From wh6p at cms.mail.virginia.edu  Tue Jul  1 05:28:23 2008
From: wh6p at cms.mail.virginia.edu (Wei Huang)
Date: Tue, 01 Jul 2008 08:28:23 -0400
Subject: [Hotspot] unexpected blocks temperatures
In-Reply-To: <554934.53575.qm@web53705.mail.re2.yahoo.com>
References: <554934.53575.qm@web53705.mail.re2.yahoo.com>
Message-ID: <EDE6DAAD66EB9AA4C15B3D4C@[192.168.1.102]>

Hi Mohamed,

>From your plot, it seems the temperature difference between 
cores are not significant. I can several reasons to the 
differences:
- you floorplan is not entirely symmetric for each 
core/unit.
- There are white spaces at the edge of the chip. Since 
your cores aren't exactly square, these white spaces are 
also asymmetry.
- Earlier versions of HotSpot have a thermal interface 
material thickness of 75um, which is higher than the 
state-of-the-art. Using HS4.1 with thinner TIM will reach 
less temperature difference across die.

Hope this helps.
-Wei

--On Monday, June 30, 2008 11:31 PM -0700 Mohamed Elsawaf 
<mohamed_elsawaf at yahoo.com> wrote:

> Hi Wei
>
> Thank you for your help, i will try Hostspot 4.1
> meanwhile please find the simulation temperatures
> distribution output [per block per core] and my floorplan
> reassembling to IBM power4 MCM [ 8 cores]
> i scale it to 45 nm technology
> i can see that cores C0 and C6 having higher temperatures
> at steady state than the other cores
> while all cores having the same power consumed in
> gcc.ptrace file also i don't have any explanation
> regarding this unexpected higher temperature
>
> thank you
> Mohamed Elsawaf
>
> --- On Fri, 6/27/08, Wei Huang
> <wh6p at cms.mail.virginia.edu> wrote:
>
> From: Wei Huang <wh6p at cms.mail.virginia.edu>
> Subject: Re: [Hotspot] unexpected blocks temperatures
> To: mohamed_elsawaf at yahoo.com
> Cc: "Karthik Sankaranarayanan" <ks4kk at cs.virginia.edu>
> Date: Friday, June 27, 2008, 12:49 PM
>
>
> Hi Mohamed,  As you can see, your chip size is 2.26cm on
> each side,  which is smaller than the sizes of spreader
> (3cm) and heat  sink (12cm). So I don't think your
> problem is caused by the  size.  How much higher the
> temperatures are than what you  expected? From your
> configuration file, it seems either you  have changed it
> (which is OK) or you are using an old  version of
> HotSpot. The newest version is 4.1, which fixes  some
> bugs in previous versions and includes a faster grid
> model solver. So I would recommend you to download v4.1
> and  try it again.  -Wei  --On Friday, June 27, 2008
> 11:07 AM -0700 Mohamed Elsawaf
> <mohamed_elsawaf at yahoo.com> wrote:  > Hi Wei > > Thank
> you for your fast response > > My chip area = 0.000514234
> m^2 > ># chip specs ># chip
>  thickness in meters > -t_chip 0.0005 > ># heat sink
> specs ># convection capacitance in J/K > -c_convec 140.4
> ># convection resistance in K/W > -r_convec 0.1 >#
> heatsink side in meters > -s_sink 0.12 ># heatsink
> thickness in meters > -t_sink 0.0069 > ># heat spreader
> specs ># spreader side in meters > -s_spreader 0.03 >#
> spreader thickness in meters > -t_spreader 0.001 > >#
> interface material specs ># interface material thickness
> in meters > -t_interface 0.000075 > > waiting for your
> feedback > > Thank you > > Mohamed Elsawaf > > > --- On
> Fri, 6/27/08, Wei Huang > <wh6p at cms.mail.virginia.edu>
> wrote: > > From: Wei Huang <wh6p at cms.mail.virginia.edu> >
> Subject: Re: [Hotspot] unexpected blocks temperatures >
> To: mohamed_elsawaf at yahoo.com > Cc: "Karthik
> Sankaranarayanan" <ks4kk at cs.virginia.edu>, >
> hotspot at mail.cs.virginia.edu >
>  Date: Friday, June 27, 2008, 8:32 AM > > > Hi Mohamed,
> HotSpot automatically aligns chip center > with heat
> sink/spreader centers. What is the size of > your chip?
> You  may want to check if it is larger than > the default
> sizes of  heat spreader and heat sink in > HotSpot. If
> that is the  case, increasing the default > values to
> proper ones should  solve your problem.  -Wei > --On
> Friday, June 27, 2008 4:15 AM -0700 Mohamed Elsawaf >
> <mohamed_elsawaf at yahoo.com> wrote:  > Hi > > can you >
> please help me? > how can i check that the center of my >
> floor-plan is on > the same access as >  heat-sink
> center? > I?am getting unexpected blocks >
> temperatures.Iam > suspecting that chip floor-plan is >
> shifted to the left of > from heat-sink center. > > Thank
> > you > Mohamed Elsawaf > > > > > >
>
>
>







