1
$\begingroup$

In Chapter 5.7 of CSAPP, author list out the throughput boundary of Intel Core i7 Haswell while executing the operation Integer addition and multiplication, floating points addition and multiplication. Here is the function units list.enter image description here The author also pointed out the equation was throughput boundary = Capacity/Issue. Here is the data.enter image description here

I wonder how did the result equals to 0.5.enter image description here Because if the capacity is 2 and the issue is 1, it should be 2 right?

Improve this question
$\endgroup$
2
  • 1
    $\begingroup$ This question is not theoretical computer science, and it is off-topic on this site. Topic-wise, it seems it might belong to electrical engineering electronics.stackexchange.com , but I do not really know if this is a good question for that site. $\endgroup$
    – Emil Jeřábek
    Commented Apr 30, 2023 at 6:42
  • $\begingroup$ Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. $\endgroup$
    – Community Bot
    Commented May 5, 2023 at 10:51

1 Answer 1

Reset to default
0
$\begingroup$

You confused the definition of maximum throughput and throughput bound:

  • Maximum throughput = Capacity/Issue time
  • Throughput bound "gives a minimum bound for the CPE based on the maximum rate at which the functional units can produce results" (quoted from CSAPP). So throughput bound is the theoretical minimum CPE(Cycles per Element) caused by maximum throughput. Throughput bound = 1/maximum throughput.
Improve this answer
$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.