Browse Prior Art Database

Optimizing the Build Process by Scaling Build Agents Based on System Need

IP.com Disclosure Number: IPCOM000247222D
Publication Date: 2016-Aug-17
Document File: 1 page(s) / 19K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to optimize the build process by scaling build agents based on system need and project dependencies.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 100% of the total text.

Page 01 of 1

Optimizing the Build Process by Scaling Build Agents Based on System Need

Build systems become backed up when multiple builds are simultaneously kicked off the system. The backup cascades due to dependencies. The problem is that the number of build agents (i.e., clients) needed is initially not known, which makes the build process suboptimal.

The solution is to optimize the build process by scaling build agents based on system need. When a project is built, the method is to build the dependency graph, traverse it, look at historical data of how long each project takes to build, and then determine how many of the projects can be built in parallel in order to determine the number of agents that need to be added to the existing pool. This optimizes the time that it will take to run the build.

The process for implementing the solution follows:

1. Access the dependency graph for the project being built 2. Traverse the graph and look at historical data for each project in the graph 3. Calculate the new average build time for each project in the graph 4. Identify the projects that can be built in parallel and must be built before developers can build the next group of dependent projects 5. Use the combination of the average build times and the list of projects that can be built in parallel to determine the number of agents that need to be added

1