A key Intel focal point is changing the narrative around the SPECrate 2017 benchmark. The high-tech industry has traditionally relied on one or two benchmarks (including SPECrate 2017) for a simple, standardized way to assess CPU performance, and consequently, overall system performance. IT decision makers use these benchmarks to choose systems that meet their needs, but performance requirements become more complex as workloads grow increasingly diverse.
Through a series of testing, we now have competitive data points to conclude that the SPECrate 2017 benchmark might not always provide all of the data customers need to analyze CPU performance. One step toward more effective benchmarking is to supplement CPU benchmarks like the SPECrate 2017 with proxy benchmarks that target specific workloads. Our workload similarity analysis concept can help customers identify more meaningful benchmarks to better assist with making the best IT purchasing decisions.
Explore the assets below to learn more about the importance of analyzing real world benchmarks in addition to SPECrate 2017 when assessing CPU performance.
- Workload Testing Report:
- Tested SPECrate 2017 as well as (8) other workloads across 3rd Gen Intel Xeon Scalable and AMD EPYC Azure instances – see where Xeon won:
- Workload Testing Infographics:
- Based on the testing results featured in the report above, explore individual infographics comparing the SPECrate 2017 result to specific workload results across both instances:
