CloudBees Accelerator GCP Marketplace Quick Start
CloudBees Accelerator (“Accelerator”) from CloudBees, Inc. is a high-performance, drop-in replacement for GNU Make and Ninja build facilities that dramatically reduces software build times. Unlike other parallel build tools, Accelerator guarantees that builds will not fail because of incomplete dependencies, thanks to patented conflict detection and correction algorithms. So your build results—including the build log—are identical to those from serial builds, but they happen with the speed of parallel builds by distributing them over numerous inexpensive machines on a LAN or a public cloud such as Amazon EC2 or Google Cloud Platform. You can also use Accelerator to accelerate static analysis (such as the Coverity, Klocwork, and Parasoft tools), unit testing, and linting tasks.
Accelerator has two primary user-facing parts: Electric Make (”eMake”)—the command-line front end to the cluster, and the Cluster Manager web UI—a web-based dashboard for managing the cluster and viewing various types of cluster activity. Also, for Visual Studio, Accelerator includes an add-in that lets you execute builds using the cluster directly from the Visual Studio IDE. In most cases, switching from another build tool to Accelerator is as easy as replacing
You do not need to request or import a license file to run Accelerator on GCP Marketplace.
Launch Accelerator on Compute Engine.
Visit the CloudBees Accelerator marketplace solution page and click the LAUNCH ON COMPUTE ENGINE button:
Choose your parameters for deploying Accelerator.
Specify your deployment name, zone, and machine parameters. The recommended machine for the best performance is an n1-standard-32 (32 vCPUs, 120GB RAM) and a 500 GB persistent SSD:
Click the Deploy button and wait for the deployment to complete. This will take a few minutes:
Proceed to the next section when the deployment is complete:
Connect to your Accelerator VM.
Use the Connect to Accelerator VM button to open a Cloud Shell terminal to your VM:
Set up your project resources.
On your GCP Marketplace instance, download or check out your source code, and install any additional tools needed to execute your build:
git clone [<options>] <repository> [<directory>]
Run an eMake “learning” build.
Change to your project directory and invoke eMake by entering
By default, eMake virtualizes just the current working directory across the Accelerator build cluster. If your source code, output files, and build tools are in different directories, specify those directories by adding the
--emake-root=<path1>:<path2> … :<pathN>option. For example, enter:
Starting build: <build_number>appears. For example,
Starting build: 1. When the build is finished, a message such as the following message appears:
Finished build: 1 Duration: 0:41 (m:s) Cluster availability: 100%
Clean your build output tree.
Clean the output (for example, by using
make clean) to prepare for a fully accelerated build. Do not delete the eMake history file (
emake.data) or the eMake asset directory (
Run an eMake accelerated build.
Run eMake again, from the same directory, using the same options you used the first time:
Sign in to the Cluster Manager web UI at
https://<server_hostname_or_IP>/ using the default administrator user name
admin and password
changeme. You should change the default password as soon as possible. (If the Cluster Manager web server does not use port 443, append
:<port> to the host name or IP address.)
A security warning appears because of an invalid certificate. If you want to eliminate this warning, replace the self-signed Apache certificate with one that is signed by a certificate authority. For details, see Installing an Apache Server Certificate.
links to documentation and other assistance, instruction for novice users, and real-time charts for agent usage and build durations, click the Home tab.
a list of builds and other build information, click the Builds tab.
a list of agents participating in builds and information about those agents, click the Agents tab.
reports using many types of analytics about your build cluster and several types of data visualization such as scatter plots and pie charts, click the Reports tab.
messages from the Cluster Manager, agents, and eMake, click the Messages tab.
status for Azure Services Platform, Amazon EC2, Google Cloud Platform, or Kubernetes instances that are active (that is, instances that are running builds), click the Cloud tab.
details about users, click the Administration tab.
You might need to open the Cluster Manager web server port by configuring the port number in the firewall rules for your GCP Compute Engine VM instance. For details, see https://cloud.google.com/vpc/docs/vpc?hl=en_US#firewall_rules.
For basic Accelerator terms and concepts, explanations of the relationships among the parts of Accelerator, and more information about the Cluster Manager web UI, see CloudBees Accelerator Terms and Concepts. For configuration procedures, see the CloudBees Accelerator Installation Guide and the CloudBees Accelerator Configuration Guide. For details about eMake, see the CloudBees Accelerator Electric Make User Guide.
For assistance, go to the CloudBees technical support site at https://support.cloudbees.com/ and submit a support request. You can also go to https://ask.cloudbees.com/—a member-moderated community forum where you can ask and answer questions as well as comment on (and vote for) the questions of others and their answers.