From dip224 at gmail.com  Tue Oct  2 22:02:31 2012
From: dip224 at gmail.com (subhadip kundu)
Date: Wed, 3 Oct 2012 10:32:31 +0530
Subject: [Hotspot] Problem in ttrace file
In-Reply-To: <CAOuN=+72yHbutpCsfAr8T0jc_TYgqvJ+AyWOXQ3WM_=CRtDktQ@mail.gmail.com>
References: <CAOuN=+72yHbutpCsfAr8T0jc_TYgqvJ+AyWOXQ3WM_=CRtDktQ@mail.gmail.com>
Message-ID: <CAOuN=+4VXiH+YbCPpN2+dQaHLxKhx8oWqp=+8TyEaFqk4dd5Xg@mail.gmail.com>

Hi,

The problem of ttrace file with transient temperature is still there...
Please Help !!!

On Fri, Sep 28, 2012 at 11:05 AM, subhadip kundu <dip224 at gmail.com> wrote:

> HI,
>
> I am having trouble in generating transient temperatures. I have given a
> power trace file containing power at different time instants. I want to
> check the change in transient temperatures. But the ttrace file generated
> by hotspot gives "nan" (not a number) values for most of the cases.
>
> Please help me solving this problem and if possible give me the details of
> how this ttrace file is generated.
>
> Thanks and Regards,
> Subhadip Kundu.
> IIT Kharagpur.
>



-- 
Subhadip Kundu.
IIT Kharagpur.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20121003/12e1c3ce/attachment.html 

From Runjie at virginia.edu  Wed Oct  3 06:19:01 2012
From: Runjie at virginia.edu (Runjie Zhang)
Date: Wed, 3 Oct 2012 09:19:01 -0400
Subject: [Hotspot] Problem in ttrace file
In-Reply-To: <CAOuN=+4VXiH+YbCPpN2+dQaHLxKhx8oWqp=+8TyEaFqk4dd5Xg@mail.gmail.com>
References: <CAOuN=+72yHbutpCsfAr8T0jc_TYgqvJ+AyWOXQ3WM_=CRtDktQ@mail.gmail.com>
	<CAOuN=+4VXiH+YbCPpN2+dQaHLxKhx8oWqp=+8TyEaFqk4dd5Xg@mail.gmail.com>
Message-ID: <CAHxZ6YAd42M_vBsUTQbSQavmGXy8TCqW6XeDmNyHTJ53hLQd_Q@mail.gmail.com>

Hi, Subhadip

  The problem will go away if you increase the thickness of either heatsink
or heatspreader.

Runjie

On Wed, Oct 3, 2012 at 1:02 AM, subhadip kundu <dip224 at gmail.com> wrote:

> Hi,
>
> The problem of ttrace file with transient temperature is still there...
> Please Help !!!
>
> On Fri, Sep 28, 2012 at 11:05 AM, subhadip kundu <dip224 at gmail.com> wrote:
>
>> HI,
>>
>> I am having trouble in generating transient temperatures. I have given a
>> power trace file containing power at different time instants. I want to
>> check the change in transient temperatures. But the ttrace file generated
>> by hotspot gives "nan" (not a number) values for most of the cases.
>>
>> Please help me solving this problem and if possible give me the details
>> of how this ttrace file is generated.
>>
>> Thanks and Regards,
>> Subhadip Kundu.
>> IIT Kharagpur.
>>
>
>
>
> --
> Subhadip Kundu.
> IIT Kharagpur.
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot
>
>


-- 
Runjie Zhang
Computer Engineering
University of Virginia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20121003/f900390e/attachment.html 

From skadron at cs.virginia.edu  Sun Oct  7 18:55:05 2012
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Sun, 07 Oct 2012 21:55:05 -0400
Subject: [Hotspot] 5-minute survey on future HotSpot development
Message-ID: <50723279.7040702@cs.virginia.edu>

Dear Hotspot user/developer,

Next year Hotspot will turn 10! This anniversary is a good opportunity 
to see what new extensions and improvements are needed, and we hope you 
are willing to help us prioritize. This 12-question survey should take 
just a few minutes, and as a small token of appreciation all responses 
will be entered in a raffle for UVa apparel (UVa T-shirt, UVa cap, UVa 
cup--if you win, we will email you for shipping information).

