0
\$\begingroup\$

I have a question about the Mark Horowitz paper: Computing’s Energy Problem (and what we can do about it). In the paper, the author breaks down the sources of energy loss in modern computing systems.

What is the methodology they used to obtain these results? or could have used? Is it some sort of simulation?

\$\endgroup\$
2
  • \$\begingroup\$ That paper contains quite a lot of numbers and statements. Which premise ones are you wondering about? Presumably, the authors got most of their information from the sources that they cite; have you checked those? \$\endgroup\$
    – Sophie Swett
    Commented Jun 5, 2023 at 18:10
  • \$\begingroup\$ I have read some of the references, yes. However, to my knowledge, they do not sufficiently explain these numbers. For example, the CPU DB reference does contain a plethora of information on microprocessors, but not on systems. Thus DRAM energy is not included. Additionally, they quote energy at a macro level of a whole processor, not broken down by cache/control/operation. I do want to note however, that my background is higher level computer architecture, and I do have a hard time reading VLSI papers. \$\endgroup\$
    – Jure Vreča
    Commented Jun 6, 2023 at 9:46

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.