Dear Hang Lu, The hardware requirements for Flexus jobs depend on the type of jobs you are running and the workload. If you are running timing simulations with sampling, a single job requires no more than 1GB of RAM per job and one core or hardware thread (if hyperthreading should be enabled for better resource utilization). These jobs are short. For trace simulations that you run to create flexpoints, the hardware requirements are similar, because you will write frequent checkpoints and flush the memory. However, if you run long trace simulations for profiling purposes, you might need several gigabytes per job, which primarily depends on the length of your simulations, but you typically have very few of these jobs (in contrast, you typically run 100s of timing jobs at a time).
Regarding 2TB: the total size of all CloudSuite images is around 1TB. You do need a storage system (disk) that can fit 1TB of data, but you are not expected to have this data in memory at all. Regards, Djordje ________________________________________ From: Lu Hang [[email protected]] Sent: Friday, October 10, 2014 7:25 AM To: cloudsuite-admin Cc: cloudsuite Subject: Flexus plus Cloudsuite image simulation requirements Dear all, I have another question. We are now setting up the Cloudsuite images with flexus on our server for simulation. Do you guys know how much physical memory and CPU usage it will cost during simulation runtime at most? I mean, the server has many other irrelevant users. If running cloudsuite will load 2T image data to the memory, then the server cannot put up with it. Please also inform me the lowest physical resources requirement when running flexus + image simulation. Thanks in advance. ________________________________ Thanks and best regards! ------------------------------------------------------------- Hang Lu Phd Candidate State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China. -------------------------------------------------------------