Numerical answers are sufficient, but feel free to add text comments.

a. Just to understand the context of your responses please state your 
level of involvement with Hotspot on a 1 to 5 scale (5 - active user and 
developer, 4 - intermitent user, 3 - past user, 2 - downloaded but never 
used, 1 - observer)

b. One possible extension would be a more elaborate floorplanning tool 
to allow rapid generation of HotSpot floorplans to support. On a 1 to 5 
scale please state your opinion about such an extension (5 - essential 
extension highly desirable, 4 - important, 3 - somewhat desirable, 2 - 
neutral, 1 - not important)

c. Another possible extension would be a more elaborate package modeling 
tool (HotPackage) to allow a thorough exploration of the package and 
cooling solution design space, from embedded systems with no heat 
spreader or heat sink, to more exotic schemes such as active fans, heat 
pipes, thermoelectric cooling, microchannels, etc. On a 1 to 5 scale 
please state your opinion about such an extension (5 - essential 
extension highly desirable, 1 - not important)

d. Yet another possible extension would be a more elaborate power 
delivery modeling tool (VoltSpot) to allow a thorough exploration of the 
power delivery design space at the architecture level, including 
on-chip and off-chip delivery, C4s, pins, and DC/DC converters. On a 1 
to 5 scale please state your opinion about such an extension (5 - 
essential extension highly desirable, 1 - not important)

e. Another extension would be to fully support 3D integration with 
Hotspot, including modeling TSVs, microfluidic cooling, etc. While many 
users (many of you) have developed various 3D extensions, those have not 
been fully integrated into the main release. On a 1 to 5 scale please 
state your opinion about such an extension (5 - essential extension 
highly desirable, 1 - not important)

f. An improvement to Hotspot would be the inclusion of faster, parallel 
solvers to allow for simulation scenarios that would be impractical with 
the current Hotspot due to simulation speed. On a 1 to 5 scale please 
state your opinion about such an improvement (5 - essential improvement 
highly desirable, 1 - not important)

g. A few thermal models exist for individual components, but we are not 
aware of any full-system (especially cluster/warehouse-scale) simulation 
capabilities.  On a 1 to 5 scale please state your opinion about such a 
resource (5 - essential resource highly desirable, 1 - not important)

h. A possible enhancement of the Hotspot infrastructure would be the 
addition of an IR camera, both as a validation and verification 
instrument, but also potentially as a shared resource for the Hotspot 
user community - this sharing could be done for example through a remote 
web-based interface. On a 1 to 5 scale please state your opinion about 
such a resource (5 - essential resource highly desirable, 1 - not important)

i. We occasionally get requests for help with integrating Hotspot with 
various other simulators. If you have encountered such difficulties 
please state briefly which simulator and what was the nature of the 
difficulty?

j. Some of the above extensions could be achieved in one of two ways: as 
a new effort on our side, or by trying to integrate existing code 
developed elsewhere. Please let us know if there are existing packages 
you would like to see integrated with HotSpot.

k. Finally, if you have developed new code based on Hotspot, are you 
willing to share it with us and offer support to be possibly included in 
a future release?

l. Are there any other improvements/extensions you would like to see?

Thank you for your time!

Yours truly,

Kevin and Mircea



From mircea at virginia.edu  Fri Oct 12 13:00:44 2012
From: mircea at virginia.edu (Mircea R. Stan)
Date: Fri, 12 Oct 2012 16:00:44 -0400
Subject: [Hotspot] Urgent and brief survey re: HotSpot
In-Reply-To: <5078733E.60306@cs.virginia.edu>
References: <50723279.7040702@cs.virginia.edu>
	<50784C9B.5010801@cs.virginia.edu>
	<web-4033272@koi.mail.virginia.edu>
	<5078733E.60306@cs.virginia.edu>
Message-ID: <web-4033418@koi.mail.virginia.edu>

Dear Hotspot user/developer,

Next year Hotspot will turn 10! This anniversary is a good opportunity to see 
what new extensions and improvements are needed, and in fact, we are applying 
for renewal funding to continue development of HotSpot.  We hope you are 
willing to help us prioritize.

We would like to hit 100% response rate from members on this list--students, 
faculty, industry--everyone!  We would like as representative a set of results 
as possible.  Even if you are a past user, or it worked fine and you have no 
suggestions for extensions, we still need your response!  Please respond by 
*Wednesday, Oct. 17*. Please fill the following on-line survey:

http://www.surveymonkey.com/s/YRV5YBX

This survey should take just a few minutes, and as a small token of 
appreciation all responses will be entered in a raffle for UVa apparel (one 
each UVa T-shirt, UVa cap, UVa cup, and three pens)--if you win, we will email 
you for shipping information. Thank you for your time and help!

Yours truly,

Kevin and Mircea

From mircea at virginia.edu  Tue Oct 16 16:35:43 2012
From: mircea at virginia.edu (Mircea R. Stan)
Date: Tue, 16 Oct 2012 19:35:43 -0400
Subject: [Hotspot] Reminder: Urgent and brief survey re: HotSpot
In-Reply-To: <web-4033418@koi.mail.virginia.edu>
References: <50723279.7040702@cs.virginia.edu>
	<50784C9B.5010801@cs.virginia.edu>
	<web-4033272@koi.mail.virginia.edu>
	<5078733E.60306@cs.virginia.edu>
	<web-4033418@koi.mail.virginia.edu>
Message-ID: <web-3719552@neon.mail.virginia.edu>

This is a quick reminder that Kevin and I would really appreciate if you
could fill in the survey below by tomorrow, Oct. 17. We have already
received a number of responses, but by no means as many as we would
like. Please try to find 5 minutes to fill the survey and help our community
to move forward. Thanks a lot for your help!