From eby_buendia at hotmail.com  Tue Jul  1 16:21:53 2008
From: eby_buendia at hotmail.com (Encarna Buendia)
Date: Tue, 1 Jul 2008 16:21:53 -0700
Subject: [Hotspot] significant difference on thermal plot v3.1 vs v4.1
Message-ID: <BLU124-W24A62E628E415D7F1652409E9E0@phx.gbl>


Hi,

I am trying to migrate from v3.1 to v4.1
Attached are two thermal plots. The power trace is exactly the same in both cases, but one is obtained with v3.1 and the other with v4.1.
The v3.1 one should be the correct one. The v4.1 one has this very suspiciously round hotspot in the middle of the die. The only difference in the commands to the hotspot tool are (first is v4.1):

6,7c6,7
< -t_chip       0.00015
< -thermal_threshold    354.95
---
> -t_chip       0.0005
> -thermal_threshold    384.95
14,15c14,15
< -t_interface  2e-05
< -ambient      318.15
---
> -t_interface  7.5e-05
> -ambient      313.15
20c20
< -base_proc_freq       3e+09
---
> -base_proc_freq       5e+08


Any idea why the huge differenence in the thermal plot?
The diff in the grid_thermal_map.pl script I use to generate the plot is:

185c190
<               $txt_start_y=$start_y+7*$txt_offset;
---
>               $txt_start_y=$start_y+3*$txt_offset;
204c209
<                       '" fill="black" text_anchor="start" style="font-size:180"> '. @{$_}[0] .' '."\n";
---
>                       '" fill="black" text_anchor="start" style="font-size:300"> '. @{$_}[0] .' '."\n";
234c239
<                               '" fill="black" text_anchor="start" style="font-size:250"> '. $scale_value .' '."\n";
---
>                               '" fill="black" text_anchor="start" style="font-size:300"> '. $scale_value .' '."\n";
241c246
<       '" fill="black" text_anchor="start" style="font-size:250"> '. $min_t .' '."\n";
---
>       '" fill="black" text_anchor="start" style="font-size:300"> '. $min_t .' '."\n";

_________________________________________________________________
The i?m Talkaton. Can 30-days of conversation change the world?
http://www.imtalkathon.com/?source=EML_WLH_Talkathon_ChangeWorld
-------------- next part --------------
A non-text attachment was scrubbed...
Name: v31.pdf
Type: application/pdf
Size: 1109228 bytes
Desc: not available
Url : http://www.cs.virginia.edu/pipermail/hotspot/attachments/20080701/954ccde5/attachment-0002.pdf 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: v41.pdf
Type: application/pdf
Size: 1106229 bytes
Desc: not available
Url : http://www.cs.virginia.edu/pipermail/hotspot/attachments/20080701/954ccde5/attachment-0003.pdf 

From ks4kk at cs.virginia.edu  Tue Jul  1 23:08:11 2008
From: ks4kk at cs.virginia.edu (Karthik Sankaranarayanan)
Date: Wed, 2 Jul 2008 02:08:11 -0400 (EDT)
Subject: [Hotspot] significant difference on thermal plot v3.1 vs v4.1
In-Reply-To: <BLU124-W24A62E628E415D7F1652409E9E0@phx.gbl>
References: <BLU124-W24A62E628E415D7F1652409E9E0@phx.gbl>
Message-ID: <Pine.SOC.4.64.0807020200130.20099@mamba.cs.Virginia.EDU>


