With the release of its Continuous Profiler to the public, Intel is demonstrating its open ecosystem strategy to spur innovation and increase developer productivity. Companies such as ironSource, ShareChat, and Snap routinely use the optimization agent to identify production bottlenecks and optimization opportunities.
Continuous Profiler is an open-source tool designed by Intel Granulate that integrates multiple profilers into a single flame graph display. This consolidated view provides an autonomous and ongoing methodology for developers, performance engineers, and DevOps teams to identify runtime inefficiencies.
“Continuous Profiler has been at the core of our work at Intel Granulate,” said Asaf Ezra, General Manager at Intel Granulate. “By assisting developers in identifying code bottlenecks, businesses can optimize their applications more effortlessly and efficiently.”
In every testing or production environment, one of the common tasks for performing analysis is understanding why central processing units (CPUs) are busy. Continuous Profiler illustrates the most active code paths in a flame graph.
“This visualized view makes it immediately obvious where CPU is consumed so you can discover cost savings, eradicate bottlenecks, enhance throughput, and diminish latency and performance regressions,” stated Brendan Gregg, Intel Fellow. “However, in the current complex environment, flame graphs can expose numerous performance benefits that applying them all becomes a cumbersome task. Intel Granulate automates this chore, enabling corporations to actualize these performance benefits presently and as Intel advances more enhancements in the future.”
– the story carries on underneath the graphic –
How does it operate? A unified, cohesive depiction of the tasks that a CPU is executing is produced by the Continuous Profiler, a creation of gProfiler that amalgamates numerous sample profilers. Easy-to-navigate flame graphs identify performance regressions, encompassing deadlocks and garbage collection, making deployments easier.
Continuous Profiler is one of the few tools that let DevOps sort services by hostname, container moniker, or Kubernetes deployment object. Selecting the resolution grade on the page is all that is required. With no necessity to initiate a profiler for each entity, these inbuilt Kubernetes filters profile from the deployment down to the pod level. Teams can inspect how diverse deployments, pods, nodes, and hosts behave in various facets and code owing to this.
Within a few minutes, the profiler can be distributed across the cluster and support multiple programming languages without the necessity for modifications to the code. It also dovetails with Intel Granulate‘s ongoing optimization services. The Constant Profiler is certified with SOC2.
The Continuous Profiler has a simple User Interface (UI) and its perfect integration allows it to recognize runtime bottlenecks, enhance application performance for improved user experiences, and make changes to meet the specific developer requirements. Intel asserts that they will persist in enhancing the tool and aiding developers in building applications that would be tougher, swifter, and more cost-effective with input from open source contributors and the wider ecosystem.
