Google today rolled out a series of new products that help developers track down performance problems in their code.
The Google Cloud Stackdriver APM suite offers the same capabilities it uses internally to monitor and tune the performance of its applications, the company said.
Google product manager Morgan McLean said: “The foundation of our APM tooling is two existing products, Stackdriver Trace and Debugger, which give you the power to analyze and debug applications while they are running in production, without impacting UX.”
Debugger simplifies root-cause analysis for hard-to-find production code issues, so that users don’t have to manually add new log statements to application code; redeploy affected services, analyse logs to determine what is going wrong.
Trace allows users to analyze how customer requests propagate through an application; useful for reducing latency and performing root cause analysis. It continuously samples requests, automatically captures their propagation and latency, presents the results for display, and finds any latency-related trends. Users can also add custom metadata to their traces for deeper analysis.
McClean added: “On top of that, we’re introducing Stackdriver Profiler to our APM toolkit, which lets you profile and explore how your code actually executes in production, to optimize performance and reduce cost of computation.”
Google also announced integrations between Stackdriver Debugger and GitHub Enterprise and GitLab, adding to its existing code mirroring functionality for GitHub, Bitbucket, Google Cloud Repositories, as well as locally-stored source code.
All of the tools work with code and applications that run on any cloud or on-premises infrastructure, providing an APM toolkit to monitor and manage the performance of applications irrespective of where and how that is accessed.
Evan Yin, a software engineer at Snap Inc. was an early user. He said: “We used Stackdriver Profiler as part of an effort to improve the scalability of our services. It helped us to pinpoint areas we can optimize and reduce CPU time, which means a lot to us at our scale.”
Others use the tools to identify performance issues driven by code.
Nicolas Fonrose, the CEO of cloud analytics specialists Teevity, said: “Profiler helped us identify very slow parts of our code which were hidden in the middle of large and complex batch processes. We run hundreds of batches every day, each with different data sets and configurations, which makes it hard to track down performance issues related to client-specific configurations.”
Stackdriver Debugger is currently free, as is the beta of Stackdriver Profiler. Stackdriver Trace includes a monthly quota of free trace submissions.