I am not sure this a HotSpot issue. It appears to be an error in usage to 
me. The default grid resolution is different between v3.1 and 4.1 (50x50 
vs. 64x64). Please check to see if you are specifying the correct grid 
resolution to grid_thermal_map.pl (type grid_thermal_map.pl without any 
arguments to see the various command line options)

Thanks,
-karthik


On Tue, 1 Jul 2008, Encarna Buendia wrote:

>
> Hi,
>
> I am trying to migrate from v3.1 to v4.1 Attached are two thermal plots. 
> The power trace is exactly the same in both cases, but one is obtained 
> with v3.1 and the other with v4.1. The v3.1 one should be the correct 
> one. The v4.1 one has this very suspiciously round hotspot in the middle 
> of the die. The only difference in the commands to the hotspot tool are 
> (first is v4.1):
>
> 6,7c6,7
> < -t_chip       0.00015
> < -thermal_threshold    354.95
> ---
>> -t_chip       0.0005
>> -thermal_threshold    384.95
> 14,15c14,15
> < -t_interface  2e-05
> < -ambient      318.15
> ---
>> -t_interface  7.5e-05
>> -ambient      313.15
> 20c20
> < -base_proc_freq       3e+09
> ---
>> -base_proc_freq       5e+08
>
>
> Any idea why the huge differenence in the thermal plot?
> The diff in the grid_thermal_map.pl script I use to generate the plot is:
>
> 185c190
> <               $txt_start_y=$start_y+7*$txt_offset;
> ---
>>               $txt_start_y=$start_y+3*$txt_offset;
> 204c209
> <                       '" fill="black" text_anchor="start" style="font-size:180"> '. @{$_}[0] .' '."\n";
> ---
>>                       '" fill="black" text_anchor="start" style="font-size:300"> '. @{$_}[0] .' '."\n";
> 234c239
> <                               '" fill="black" text_anchor="start" style="font-size:250"> '. $scale_value .' '."\n";
> ---
>>                               '" fill="black" text_anchor="start" style="font-size:300"> '. $scale_value .' '."\n";
> 241c246
> <       '" fill="black" text_anchor="start" style="font-size:250"> '. $min_t .' '."\n";
> ---
>>       '" fill="black" text_anchor="start" style="font-size:300"> '. $min_t .' '."\n";
>
> _________________________________________________________________
> The i?m Talkaton. Can 30-days of conversation change the world?
> http://www.imtalkathon.com/?source=EML_WLH_Talkathon_ChangeWorld

From eby_buendia at hotmail.com  Wed Jul  2 06:59:07 2008
From: eby_buendia at hotmail.com (Encarna Buendia)
Date: Wed, 2 Jul 2008 06:59:07 -0700
Subject: [Hotspot] significant difference on thermal plot v3.1 vs v4.1
In-Reply-To: <Pine.SOC.4.64.0807020200130.20099@mamba.cs.Virginia.EDU>
References: <BLU124-W24A62E628E415D7F1652409E9E0@phx.gbl>
	<Pine.SOC.4.64.0807020200130.20099@mamba.cs.Virginia.EDU>
Message-ID: <BLU124-W38449BB0AEFCE25ECDEF129E990@phx.gbl>


Hi Karthik,

the only differences in grid_thermal_map.pl that I use in 4.1 with respect the original release in 4.1 are there:

179c179
<               if ($unit_cnt < 1024) {
---
>               if ($unit_cnt < 121) {
185c185
<               $txt_start_y=$start_y+7*$txt_offset;
---
>               $txt_start_y=$start_y+2*$txt_offset;

The grid is 64x64 in both grid_thermal_map.pl and hotspot.config

thanks.