Mircea

http://www.surveymonkey.com/s/YRV5YBX

On Fri, 12 Oct 2012 16:00:44 -0400
  "Mircea R. Stan" <mircea at virginia.edu> wrote:
> Dear Hotspot user/developer,
> 
> Next year Hotspot will turn 10! This anniversary is a good opportunity to 
>see what new extensions and improvements are needed, and in fact, we are 
>applying for renewal funding to continue development of HotSpot.  We hope you 
>are willing to help us prioritize.
> 
> We would like to hit 100% response rate from members on this list--students, 
>faculty, industry--everyone!  We would like as representative a set of 
>results as possible.  Even if you are a past user, or it worked fine and you 
>have no suggestions for extensions, we still need your response!  Please 
>respond by *Wednesday, Oct. 17*. Please fill the following on-line survey:
> 
> http://www.surveymonkey.com/s/YRV5YBX
> 
> This survey should take just a few minutes, and as a small token of 
>appreciation all responses will be entered in a raffle for UVa apparel (one 
>each UVa T-shirt, UVa cap, UVa cup, and three pens)--if you win, we will 
>email you for shipping information. Thank you for your time and help!
> 
> Yours truly,
> 
> Kevin and Mircea

Mircea Stan

The views expressed in this email represent the personal
views of the sender and do not represent the official
position of the University of Virginia

From skadron at cs.virginia.edu  Wed Oct 17 19:14:03 2012
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Wed, 17 Oct 2012 22:14:03 -0400
Subject: [Hotspot] New ArchFP floorplanning tool
Message-ID: <507F65EB.5090005@cs.virginia.edu>

Dear HotSpot users,

There has been a lot of demand for a more convenient way to generate 
floorplans for use with HotSpot, especially for people who want to 
explore a design space that may encompass many floorplans (eg, different 
#s cores, different cache sizes, etc.)  The current HotFloorplan is 
fairly heavyweight, so we have developed a tool that we call ArchFP. 
You can generate a floorplan by writing a few lines of code.  The API 
allows you to define blocks, repeat them, mirror them, etc.  The 
programmatic nature also lets you use it within scripts to generate 
floorplans on the fly.

We have now released this new floorplanning tool. Please see 
http://lava.cs.virginia.edu/archfp/ for more detail and for downloads. 
We hope you will find it useful, and welcome your feedback!

--Kevin, Mircea, Runjie, and Greg

From skadron at cs.virginia.edu  Wed Oct 17 19:27:11 2012
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Wed, 17 Oct 2012 22:27:11 -0400
Subject: [Hotspot] New ArchFP floorplanning tool
In-Reply-To: <507F65EB.5090005@cs.virginia.edu>
References: <507F65EB.5090005@cs.virginia.edu>
Message-ID: <507F68FF.1080800@cs.virginia.edu>

Sorry to bombard the list, but I forgot to mention Brett Meyer (McGill) 
as one of the authors!
--Kevin

On 10/17/2012 10:14 PM, Kevin Skadron wrote:
> Dear HotSpot users,
>
> There has been a lot of demand for a more convenient way to generate
> floorplans for use with HotSpot, especially for people who want to
> explore a design space that may encompass many floorplans (eg, different
> #s cores, different cache sizes, etc.)  The current HotFloorplan is
> fairly heavyweight, so we have developed a tool that we call ArchFP. You
> can generate a floorplan by writing a few lines of code.  The API allows
> you to define blocks, repeat them, mirror them, etc.  The programmatic
> nature also lets you use it within scripts to generate floorplans on the
> fly.
>
> We have now released this new floorplanning tool. Please see
> http://lava.cs.virginia.edu/archfp/ for more detail and for downloads.
> We hope you will find it useful, and welcome your feedback!
>
> --Kevin, Mircea, Runjie, and Greg

From skadron at cs.virginia.edu  Thu Oct 18 06:05:51 2012
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Thu, 18 Oct 2012 09:05:51 -0400
Subject: [Hotspot] Reminder: Urgent and brief survey re: HotSpot
In-Reply-To: <web-3719552@neon.mail.virginia.edu>
References: <50723279.7040702@cs.virginia.edu>
	<50784C9B.5010801@cs.virginia.edu>
	<web-4033272@koi.mail.virginia.edu>
	<5078733E.60306@cs.virginia.edu>
	<web-4033418@koi.mail.virginia.edu>
	<web-3719552@neon.mail.virginia.edu>
Message-ID: <507FFEAF.8050604@cs.virginia.edu>

I apologize to those of you who have already received several emails 
about this.  But we just discovered that the "daily digest" feature of 
our listserv software is not working, and that people with the digest 
option enabled (many people) may not have received anything at all about 
this.  (We may have to disable this feature--stay tuned.)

