Congrats! You’ve got kubectl installed and ran your first job using Kubernetes. Just continue like this and the world will be your oyster. So let’s start on some more serious work.
Scaling your jobs
Jobs on the Playground Cluster default to using just 1 vCPU and 1 GB memory. You can use all of your free resources by specifying resources in the Spec file, but your free resource are limited.
To scale further, you can easily create a larger cluster in the portal and running your job on it. Our clusters support jobs up to 30 vCPU and 50 GB memory per job.
Diving deeper into jobs
Recurring jobs using Cron
If you need to run a job automatically at a certain interval, Kubernetes has a special Cron Job. This job type allows you to schedule a job, just like you with Cron.
Batch processing using a work queue
If you need to process many jobs, you might want to automate that using a queue. Because jobs in Kubernetes provide a lot of flexibility you can use them to set up a batch processing queue.
Other Kubernetes features
Jobs are just the tip of the iceberg when it comes to Kubernetes. Take a look at the Kubernetes documentation and at our list of supported features to see what else your Nerdalize cluster can do.
Creating custom Docker Images
If you’re looking for a Docker image to run, there’s a good chance someone already made and it’s available on Docker Hub. If it’s not, you can easily get started making your own images.
Need some help making your preferred software available in Docker? We can help out. Just contact us to discuss the possibilities.