----------------------------------------
> Date: Wed, 2 Jul 2008 02:08:11 -0400
> From: ks4kk at cs.virginia.edu
> To: eby_buendia at hotmail.com
> CC: hotspot at mail.cs.virginia.edu
> Subject: Re: [Hotspot] significant difference on thermal plot v3.1 vs v4.1
> 
> 
> I am not sure this a HotSpot issue. It appears to be an error in usage to 
> me. The default grid resolution is different between v3.1 and 4.1 (50x50 
> vs. 64x64). Please check to see if you are specifying the correct grid 
> resolution to grid_thermal_map.pl (type grid_thermal_map.pl without any 
> arguments to see the various command line options)
> 
> Thanks,
> -karthik
> 
> 
> On Tue, 1 Jul 2008, Encarna Buendia wrote:
> 
> >
> > Hi,
> >
> > I am trying to migrate from v3.1 to v4.1 Attached are two thermal plots. 
> > The power trace is exactly the same in both cases, but one is obtained 
> > with v3.1 and the other with v4.1. The v3.1 one should be the correct 
> > one. The v4.1 one has this very suspiciously round hotspot in the middle 
> > of the die. The only difference in the commands to the hotspot tool are 
> > (first is v4.1):
> >
> > 6,7c6,7
> > < -t_chip       0.00015
> > < -thermal_threshold    354.95
> > ---
> >> -t_chip       0.0005
> >> -thermal_threshold    384.95
> > 14,15c14,15
> > < -t_interface  2e-05
> > < -ambient      318.15
> > ---
> >> -t_interface  7.5e-05
> >> -ambient      313.15
> > 20c20
> > < -base_proc_freq       3e+09
> > ---
> >> -base_proc_freq       5e+08
> >
> >
> > Any idea why the huge differenence in the thermal plot?
> > The diff in the grid_thermal_map.pl script I use to generate the plot is:
> >
> > 185c190
> > <               $txt_start_y=$start_y+7*$txt_offset;
> > ---
> >>               $txt_start_y=$start_y+3*$txt_offset;
> > 204c209
> > <                       '" fill="black" text_anchor="start" style="font-size:180"> '. @{$_}[0] .' '."\n";
> > ---
> >>                       '" fill="black" text_anchor="start" style="font-size:300"> '. @{$_}[0] .' '."\n";
> > 234c239
> > <                               '" fill="black" text_anchor="start" style="font-size:250"> '. $scale_value .' '."\n";
> > ---
> >>                               '" fill="black" text_anchor="start" style="font-size:300"> '. $scale_value .' '."\n";
> > 241c246
> > <       '" fill="black" text_anchor="start" style="font-size:250"> '. $min_t .' '."\n";
> > ---
> >>       '" fill="black" text_anchor="start" style="font-size:300"> '. $min_t .' '."\n";
> >
> > _________________________________________________________________
> > The i?m Talkaton. Can 30-days of conversation change the world?
> > http://www.imtalkathon.com/?source=EML_WLH_Talkathon_ChangeWorld

_________________________________________________________________
It?s a talkathon ? but it?s not just talk.
http://www.imtalkathon.com/?source=EML_WLH_Talkathon_JustTalk

From wh6p at cms.mail.virginia.edu  Wed Jul  2 08:08:08 2008
From: wh6p at cms.mail.virginia.edu (Wei Huang)
Date: Wed, 02 Jul 2008 11:08:08 -0400
Subject: [Hotspot] significant difference on thermal plot v3.1 vs v4.1
In-Reply-To: <BLU124-W38449BB0AEFCE25ECDEF129E990@phx.gbl>
References: <BLU124-W24A62E628E415D7F1652409E9E0@phx.gbl>
	<Pine.SOC.4.64.0807020200130.20099@mamba.cs.Virginia.EDU>
	<BLU124-W38449BB0AEFCE25ECDEF129E990@phx.gbl>
Message-ID: <8E0295992813565D55B702E1@mstu2>

By looking at the temperature index sidebars, it seems that 
your power distribution is quite uniform across the die, 
because the temperature diff is only ~1 degree. Is that the 
case? Can you also tell us whether your power map is 
symmetrical w.r.t to the chip center? The scattered 
"hotspots" in the v3.1 plot seem to be numerical 
oscillations, which we managed to improve in v4.1. So, if 
your power map is actually uniform, the plot of v4.1 should 
be the more "correct" one.

