Stats

  • Locked Locked
  • Replies 4
  • Subscribers 125
  • Views 19511
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

AMS simulator with APS and multi-thread

SteveRFIC

I am using the AMS simulator to run AMS sims.

In the "setup-high performance" menu, I choose APS. I enable ++aps and in the multi-threading part I choose auto.

Now I have a 4-core, 8-thread CPU and I am running simulation locally on my machine. 

I expect the simulator to use all the cpus for simulation. When I look at the sim log file, it says

"Multithreading Enabled: 8 threads in the system with 8 available processors."

However, I also monitored the CPU usage during the simulation, and it only uses 140% cpu, instead of 600-800%.

I also manually chose to use 4, or 8 threads (second screenshot below). It still just uses 140% cpu and the simulation time is the same.

The simulation takes one hour, I wonder why the simulator does not use all of the cpu to speed up the simulation.

How does it choose how many cores to use? what can I do to make it use the full capacity of the cpu?

Also, I noticed that if I set the multi-thread option to auto, then the simulator always use 8 threads, no matter how many cores I have. Is this true? WIll the simulator dynamically adjust the core/thread to use if I run it on different machines with different cpus?

Thanks!

Parents
  • SteveRFIC

    Thank you Andew!

    It is good to know that the limitation is gone now, so we can still choose more threads when resource is available. 

    What you explained about the parallel computing completely makes sense to me. Whether things can be computed in parallel depends on the circuit itself. Actually I was feeling the same way, but it is good to hear from the expert :) As a user who uses cadence everyday, I just want to make sure that I am using the tool properly and efficiently. Hearing from internal experts definitely helps :)

    By the way, I am using ic616_isr13. How can I get detailed information on the size of the circuit?

    Thanks!

Reply
  • SteveRFIC

    Thank you Andew!

    It is good to know that the limitation is gone now, so we can still choose more threads when resource is available. 

    What you explained about the parallel computing completely makes sense to me. Whether things can be computed in parallel depends on the circuit itself. Actually I was feeling the same way, but it is good to hear from the expert :) As a user who uses cadence everyday, I just want to make sure that I am using the tool properly and efficiently. Hearing from internal experts definitely helps :)

    By the way, I am using ic616_isr13. How can I get detailed information on the size of the circuit?

    Thanks!

Children
No Data

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.