Slashdot Mirror
Benchmarking Power-Efficient Servers
modapi writes "According to the EPA, data centers — not including Google et al. — are on track to double power consumption in the next five years, to 3% of the US energy budget. That is a lot of expensive power. Can we cut the power requirement? We could, if we had a reliable way to benchmark power consumption across architectures. Which is what JouleSort: A Balanced Energy-Efficiency Benchmark (PDF), by a team from HP and Stanford, tries to do. StorageMojo summarizes the key findings of the paper and contrasts it with the recent Google paper, Power Provisioning for a Warehouse-sized Computer (PDF). The HP/Stanford authors use the benchmark to design a power-efficient server — with a mobile processor and lots of I/O — and to consider the role of software, RAM, and power supplies in power consumption."
I'm hoping the units are going to be kWh/slasdotting.
ccalam - acoustic versions of new songs.
Your basic point, which is that we need to consider not just operating costs also manufacturing and disposal costs, is a valid one.
However the way you've worded it amounts to "since we can't account for all aspects of impact, I'm not going to worry about any aspect of impact." That's a bit extreme. Surely reducing our power consumption during the operating lifetime of our servers is a step towards greater environmental and fiscal responsibility.
Now, if you can show that the "energy saving" chips generate more pollution during production than the "normal" chips (and that this increase in pollution/energy-use/cost is greater than the savings during the lifetime operation of the chip), that's important. However I doubt that is the case. Thus, to ignore the potential advantages of power-saving measures in the data-center, simply because such measures don't address the orthogonal concerns of production impact, is silly.
Answer me this : how much power is lost through the use of inefficient programming languages and architectures which only emphasize processor speed, instead of balancing memory, processor and IO ?
Python, Perl and PHP all suffer from one big drawback : when you scale up you need that much extra processor power. One programming language I know (Common Lisp) offers the advantages of them, but can be compiled to near C/C++ speeds. I suppose there are others. Don't come saying that programmers are expensive. It seems that what you gain on programmers, you lose in the cost of your datacenter. I don't know how Java matches here, it probably depends upon the deployment of more recent JIT compilers.
If you see how much a process has to wait on IO, how come there are still no good solutions in providing enough IO bandwidth that the processor can use fully ? (Unless you buy a mainframe or iSeries system that is)
Just asking.