We would really appreciate the highest possible response rate; it will 
really help guide us in general, and for the renewal funding, a high 
response rate is important to demonstrate strong user interest in these 
features!!!  So we would greatly appreciate it if you would help us out 
and take just 5 minutes for this survey!

Given the snafu with the mailing list, we'll extend the deadline - 
please respond by this Friday (10/19).

Thanks!!!
Kevin and Mircea


-------- Original Message --------
Subject: Urgent and brief survey re: HotSpot
Date: Fri, 12 Oct 2012 16:00:44 -0400
From: Mircea R. Stan <mircea at virginia.edu>
To: hotspot at cs.virginia.edu
CC: mircea at virginia.edu, skadron at cs.virginia.edu

Dear Hotspot user/developer,

Next year Hotspot will turn 10! This anniversary is a good opportunity 
to see  what new extensions and improvements are needed, and in fact, we 
are applying for renewal funding to continue development of HotSpot.  We 
hope you are willing to help us prioritize.

We would like to hit 100% response rate from members on this 
list--students, faculty, industry--everyone!  We would like as 
representative a set of results as possible.  Even if you are a past 
user, or it worked fine and you have no suggestions for extensions, we 
still need your response!  Please respond by *Wednesday, Oct. 17* [now 
Friday, Oct. 19]. Please fill the following on-line survey:

http://www.surveymonkey.com/s/YRV5YBX

This survey should take just a few minutes, and as a small token of
appreciation all responses will be entered in a raffle for UVa apparel 
(one each UVa T-shirt, UVa cap, UVa cup, and three pens)--if you win, we 
will email you for shipping information. Thank you for your time and help!

Yours truly,

Kevin and Mircea

From Yuanqing.Cheng at lirmm.fr  Wed Oct 31 05:50:05 2012
From: Yuanqing.Cheng at lirmm.fr (yuanqing)
Date: Wed, 31 Oct 2012 14:50:05 +0100
Subject: [Hotspot] hotspot for 3D IC simulation
Message-ID: <201210311450049276782@lirmm.fr>

Hi,
I want to ask a question on Hotspot. Can it simulate the 3D ICs? Does it consider the TSV effects for thermal dissipation during the thermal simulation? Thanks!




Best Regards!
-- 
Yuanqing Cheng, Ph.D.
LIRMM - CNRS / Universit? Montpellier 2 
UMR 5506 - CC 477 
161 rue Ada 
34095 Montpellier Cedex 5 France
Email: Yuanqing.Cheng at lirmm.fr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20121031/57e4df07/attachment.html 

From Runjie at virginia.edu  Wed Oct 31 06:16:13 2012
From: Runjie at virginia.edu (Runjie Zhang)
Date: Wed, 31 Oct 2012 10:16:13 -0400
Subject: [Hotspot] hotspot for 3D IC simulation
In-Reply-To: <201210311450049276782@lirmm.fr>
References: <201210311450049276782@lirmm.fr>
Message-ID: <CAHxZ6YDTatKJVD9BT44swnuhHsv8gEjaFcwXuB+-g-BTUH5RZA@mail.gmail.com>

Hello

  HotSpot supports 3D structure as long as each layer's thermal resistance
is uniform. So TSV or micro channel cooling are currently not supported by
HotSpot.

  However, Prof. Asye Coskun in Boston University has built an extended
version of HotSpot, which can model micro channel cooling and etc. You can
find related links here:

  http://lava.cs.virginia.edu/HotSpot/links.htm

Runjie

On Wed, Oct 31, 2012 at 9:50 AM, yuanqing <Yuanqing.Cheng at lirmm.fr> wrote:

> **
> Hi,
> I want to ask a question on Hotspot. Can it simulate the 3D ICs? Does it
> consider the TSV effects for thermal dissipation during the thermal
> simulation? Thanks!
>
> ------------------------------
>  Best Regards!
> --
> Yuanqing Cheng, Ph.D.
> LIRMM - CNRS / Universit? Montpellier 2
> UMR 5506 - CC 477
> 161 rue Ada
> 34095 Montpellier Cedex 5 France
> Email: Yuanqing.Cheng at lirmm.fr
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot
>
>


-- 
Runjie Zhang
Computer Engineering
University of Virginia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20121031/1c8aa792/attachment.html