Thanks,
-Wei

--On Wednesday, July 02, 2008 6:59 AM -0700 Encarna Buendia 
<eby_buendia at hotmail.com> wrote:

>
> Hi Karthik,
>
> the only differences in grid_thermal_map.pl that I use in
> 4.1 with respect the original release in 4.1 are there:
>
> 179c179
> <               if ($unit_cnt < 1024) {
> ---
>>               if ($unit_cnt < 121) {
> 185c185
> <               $txt_start_y=$start_y+7*$txt_offset;
> ---
>>               $txt_start_y=$start_y+2*$txt_offset;
>
> The grid is 64x64 in both grid_thermal_map.pl and
> hotspot.config
>
> thanks.
>
>
> ----------------------------------------
>> Date: Wed, 2 Jul 2008 02:08:11 -0400
>> From: ks4kk at cs.virginia.edu
>> To: eby_buendia at hotmail.com
>> CC: hotspot at mail.cs.virginia.edu
>> Subject: Re: [Hotspot] significant difference on thermal
>> plot v3.1 vs v4.1
>>
>>
>> I am not sure this a HotSpot issue. It appears to be an
>> error in usage to  me. The default grid resolution is
>> different between v3.1 and 4.1 (50x50  vs. 64x64).
>> Please check to see if you are specifying the correct
>> grid  resolution to grid_thermal_map.pl (type
>> grid_thermal_map.pl without any  arguments to see the
>> various command line options)
>>
>> Thanks,
>> -karthik
>>
>>
>> On Tue, 1 Jul 2008, Encarna Buendia wrote:
>>
>> >
>> > Hi,
>> >
>> > I am trying to migrate from v3.1 to v4.1 Attached are
>> > two thermal plots.  The power trace is exactly the
>> > same in both cases, but one is obtained  with v3.1 and
>> > the other with v4.1. The v3.1 one should be the
>> > correct  one. The v4.1 one has this very suspiciously
>> > round hotspot in the middle  of the die. The only
>> > difference in the commands to the hotspot tool are
>> > (first is v4.1):
>> >
>> > 6,7c6,7
>> > < -t_chip       0.00015
>> > < -thermal_threshold    354.95
>> > ---
>> >> -t_chip       0.0005
>> >> -thermal_threshold    384.95
>> > 14,15c14,15
>> > < -t_interface  2e-05
>> > < -ambient      318.15
>> > ---
>> >> -t_interface  7.5e-05
>> >> -ambient      313.15
>> > 20c20
>> > < -base_proc_freq       3e+09
>> > ---
>> >> -base_proc_freq       5e+08
>> >
>> >
>> > Any idea why the huge differenence in the thermal plot?
>> > The diff in the grid_thermal_map.pl script I use to
>> > generate the plot is:
>> >
>> > 185c190
>> > <               $txt_start_y=$start_y+7*$txt_offset;
>> > ---
>> >>               $txt_start_y=$start_y+3*$txt_offset;
>> > 204c209
>> > <                       '" fill="black"
>> > text_anchor="start" style="font-size:180"> '. @{$_}[0]
>> > .' '."\n"; ---
>> >>                       '" fill="black"
>> >>                       text_anchor="start"
>> >>                       style="font-size:300"> '.
>> >>                       @{$_}[0] .' '."\n";
>> > 234c239
>> > <                               '" fill="black"
>> > text_anchor="start" style="font-size:250"> '.
>> > $scale_value .' '."\n"; ---
>> >>                               '" fill="black"
>> >>                               text_anchor="start"
>> >>                               style="font-size:300">
>> >>                               '. $scale_value .'
>> >>                               '."\n";
>> > 241c246
>> > <       '" fill="black" text_anchor="start"
>> > style="font-size:250"> '. $min_t .' '."\n"; ---
>> >>       '" fill="black" text_anchor="start"
>> >>       style="font-size:300"> '. $min_t .' '."\n";
>> >
>> > ______________________________________________________
>> > ___________ The i?m Talkaton. Can 30-days of
>> > conversation change the world?
>> > http://www.imtalkathon.com/?source=EML_WLH_Talkathon_C
>> > hangeWorld
>
> _________________________________________________________
> ________ It?s a talkathon ? but it?s not just talk.
> http://www.imtalkathon.com/?source=EML_WLH_Talkathon_Just
> Talk _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot







From sgeorge at coe.neu.edu  Tue Jul 15 11:47:51 2008
From: sgeorge at coe.neu.edu (Sarmad George)
Date: Tue, 15 Jul 2008 14:47:51 -0400
Subject: [Hotspot] Queries
Message-ID: <1216147671.487cf0d75eb4b@www.coe.neu.edu>

Dear all,
I would like to inquire about the hotspot 4.1 - please find the attached
inquiries on the doc file.
The questions are on the 3rd page

