SD Times Open-Source Project of the Week: G-Profiler

Workload optimization firm Granulate lately introduced a brand new open-source venture referred to as G-Profiler that goals to assist firms measure the efficiency of code that’s in manufacturing. 

In line with Granulate, present profiling options require code modifications, and are usually arduous to make use of, resource-intensive, or costly. These issues are intensified in trendy environments, akin to Kubernetes-based environments or huge knowledge workloads, that require steady profiling of knowledge throughout whole clusters and for jobs throughout a number of batches. This typically forces groups to make use of these options just for restricted quantities of time. 

Granulate hopes that with its new open-source resolution, these challenges may be eradicated. G-Profiler aggregates profiling knowledge throughout a number of nodes and utility languages, with out requiring modifications to code. This enables groups to establish and optimize efficiency bottlenecks extra effectively, the corporate defined.  

G-Profiler is modeled after inner instruments Granulate created for its personal real-time steady optimization resolution. It determined to open supply this venture with a purpose to help the group in addition to improve trade consciousness of computing inefficiencies. 

“On this cloud-native age, code profiling is extra essential than ever for bettering utility efficiency, taming cloud prices, and rising margins. The G-Profiler is a really highly effective software that enables growth groups to realize visibility and enhance efficiency,” stated Asaf Ezra, CEO of Granulate. “As a consequence of complexity of implementation and efficiency overhead, many groups couldn’t afford to make the most of such instruments, so we’re releasing the G-Profiler to permit them to reap the advantages of a code profiler with out having to make modifications of their code. This can be a main milestone in our effort to commoditize real-time steady optimization with a purpose to allow hyperscale efficiency and cost-efficiency for all.”

G-Profiler is offered on GitHub, or as a free picture in AWS, Azure, GCP, or the Docker registry.