Hi all, Here is a thesis work conducted at NTNU about power estimation based on stat file generated by Gem5. http://www.diva-portal.org/smash/get/diva2:751714/FULLTEXT01.pdf
Source code is available at https://github.com/terjr/thesis.git I have not use this tool, but it may prove helpful for you. > Hi all, > > I merely wanted to caution you guys before you get too excited. Even one > of these power models, McPat, Orion, DSENT, has hundreds if not thousands > of undocumented and hidden assumptions. This makes it very challenging to > use even a single one of these tools with any confidence what so ever > (yes, even relative numbers). As they are bottom up tools, the assumptions > are fundamental in the numbers they eventually generate, and accumulate to > error bars measured in 100s or even 1000s of percent. > > When you combine them as you suggest you make this even worse. The models > are with all certainty not composable. Most likely there is no alignment > of assumptions between the tools, and a number generated by one tools > means even less in comparison to the numbers generated by another tool. > I’m not trying to dissuade you from using energy models, but be very > cautious and critical when it comes to these bottom up models. > > There are some very promising top-down regressions models out there that > probably are far less risky. In any case, keep the user group updated with > this work as there currently is not too much visibility of how people use > power/energy/thermal models with gem5. > > Andreas > > From: Matheus Alcântara Souza via gem5-users > <gem5-users@gem5.org<mailto:gem5-users@gem5.org>> > Reply-To: Matheus Alcântara Souza > <ticks...@gmail.com<mailto:ticks...@gmail.com>>, gem5 users mailing list > <gem5-users@gem5.org<mailto:gem5-users@gem5.org>> > Date: Tuesday, 25 November 2014 08:39 > To: Sahar Pilevar <saharpile...@yahoo.com<mailto:saharpile...@yahoo.com>> > Cc: gem5 users mailing list > <gem5-users@gem5.org<mailto:gem5-users@gem5.org>> > Subject: Re: [gem5-users] Energy and Power > > Hi, > > Not yet. The tool I should check is DSENT. > > Best, > > Matheus Alcântara Souza > > > Em 25/11/2014, às 05:51, Sahar Pilevar > <saharpile...@yahoo.com<mailto:saharpile...@yahoo.com>> escreveu: > > Hello, I have this question too. Do you find any answer for your question > ,Matheus ? > > > On Saturday, November 22, 2014 9:43 PM, Matheus Alcântara Souza via > gem5-users <gem5-users@gem5.org<mailto:gem5-users@gem5.org>> wrote: > > > Hi everyone! > > I would like to know how to get energy/power results from gem5 using Ruby > Mem System and Garnet. > > I already know McPAT, that was useful when using Classic Mem System, but > with Garnet, I don't know how to use it. > > Is Orion still working? > > Have a nice weekend! > Matheus > _______________________________________________ > gem5-users mailing list > gem5-users@gem5.org<mailto:gem5-users@gem5.org> > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > > > > -- IMPORTANT NOTICE: The contents of this email and any attachments are > confidential and may also be privileged. If you are not the intended > recipient, please notify the sender immediately and do not disclose the > contents to any other person, use it for any purpose, or store or copy the > information in any medium. Thank you. > > ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, > Registered in England & Wales, Company No: 2557590 > ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, > Registered in England & Wales, Company No: > 2548782_______________________________________________ > gem5-users mailing list > gem5-users@gem5.org > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users Naveed Ul Mustafa _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