Thanks alot,
All the best,

Sarmad Edward George
Electrical Computer Eng
Noetheastern Uni
Boston MA - USA
-------------- next part --------------
A non-text attachment was scrubbed...
Name: HotSpot 4.doc
Type: application/msword
Size: 31232 bytes
Desc: not available
Url : http://www.cs.virginia.edu/pipermail/hotspot/attachments/20080715/06b91767/attachment-0001.doc 

From wh6p at cms.mail.virginia.edu  Tue Jul 15 12:06:33 2008
From: wh6p at cms.mail.virginia.edu (Wei Huang)
Date: Tue, 15 Jul 2008 15:06:33 -0400
Subject: [Hotspot] answers to your questions
In-Reply-To: <mailman.905.1216147726.502.hotspot@mail.cs.virginia.edu>
References: <mailman.905.1216147726.502.hotspot@mail.cs.virg
 inia.edu>
Message-ID: <7AF110A384666591EE5F46D0@mstu2>

Hi Sarmad,

What you drew was correct. The equations are used to 
calculate the dimensions of the for peripheral parts of the 
spreader/sink. In HotSpot, we assume the spreader and sink 
are square (rectangular shapes don't make much difference) 
and the chip can be rectangular.
For more details, you can read our papers that are in the 
HotSpot webpage.

-Wei



From ss_majzoub at yahoo.com  Tue Jul 15 17:57:41 2008
From: ss_majzoub at yahoo.com (Sohaib Majzoub)
Date: Tue, 15 Jul 2008 17:57:41 -0700 (PDT)
Subject: [Hotspot] first timer questons
Message-ID: <666178.20317.qm@web50206.mail.re2.yahoo.com>

Hi Everyone,
I am trying to use HotSpot in my work. Please, where can I find detailed info about the file hotspot.config?
thanks
Sohaib.
www.ece.ubc.ca/~sohaibm 



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20080715/5683ae63/attachment.html 

From ks4kk at cs.virginia.edu  Tue Jul 15 23:07:45 2008
From: ks4kk at cs.virginia.edu (Karthik Sankaranarayanan)
Date: Wed, 16 Jul 2008 02:07:45 -0400 (EDT)
Subject: [Hotspot] first timer questons
In-Reply-To: <666178.20317.qm@web50206.mail.re2.yahoo.com>
References: <666178.20317.qm@web50206.mail.re2.yahoo.com>
Message-ID: <Pine.SOC.4.64.0807160203001.23034@cobra.cs.Virginia.EDU>


Hello Sohaib,

To start with, please take a look at

http://lava.cs.virginia.edu/HotSpot/HotSpot-HOWTO.htm

For more technical detail, please take a look at

http://www.cs.virginia.edu/~skadron/Papers/hotspot_tr2003_08.pdf

-karthik


On Tue, 15 Jul 2008, Sohaib Majzoub wrote:

> Hi Everyone,
> I am trying to use HotSpot in my work. Please, where can I find detailed info about the file hotspot.config?
> thanks
> Sohaib.
> www.ece.ubc.ca/~sohaibm
>
>
>
>

From wh6p at cms.mail.virginia.edu  Wed Jul 16 06:57:12 2008
From: wh6p at cms.mail.virginia.edu (Wei Huang)
Date: Wed, 16 Jul 2008 09:57:12 -0400
Subject: [Hotspot] Queries
In-Reply-To: <1216147671.487cf0d75eb4b@www.coe.neu.edu>
References: <1216147671.487cf0d75eb4b@www.coe.neu.edu>
Message-ID: <C8D0C3333AE9A2BA19AC3D01@mstu2>

Hi Sarmad,

Just make it more clear-- if you want to know more about 
how to get HotSpot running and the version history of 
HotSpot, please go to
<http://lava.cs.virginia.edu/HotSpot/HotSpot-HOWTO.htm>

if you want to know the details how the model is 
constructed and how the equations are derived, please 
browse through the papers listed on the HotSpot 
documentation webpage:
<http://lava.cs.virginia.edu/HotSpot/documentation.htm>

As usual, it is always a good practice to look through the 
FAQ's and the archive of the mailing list for answers to 
your questions. There is a good chance somebody already 
asked them.
<http://lava.cs.virginia.edu/HotSpot/faq.htm>

Hope this is helpful.

