From mcpatspot at gmail.com Mon Aug 19 09:03:54 2013 From: mcpatspot at gmail.com (Hotspot McPat) Date: Mon, 19 Aug 2013 14:03:54 +0100 Subject: [Hotspot] Getting the entire chip temperature Message-ID: Hi All, Anybody knows where I can get the entire chip temperature (just one number for the whole chip or core)? Many thanks Negar -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130819/60ce4d3e/attachment.html From mcpatspot at gmail.com Mon Aug 19 10:11:57 2013 From: mcpatspot at gmail.com (Hotspot McPat) Date: Mon, 19 Aug 2013 15:11:57 +0100 Subject: [Hotspot] Getting the entire chip temperature In-Reply-To: References: Message-ID: Hi Simone, Thanks for your prompt reply. No, I have the average temperatures for each block. I'm wondering if HotSpot is also calculating just one number for the entire chip temperature? Cheers Negar On Mon, Aug 19, 2013 at 2:45 PM, Simone Corbetta wrote: > Dear Negar, > > Do you mean average temperature? Well starting from the block-level > samples, it is straightforward to derive an average value weighting > temperature in each block to its area for instance. > > Cheers > > -- > "I want to remain an eternal mystery to myself and others" > Ludwig Otto Friedrich Wilhelm > > > On Mon, Aug 19, 2013 at 3:03 PM, Hotspot McPat wrote: > >> Hi All, >> >> Anybody knows where I can get the entire chip temperature (just one >> number for the whole chip or core)? >> >> Many thanks >> Negar >> >> _______________________________________________ >> HotSpot mailing list >> HotSpot at mail.cs.virginia.edu >> http://www.cs.virginia.edu/mailman/listinfo/hotspot >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130819/73a26947/attachment.html From skadron at cs.virginia.edu Mon Aug 19 21:52:51 2013 From: skadron at cs.virginia.edu (Kevin Skadron) Date: Mon, 19 Aug 2013 21:52:51 -0400 Subject: [Hotspot] Getting the entire chip temperature In-Reply-To: References:

Message-ID: <5212CBF3.8070709@cs.virginia.edu> HotSpot does not currently report a single average temperature, but as Simone noted, it's easy to compute. I would be careful of such a number, however, as it has very limited physical meaning. /K On 8/19/2013 10:11 AM, Hotspot McPat wrote: > Hi Simone, > > Thanks for your prompt reply. No, I have the average temperatures for > each block. I'm wondering if HotSpot is also calculating just one number > for the entire chip temperature? > > Cheers > Negar > > > On Mon, Aug 19, 2013 at 2:45 PM, Simone Corbetta > > wrote: > > Dear Negar, > Do you mean average temperature? Well starting from the block-level > samples, it is straightforward to derive an average value weighting > temperature in each block to its area for instance. > Cheers > > -- > "I want to remain an eternal mystery to myself and others" > Ludwig Otto Friedrich Wilhelm > > > On Mon, Aug 19, 2013 at 3:03 PM, Hotspot McPat > wrote: > > Hi All, > > Anybody knows where I can get the entire chip temperature (just > one number for the whole chip or core)? > > Many thanks > Negar > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot > > > > > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot > From yhu14 at tigers.lsu.edu Tue Aug 27 11:20:43 2013 From: yhu14 at tigers.lsu.edu (Yue Hu) Date: Tue, 27 Aug 2013 10:20:43 -0500 Subject: [Hotspot] Traditional Liquid Cooling Message-ID: Hi All, Can HotSpot simulate traditional liquid cooling, like backplate liquid cooling? It would be very helpful if someone could also provide some related references for simulation parameters. Regards, -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130827/a066d342/attachment.html From wh6p at virginia.edu Thu Aug 29 11:28:52 2013 From: wh6p at virginia.edu (Wei Huang) Date: Thu, 29 Aug 2013 10:28:52 -0500 Subject: [Hotspot] Traditional Liquid Cooling In-Reply-To: References: Message-ID: Hi, HotSpot is able to simulate traditional liquid cooling. If you know the flow property (flow speed, viscosity, thermal conductance etc), you should be able to calculate a heat transfer coefficient (h), and use 1/(h*A) to update the convection thermal resistance in the hot spot config file (where A is the surface area of your back plate). For calculation of h, most heat transfer textbook has the equations. Or you can refer to package.c file in the HotSpot release. A quick search in Wikipedia leads to this link: http://en.wikipedia.org/wiki/Heat_transfer_coefficient, where it mentions typical h value for water is 500 to 10,000 W/(m2K). You see it is a wide range, so it really depends on your flow property to narrow it down. Hope this is helpful. -Wei On Tue, Aug 27, 2013 at 10:20 AM, Yue Hu wrote: > Hi All, > > Can HotSpot simulate traditional liquid cooling, like backplate liquid > cooling? It would be very helpful if someone could also provide some > related references for simulation parameters. > > Regards, > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130829/2229f293/attachment.html