Read/Download White Paper (PDF)

After our successful, proof-of-concept (POC) testing of our on-premise platform, where we found that the Granulate software reduced 25% of memory and 44% of CPU utilization, while offering potential cost savings, we began testing in the cloud Databricks data lake. Collaborating with our technology partner Accenture, Intel IT performed a four-week POC with the active Granulate agent in our Production and Quality Assurance (QA) environment running on the Databricks platform securely in the public cloud. The results included a 23% improvement in vCore compute consumption, a 17% increase in data throughput, and reduced CO2 emissions, aligning with the on-premises benefits we observed previously for another big data workload.

Tags: autonomous optimization, Big Data, carbon emissions reduction, cloud cost savings, cloud optimization, compute optimization, data throughput, databricks, energy efficiency, granulate, Green Computing, Hybrid Cloud, machine learning optimization, resource utilization, spark optimization, sustainability, vcore utilization, workload efficiency
 
Posted in: Intel, Intel IT, IT White Papers, IT@Intel