-Wei



--On Tuesday, July 15, 2008 2:47 PM -0400 Sarmad George 
<sgeorge at coe.neu.edu> wrote:

> Dear all,
> I would like to inquire about the hotspot 4.1 - please
> find the attached inquiries on the doc file.
> The questions are on the 3rd page
>
> Thanks alot,
> All the best,
>
> Sarmad Edward George
> Electrical Computer Eng
> Noetheastern Uni
> Boston MA - USA






From skadron at cs.virginia.edu  Wed Jul 16 07:02:16 2008
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Wed, 16 Jul 2008 10:02:16 -0400
Subject: [Hotspot] first timer questons
In-Reply-To: <Pine.SOC.4.64.0807160203001.23034@cobra.cs.Virginia.EDU>
References: <666178.20317.qm@web50206.mail.re2.yahoo.com>
	<Pine.SOC.4.64.0807160203001.23034@cobra.cs.Virginia.EDU>
Message-ID: <487DFF68.1080108@cs.virginia.edu>

PS  If you find topics that the documentation is not addressing, such as 
what certain aspects of the config file are for, let us know so that we 
can improve the documentation!
/K

Karthik Sankaranarayanan wrote:
> Hello Sohaib,
> 
> To start with, please take a look at
> 
> http://lava.cs.virginia.edu/HotSpot/HotSpot-HOWTO.htm
> 
> For more technical detail, please take a look at
> 
> http://www.cs.virginia.edu/~skadron/Papers/hotspot_tr2003_08.pdf
> 
> -karthik
> 
> 
> On Tue, 15 Jul 2008, Sohaib Majzoub wrote:
> 
>> Hi Everyone,
>> I am trying to use HotSpot in my work. Please, where can I find detailed info about the file hotspot.config?
>> thanks
>> Sohaib.
>> www.ece.ubc.ca/~sohaibm
>>
>>
>>
>>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot


From sgeorge at coe.neu.edu  Tue Jul 22 07:44:08 2008
From: sgeorge at coe.neu.edu (Sarmad George)
Date: Tue, 22 Jul 2008 10:44:08 -0400
Subject: [Hotspot] Multiprocessor Power Trace
Message-ID: <1216737848.4885f238eec06@www.coe.neu.edu>

Dear all,
I would like to use the Hotspot 4.1 to trace the temperature of GPGPUs (e.g.
NVIDIA CUDA) , but still I need a recommendation for the best software to trace
the power consumption per each unit in that card (with its associated
processors - memories - texture & constant caches, etc)

Or in other words - a software that when executed (to emulate the NVIDIA CUDA) -
outputs a power trace file as gcc.ptrace

Please can you advice?
Thanks

Sarmad Edward George
Electrical Computer Eng
Noetheastern Uni
Boston MA - USA




From sohaib.majzoub at ieee.org  Tue Jul 22 18:55:01 2008
From: sohaib.majzoub at ieee.org (Sohaib Majzoub)
Date: Tue, 22 Jul 2008 18:55:01 -0700 (PDT)
Subject: [Hotspot] hotspot compilation problem on Solaris 10
In-Reply-To: <487DFF68.1080108@cs.virginia.edu>
Message-ID: <441023.58496.qm@web50208.mail.re2.yahoo.com>

Hi Everyone,
I am trying to compile HostSpot on Solaris 10, but I am getting this error:
make: Fatal error in reader: Makefile, line 51: Unexpected end of line seen
?
please can you advise?
The OS is Solaris 10, and the SUN Studio links are:
INCDIR????????? = /SunStudio/SUNWspro/include
LIBDIR????????? = /SunStudio/SUNWspro/lib

thank you
Sohaib


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20080722/ba41121d/attachment.html 

From skadron at cs.virginia.edu  Tue Jul 22 20:57:16 2008
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Tue, 22 Jul 2008 20:57:16 -0700
Subject: [Hotspot] hotspot compilation problem on Solaris 10
In-Reply-To: <441023.58496.qm@web50208.mail.re2.yahoo.com>
References: <441023.58496.qm@web50208.mail.re2.yahoo.com>
Message-ID: <4886AC1C.5010000@cs.virginia.edu>

Are you using gnu make (gmake)?  I seem to recall that this error occurs 
if you don't use gmake.

