisy from the systematic.;Cross-correlate information between multiple sources. Want to find the controller and action that scans entire MongoDB collections? Done, in a single screen.;Zoom-in on trends based on performance - no guessing involved.;From mobile to desktops, understand your customer’s true web application experience from anywhere around the world.;Track server, network, and in-browser latency based on exactly what the user sees, as measured from directly inside your users’ browsers.;Tie end user experiences back to actual server side transactions.

Standard wire protocols; Consistent APIs for handling trace and metric data