New VM image deployment can take minutes to complete as it is directly dependent on many quantifiable and unavoidable factors such as disk image size. However, one component of image deployment duration introduces both significant and inconsistent delay: network latency! Pulling images from local or cloud container registries into your Orka Cluster, while necessary, obstructs a primary objective for development and DevOps teams which is to minimize build times.



The Orka Cluster 3.2 Scheduled Caching feature now allows you to mitigate this costly and variable delay by scheduling the image caching before your automated pipeline jobs kickoff providing the shortest possible VM deployment time and doing so consistently. Scheduled Caching ensures VM deployment is both predictable and only takes a matter of seconds to execute. You can shift time consuming cluster node caching to more opportune times outside of your scheduled build jobs.

The fastest possible VM deployment times with Orka Cluster are achieved once images are cached in cluster node memory. It takes only a matter of seconds for the VM to boot and the network latency component of deploying a new VM image is eliminated from your CI pipeline process.

Teams concerned with monitoring and optimizing the speed of individual tasks will love the ability to separate out image caching duration, from their overall build execution time. With a more consistent means to measure your build time devoid of unnecessary VM deployment latency you can focus on other sources of delay to your CI pipeline’s overall execution time.

Configure and automate your Orka Cluster 3.2 node image caching directly from the Orka3 CLI. Once you’ve determined the optimal time and which images to cache, automate in your CI runners or your deployment scripts and you’re on your way.

Orka customers can get started with scheduled caching by upgrading to Orka Cluster 3.2. Submit a ticket to the MacStadium support team on the Portal to start the process.