/K

Sohaib Majzoub wrote:
> Hi Everyone,
> 
> I am trying to compile HostSpot on Solaris 10, but I am getting this error:
> 
> make: Fatal error in reader: Makefile, line 51: Unexpected end of line seen
> 
>  
> 
> please can you advise?
> 
> The OS is Solaris 10, and the SUN Studio links are:
> 
> INCDIR          = /SunStudio/SUNWspro/include
> LIBDIR          = /SunStudio/SUNWspro/lib
> 
> thank you
> 
> Sohaib
> 
> 

From mrg at soe.ucsc.edu  Thu Jul 24 13:53:11 2008
From: mrg at soe.ucsc.edu (Matthew Guthaus)
Date: Thu, 24 Jul 2008 13:53:11 -0700
Subject: [Hotspot] Hotspot results
Message-ID: <3987648F-F7EB-434E-AE45-A7D2CF40DF47@soe.ucsc.edu>

Hi,

Thank you for releasing your program. I have a question about results  
from Hotspot and Hotfloorplanner. The hotfloorplanner paper shows  
results of 120C for benchmark gcc on the ev6.

When I run hotspot on the ev6 given the gcc.ptrace (using the initial  
steady state temp method as described in the README):

./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace -o gcc.ttrace - 
steady_file gcc.steady -model_type grid -grid_layer_file layer.lcf
cp gcc.steady gcc.init
./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace -o gcc.ttrace - 
init_file gcc.init -model_type grid -grid_layer_file layer.lcf

I get max temperatures of about 341K (68C).

Did you alter the heatsink or other parameters to get these 120 degree  
temperatures? Would you share those values? What is your guideline for  
selecting appropriate parameters?

Thanks,

Matt




_______________________________________________
Matthew Guthaus
Assistant Professor, Computer Engineering
University of California Santa Cruz
http://www.soe.ucsc.edu/~mrg


From wh6p at cms.mail.virginia.edu  Thu Jul 24 14:21:50 2008
From: wh6p at cms.mail.virginia.edu (Wei Huang)
Date: Thu, 24 Jul 2008 17:21:50 -0400
Subject: [Hotspot] Hotspot results
In-Reply-To: <3987648F-F7EB-434E-AE45-A7D2CF40DF47@soe.ucsc.edu>
References: <3987648F-F7EB-434E-AE45-A7D2CF40DF47@soe.ucsc.e
 du>
Message-ID: <4A3EEF5ACD01EDE390571CDA@mstu2>

Matt,

We have updated the heatsink convection resistance and the 
thermal interface material thickness to smaller values to 
keep up with the trend in the industry (0.8K/W to 0.3K/W 
to 0.1K/W for heatsink, and ~70um to ~10um for TIM, if I 
remember correctly). Over the last few years, some old 
configurations in HotSpot 1.0 and 2.0 became obsolete. 
These changes are documented in the "what's new" part of 
the README files for each HotSpot release.

Thanks,
-Wei

--On Thursday, July 24, 2008 1:53 PM -0700 Matthew Guthaus 
<mrg at soe.ucsc.edu> wrote:

> Hi,
>
> Thank you for releasing your program. I have a question
> about results   from Hotspot and Hotfloorplanner. The
> hotfloorplanner paper shows   results of 120C for
> benchmark gcc on the ev6.
>
> When I run hotspot on the ev6 given the gcc.ptrace (using
> the initial   steady state temp method as described in
> the README):
>
> ./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace -o
> gcc.ttrace -  steady_file gcc.steady -model_type grid
> -grid_layer_file layer.lcf cp gcc.steady gcc.init
> ./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace -o
> gcc.ttrace -  init_file gcc.init -model_type grid
> -grid_layer_file layer.lcf
>
> I get max temperatures of about 341K (68C).
>
> Did you alter the heatsink or other parameters to get
> these 120 degree   temperatures? Would you share those
> values? What is your guideline for   selecting
> appropriate parameters?
>
> Thanks,
>
> Matt
>
>
>
>
> _______________________________________________
> Matthew Guthaus
> Assistant Professor, Computer Engineering
> University of California Santa Cruz
> http://www.soe.ucsc.edu/~mrg
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot