Real-World Power

In the real world you do not run your virtualized servers at their maximum load just to measure the potential performance, but neither do they run idle. The user base will create a certain workload and expect this workload to be performed with the lowest response times. We created a real world “equal load” scenario as we described here.

vApus Mark II Real world energy

The numbers above show that there is more to energy consumption than just measuring idle and full load. The quad Xeon needs 67% more energy to run the same workload. Granted, our methodology favors the Opteron a bit: the average CPU load is around 5-30% for the 80 thread Xeon E7, while the Opteron is running at an average of 10-40%. So the load on the Xeon E7 is a bit too low. Still, the real problem is that the Xeon E7 power consumption is high at low loads.

Power Extremes Conclusion
Comments Locked

62 Comments

View All Comments

  • Soulnibbler - Thursday, May 26, 2011 - link

    Hey I'm thinking of building up a multicore box for raw processing. I'm wondering if you could benchmark bibble on these supercore systems.
    Preferably with an A900 or a 5dmkII
    Also using the wavelet denoise and wavelet sharpening plugins as these are what I use most often.
    I'm wondering about import and preview speed and also speed to export as jpg.
    Let me know if its possible to do these benchmarks, also if you need source files and config sets I have some 8-16gig sets.
  • JohanAnandtech - Thursday, May 26, 2011 - link

    Questions: are you sure bibble can handle enough threads? And can you provide me with a benchmarkable scenario? mail me at Johan AT anandtech

Log in

Don't have an account? Sign up now