Why Intel Added Cache Partitioning

The SPECint and SPECfp benchmark suites were never very tightly connected to real-world performance, but I wonder what these figures look like for SPECjbb, the java benchmark.

(If I recall correctly, SPECjbb spins up an app that looks a lot like "java pet store" and runs clients against it.)

Re: Why Intel Added Cache Partitioning

#7

Post by uxcn » Mon, Oct 05, 2015, 10:00 PM UTC

Regarding cpusets, I wonder how much contention is being created just by software assuming the number of cpus match the number of cpus in the affinity set. For example, C++ has hardware_concurrency now, Python has cpu_count, Java has fork-join, etc...

Re: Why Intel Added Cache Partitioning

#8

Post by thrownaway2424 » Mon, Oct 05, 2015, 10:02 PM UTC

uxcn wrote: post #7

Regarding cpusets, I wonder how much contention is being created just by software assuming the number of cpus match the number of cpus in the affinity set. For example, C++ has hardware_concurrency now, Python has cpu_count , Java has fork-join, etc...

Golang also has a totally inaccurate routine for counting the CPUs in the machine.

Re: Why Intel Added Cache Partitioning

#9

Post by amelius » Mon, Oct 05, 2015, 10:13 PM UTC

> Xkcd estimated that Google owns 2 million machines

Xkcd doesn't sound like a very convincing source. And at least this estimation should have some error-bars.

Re: Why Intel Added Cache Partitioning

#10

Post by ajkjk » Mon, Oct 05, 2015, 10:21 PM UTC

amelius wrote: post #9

> Xkcd estimated that Google owns 2 million machines Xkcd doesn't sound like a very convincing source. And at least this estimation should have some error-bars.

Do you imagine that the error bars would be so large as to significantly change the order of magnitude? I don't.

Anyway, it's an offhanded comment and its precision is almost completely irrelevant to the article. I don't fault the author for not spending too much time on it.