NPPS Leadership Team

[Torre Wenaus <wenaus AT gmail.com>]

The next step on the NPP side comes soon, May 25, below. Meifeng will inform Kerstin we need a CSI decision soon so we know whether we should present on the 25th.

  Torre

Dear PI’s,

 

Your FY2022 LDRD Type A abstract proposal has been selected to move to the next step.  I am scheduling a 1^st full draft presentation on May 25^th at 3:00 pm.  An indico web site has been created: https://indico.bnl.gov/event/11953/.

The presentations are to be 8 minutes in duration and should be uploaded prior to the meeting. Please conform to the time constraints so that all proposals can be reviewed. Kindly review all the attachments, in order to ensure that your proposal will meet the requirements. Below are some points; again to be noted along with the schedule and timeline.

 

Please reach out to your Hong Ma or Thomas Roser;  your department chairs for feedback and recommendations.

 

A meeting notice will the Zoom details will follow.

 

Thank you,

Fran

On Thu, May 20, 2021 at 9:11 AM Laycock, Paul <laycock AT bnl.gov> wrote:

Excellent, good luck moving forward !

On 20 May 2021, at 14:43, Torre Wenaus <wenaus AT gmail.com> wrote:

Hong informs me that we made the NPP cut. And some others were cut so it is good we put up the materials. Thanks for the contributions!

Meifeng is working on the CSI pitch.

  Torre

On Wed, May 19, 2021 at 1:17 PM Laycock, Paul <laycock AT bnl.gov> wrote:

Added my two cents on the focus, deployment of a service obviously raises the question of “what about SDCC” and will get shot down, having computer science expertise in CSI work with NPPS experts who are trying to solve this problem in terms of s/w for the LHC and beyond makes perfect sense, this should ensure the solution to that problem is generalised.

End of two cents :).  Good luck, I think it’s a great idea

Paul

On 19 May 2021, at 19:06, Torre Wenaus via Phys-npps-mgmt-l <phys-npps-mgmt-l AT lists.bnl.gov> wrote:

...and here is a 1 paragraph summary asked for.

https://docs.google.com/document/d/1uxoj_pUBG17Qo-4LExeE2_pUGLDoN41Kn5CuTIZ1KsI/edit?usp=sharing

On Wed, May 19, 2021 at 12:57 PM Torre Wenaus <wenaus AT gmail.com> wrote:

She's not too keen to build in google, me neither frankly. My slides for NPP thus far, editing on... note that the results includes commercial clouds etc...

https://docs.google.com/presentation/d/1ZY7-s1I4iUmGE-vNbzpD3bdQEZxf1I7xS5YrvMuSoZc/edit?usp=sharing

 Torre

On Wed, May 19, 2021 at 12:07 PM Alexei Klimentov <alexei.klimentov AT cern.ch> wrote:

Yes, but this can also be interpreted as that we invite them to our project (if they are interested), may be Meifeng can check with her boss directly, before it will be publicly available

On May 19, 2021, at 6:02 PM, Torre Wenaus <wenaus AT gmail.com> wrote:

Will mention google, grid, clusters... HPC is primary because that's what CSI HPC group brings

On Wed, May 19, 2021 at 11:57 AM Alexei Klimentov via Phys-npps-mgmt-l <phys-npps-mgmt-l AT lists.bnl.gov> wrote:

BTW, why not to expand it to Google (not only HPC) 

We propose a collaboration between CSI's HPC group and the Physics Department's NPPS group to combine their expertise towards developing, deploying and applying distributed machine learning services. CSI's HPC expertise combined with NPPS's cloud expertise and experience building ML services using PanDA and iDDS and working with physicists in identifying and supporting physics analysis use cases will enable us to effectively leverage large scale HPC and commercial clouds resources for ML to a much greater extent than either CSI or the NPPS team can on their own.

On May 19, 2021, at 5:47 PM, Brett Viren via Phys-npps-mgmt-l <phys-npps-mgmt-l AT lists.bnl.gov> wrote:

Hi Torre,

Torre Wenaus writes:

I believe she's asked Brett to participate too.

Yes, I've just read it.

It looks good but I think we need to sharpen the argument a bit.

For example, billing it as an "AI/ML service" puts it in competition
with established SaaS (AIMLaaS?) eg such as provided by Google.  We'll
need to indicate why our idea is novel.

To me, what is novel is that we need GPUs and other special hardware in
the otherwise "historical" HTC environment.  We need GPUs not just to
train AI/ML and run AI/ML inference but also to accelerate "mundane"
algorithms (eg FFTs in DUNE's case).  Training can be done in HPC but
these latter two are applied in HTC.

More generally speaking, I think we are working to balance the "ratio of
ratios" (CPU/GPU required by software vs provided by hardware).

We should invent a new buzzword for this marriage of HTC/HPC ("HT/PC"?)


-Brett.
_______________________________________________
Phys-npps-mgmt-l mailing list
Phys-npps-mgmt-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/phys-npps-mgmt-l

_______________________________________________
Phys-npps-mgmt-l mailing list
Phys-npps-mgmt-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/phys-npps-mgmt-l

-- 

-- Torre Wenaus, BNL NPPS Group, ATLAS Experiment

-- BNL 510A 1-222 | 631-681-7892 |  wenaus AT gmail.com | npps.bnl.gov | wenaus.com

-- NPPS Mattermost room: https://chat.sdcc.bnl.gov/npps/channels/town-square

-- 

-- Torre Wenaus, BNL NPPS Group, ATLAS Experiment

-- BNL 510A 1-222 | 631-681-7892 |  wenaus AT gmail.com | npps.bnl.gov | wenaus.com

-- NPPS Mattermost room: https://chat.sdcc.bnl.gov/npps/channels/town-square

-- 

-- Torre Wenaus, BNL NPPS Group, ATLAS Experiment

-- BNL 510A 1-222 | 631-681-7892 |  wenaus AT gmail.com | npps.bnl.gov | wenaus.com

-- NPPS Mattermost room: https://chat.sdcc.bnl.gov/npps/channels/town-square

_______________________________________________
Phys-npps-mgmt-l mailing list
Phys-npps-mgmt-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/phys-npps-mgmt-l

-- 

-- Torre Wenaus, BNL NPPS Group, ATLAS Experiment

-- BNL 510A 1-222 | 631-681-7892 |  wenaus AT gmail.com | npps.bnl.gov | wenaus.com

-- NPPS Mattermost room: https://chat.sdcc.bnl.gov/npps/channels/town-square

--

-- Torre Wenaus, BNL NPPS Group, ATLAS Experiment

-- BNL 510A 1-222 | 631-681-7892 |  wenaus AT gmail.com | npps.bnl.gov | wenaus.com

-- NPPS Mattermost room: https://chat.sdcc.bnl.gov/npps/channels/town-square