The utilization of Compute Optimizer resulted in the successful running of nearly 20,000 jobs, with a total of 90,103 live migrations performed. Notably, 91% of the total job execution time was in the spot market. This data highlights the critical role of Compute Optimizer in ensuring the high reliability of successful execution of a significant majority of jobs. In contrast, running these jobs without Compute Optimizer would have led to failure for more than half of them.
The adoption of Exostellar’s Compute Optimizer solution by Astera Labs has proven to be a transformative move with significant business implications, resulting in enhanced cash flow and decreased capital expenditure (CapEx) requirements.
1. Cost Reduction: By optimizing resource utilization and leveraging spot instances, Compute Optimizer helps Astera Labs achieve significant cost reductions in its AWS spending. Exostellar achieved cost reductions of up to 50% for on-demand compute resources which directly translates into lower expenses for Astera Labs, allowing the company to allocate its financial resources more efficiently. This also allowed Astera Labs to run more workloads in a shorter time period for the same cost as before, thus accelerating the chip development process and reducing time to market.
2. Operational Expenditure (OpEx) Model: Compute Optimizer operates on an OpEx model, which means that Astera Labs pays for the cloud resources it uses on a consumption basis. This model eliminates the need for upfront capital investments and long-term commitments associated with buying reserved instances (or compute savings plans) from public cloud providers (AWS/ Azure/ GCP). By shifting to an OpEx model, Astera Labs can reduce its CapEx requirements and free up cash flow that would have otherwise been tied to infrastructure investments.
3. Scalability without Capital Investment: Astera Labs can scale its compute resources without the need for additional capital investments. Compute Optimizer’s ability to seamlessly migrate workloads between on-demand and spot instances enables Astera Labs to increase its compute capacity by up to 2x within its existing budget. This scalability allows Astera Labs to meet the fluctuating demands of its projects.