As part of your CodePipeline, you are running multiple test suites. Two are bundled as Docker containers and run directly on CodeBuild, while another one runs as a Lambda function executing Python code. All these test suites are based on HTTP requests and upon analyzing, these are found to be network bound, not CPU bound. Right now, the CodePipeline takes a long time to execute as these actions happen one after the other. They prevent the company from adding further tests. The whole pipeline is managed by CloudFormation. As a DevOps Engineer, which of the following would you recommend improving the completion time of your pipeline?