3 ms·

Read the article. Their conclusion is that 24 core power8 systems can beat 36 core Xeons in some important benchmarks, both in performance and power-per-dollar

Read the article. Their conclusion is that 24 core power8 systems can beat 36 core Xeons in some important benchmarks, both in performance and power-per-dollar (but not power-per-watt). If you want to see the reason, just look at the memory bandwidth - Power8 has about double that of the latest and greatest Xeon. That's very important, especially for fixed grid computational tasks (e.g. weather, climate) that traditionally have a very low arithmetic intensity, i.e. what really counts is the memory bandwidth.

throwaway2048 12y ago
also the code tends to be highly SIMD, code with a lot of data dependent branches benefits very little from higher memory bandwidth when the fetch latency from ram is on the order of hundreds or even thousands of cycles.
- m_mueller 12y ago
  Yes that's a good point. So low arithmetic intensity + SIMD benefits POWER mostly.
dragontamer 12y ago
I did read the article. Power-per-dollar are the smaller 2S Xeons btw. Not the massive 36-core Xeons. However, the memory-bandwidth argument makes sense to me. So I'll accept that as a good reason to go Power8.
- m_mueller 11y ago
  > Power-per-dollar are the smaller 2S Xeons btw. Not the massive 36-core Xeons. Could you give a quote? All I could find about performance-per-dollar was this excerpt from [1]: > Ultimately, the POWER8 is able to offer slightly higher raw performance than the Intel CPUs, however it just won't be able to do so at the same performance/watt. Meanwhile the reasonable pricing of the POWER8 chips should result in third party servers that are strongly competitive with the Xeon on a performance-per-dollar basis. This was in conclusion to the E5-vs-Power8 comparison. For some reason they didn't use E7, however this shouldn't change the outcome, since according to [2] the E7 is basically the same chip with the same number of cores, just with a few more features enabled (4 and 8 socket systems possible and support for more memory). So AFAIK the E5 shouldn't be looked at as a "small" version of the E7, since taken for itself in isolation for a given benchmark that doesn't have extremely high memory demand, it should deliver the same performance. Whether you need an E7 or not should be decided in terms of the architecture you want to build, not the performance you aim for. [1]http://anandtech.com/show/9193/the-xeon-e78800-v3-review/7 http://anandtech.com/show/9193/the-xeon-e78800-v3-review/7 [2]http://anandtech.com/show/9193/the-xeon-e78800-v3-review/3 http://anandtech.com/show/9193/the-xeon-e78800-v3-review/3

throwaway2048 12y ago

also the code tends to be highly SIMD, code with a lot of data dependent branches benefits very little from higher memory bandwidth when the fetch latency from ram is on the order of hundreds or even thousands of cycles.

m_mueller 12y ago
Yes that's a good point. So low arithmetic intensity + SIMD benefits POWER mostly.

dragontamer 12y ago

I did read the article. Power-per-dollar are the smaller 2S Xeons btw. Not the massive 36-core Xeons. However, the memory-bandwidth argument makes sense to me. So I'll accept that as a good reason to go Power8.

m_mueller 11y ago
> Power-per-dollar are the smaller 2S Xeons btw. Not the massive 36-core Xeons. Could you give a quote? All I could find about performance-per-dollar was this excerpt from [1]: > Ultimately, the POWER8 is able to offer slightly higher raw performance than the Intel CPUs, however it just won't be able to do so at the same performance/watt. Meanwhile the reasonable pricing of the POWER8 chips should result in third party servers that are strongly competitive with the Xeon on a performance-per-dollar basis. This was in conclusion to the E5-vs-Power8 comparison. For some reason they didn't use E7, however this shouldn't change the outcome, since according to [2] the E7 is basically the same chip with the same number of cores, just with a few more features enabled (4 and 8 socket systems possible and support for more memory). So AFAIK the E5 shouldn't be looked at as a "small" version of the E7, since taken for itself in isolation for a given benchmark that doesn't have extremely high memory demand, it should deliver the same performance. Whether you need an E7 or not should be decided in terms of the architecture you want to build, not the performance you aim for. [1]http://anandtech.com/show/9193/the-xeon-e78800-v3-review/7 http://anandtech.com/show/9193/the-xeon-e78800-v3-review/7 [2]http://anandtech.com/show/9193/the-xeon-e78800-v3-review/3 http://anandtech.com/show/9193/the-xeon-e78800-v3-review/3