How can you save money with Preemptible VMs?

You have your workload in production on Google Cloud, so now what? The next step is to do more work without going over budget. Google Cloud offers reduced-rate cloud instances, called Preemptible VMs. Preemptible VM pricing varies by machine type, but can represent a cost savings of up to 80%. Since the prices are fixed, you can easily predict your spend. The tradeoff for the lower prices is that the instances are subject to being taken away on short notice in order to meet demand from the regularly-priced instances and they are terminated after 24 hours. Preemptible VMs also cannot have GPUs attached and do not receive sustained use discounts.

For customers looking to get the most out of Preemptible VMs, CycleCloud™ makes using them easy and effective with a single click, with features like:

  • Easy requests across multiple machine types and Zones
  • Automatic replacement of lost instances

Using Preemptible VMs in CycleCloud

CycleCloud software has unique features that makes it easy to not only use Preemptible VMs, but use them effectively. Preemptible VM capacity varies by machine type and Zone. With CycleCloud, you can specify multiple machine types and Zones to make it easy to get the capacity you need. CycleCloud software automatically spreads the requests out across all combinations you choose. If you lose some instances, CycleCloud automatically requests replacement instances from the remaining combinations that have capacity.

CycleCloud cluster view showing 50,000 cores from Google Cloud Preemptible VMs spread across four zones.

To use Preemptible VMs in CycleCloud, set Preemptible = true in the nodearray section of the cluster template.

Job considerations

When using Preemptible VMs, your jobs need to be interruptible. This means that they can either restart from the beginning or perform checkpoints to save state. Application data must be resilient and external actions like database operations have to be idempotent (i.e. re-running will not change the output).

In order to reduce “badput” (i.e. bad output) due to instance loss, the most common approach is to modify the workflow to go wider with smaller, shorter-running jobs. This helps to minimize the impact of any one instance being terminated. For jobs that can’t be easily made to go wider, it’s sometimes possible to run the jobs for shorter periods and use the output of one job as the input of the next. For example, instead of running 100 timesteps in a simulation as a single job, break it into 10 jobs of 10 timesteps where the final state of job 1 is the initial conditions for job 2 and so on.

With its set of focused features, CycleCloud software provides several features to make using Preemptible VMs easy and effective. These include:

  • Easy requests across multiple Zones and machine types
  • Automatic replacement of lost instances

Contact us to see how CycleCloud can help you get more out of your cloud usage.

Share this: