Skip to content

Commit

Permalink
update to common structure
Browse files Browse the repository at this point in the history
  • Loading branch information
pelmer committed Jul 17, 2024
1 parent 4949a58 commit 5e07082
Show file tree
Hide file tree
Showing 11 changed files with 80 additions and 38 deletions.
6 changes: 4 additions & 2 deletions _trainees/Lael-Verace.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,15 +15,15 @@ networks:
github-username: lverace
---

### Biography
### Biography and Interests

I am a second-year graduate student in the CMS group at the University of Wisconsin-Madison, advised by Professor Kevin Black.

### Project

I work with Fermilab scientists Dr. Syed Asif Raza Shah and Dr. Oliver Gutsche on a networking R&D project titled "Performance Testing of High-Speed Scientific Networks through A Distributed Load Generation Approach." Our goal is to develop an infrastructure which allows us to test the performance of wide area network connections among scientific computing sites on a large scale. We aim to do this through the dynamic deployment of lightweight [perfSONAR](https://www.perfsonar.net/) Testpoint containers across multiple machines which generate network traffic to a given site. This approach allows us to send enough traffic to saturate the network pipe, enabling tests of the network performance in realistic high-volume scenarios. Monitoring the resulting throughput and comparing it to expected values enables us to diagnose unforeseen bottlenecks that might be limiting the performance of scientific network connections once data-volumes increase past their current amounts. By finding and fixing those bottlenecks now, we can prepare and optimize our high-speed scientific network for the future.

#### Recent Accomplishments
### Recent Accomplishments
* Since starting the project, I've gained a vast amount of valuable knowledge and hands-on experience in containerization, networking, and server management. It really amazes me when I look back on how much I've learned in a relatively short time just by working on this project!
* Developed a system for conducting automated [iPerf3](https://iperf.fr/) network performance tests between [Podman](https://podman.io/) containers hosted on two separate servers using systemd timers.
* Utilized perfSONAR to conduct coordinated throughput measurements between multiple testbed servers. Currently working to expand this and implement a centralized monitoring tool to keep track of all ongoing measurements.
Expand All @@ -32,3 +32,5 @@ I work with Fermilab scientists Dr. Syed Asif Raza Shah and Dr. Oliver Gutsche o
* Kevin Black (University of Wisconsin-Madison)
* Syed Asif Raza Shah (Fermilab)
* Oliver Gutsche (Fermilab)

### Traineeship dates
10 changes: 8 additions & 2 deletions _trainees/ajtoler.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,10 +14,16 @@ networks:
presentations:
---

### Biography
### Biography and Interests

I am a second year graduate student currently working in the UMass Amherst ATLAS group, advised by Carlo Dallapiccola.

### Project interests
### Project

### Recent Accomplishments

### Mentors

### Traineeship dates


9 changes: 7 additions & 2 deletions _trainees/ashq7.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,11 +14,16 @@ networks:
presentations:
---

### Biography
### Biography and Interests
I am a second-year PhD student in the CMS group at Princeton University working with Professor Isobel Ojalvo. I joined the TAC-HEP project in the spring. During this time I took Introduction to programming systems at Princeton which includes modular programming, advanced program design, programming style, test, debugging and performance tuning; machine languages and assembly language; and use of system call services. This is my first summer with the TAC-HEP project.

### Project interests
### Project
I am working on Phase 2 L1 Calorimeter Trigger upgrades, algorithms and Fast-Machine Learning.

### Recent Accomplishments

### Mentors

### Traineeship dates


10 changes: 8 additions & 2 deletions _trainees/ekauffma.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,13 +14,19 @@ networks:
presentations:
---

### Biography
### Biography and Interests
I am a second-year graduate student at Princeton University working in the CMS group and advised by Prof. Isobel Ojalvo. Previously I was working with IRIS-HEP on a Primary Verted Finder Algorithm known as PV-Finder. PV-Finder is a hybrid deep learning algorithm which identifies primary vertices. This algorithm was developed for use in conjunction with the LHCb detector in Run 3 of the LHC, which will experience a luminosity that is 5.5 times that of Run 2. In LHCb data, the efficiency of the CNN has inreased from to 90% to past 98% over the course of the past few years. The success of PV-Finder motivates its extension to both the ATLAS and CMS experiments. This project is concerned with the adaptation of the PV-Finder algorithm to ATLAS and CMS.

Now as a student at Princeton enrolled in the TAC-HEP program I have taken Introduction to programming systems. This class includes modular programming, advanced program design, programming style, test, debugging and performance tuning; machine languages and assembly language; and use of system call services.

### Project interests
I am interested in how the field of high energy physics motivates advances in computing. I am currently working on a fast anomaly detection using machine learning in the CMS L1 Trigger.

### Project

### Recent Accomplishments

### Mentors

### Traineeship dates


9 changes: 6 additions & 3 deletions _trainees/matthewharri.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,13 +14,16 @@ networks:
presentations:
---

### Biography
### Biography and Interests

I am a third year graduate student currently working in the UMass Amherst ATLAS group, advised by Stephane Willocq.
I am a third year graduate student currently working in the UMass Amherst ATLAS group, advised by Stephane Willocq.

### Project interests
### Project

I currently work on ACTS (A Common Tracking Software) developing a prototype navigation model for the ATLAS muon spectrometer that could potentially be used for HL-LHC. I also work on a standard model VH physics analysis, more precisely WH(WW) with a three lepton final state.

### Recent Accomplishments

### Mentors

### Traineeship dates
21 changes: 18 additions & 3 deletions _trainees/mgorsuch.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,13 +14,26 @@ github-username: mirarenee
presentations:
---

### Biography
### Biography and Interests

I am a third year graduate student working in the Observational Cosmology group at University of Wisconsin - Madison.

### Project Description
### Project

Miranda Gorsuch is contributing to the development and testing of novel algorithms for combining multiple optical/near-IR pixel-level images in the same region of the sky to make robust, high signal-to-noise measurements of faint galaxies for weak gravitational lensing and galaxy clustering probes of dark energy with the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST). Specifically, she is working on the implementation of the cell-based (or “edgeless”) image coaddition algorithm in the LSST Science Pipelines. Cell-based coadds are substantially more computationally efficient when working with large numbers of overlapping images and enable more robust modeling of the point-spread function (PSF). The approach will be tested and optimized for computational and science performance on increasingly large datasets until it is validated for deployment on survey-scale LSST data release processing.
I am contributing to the development and testing of novel algorithms
for combining multiple optical/near-IR pixel-level images in the
same region of the sky to make robust, high signal-to-noise
measurements of faint galaxies for weak gravitational lensing and
galaxy clustering probes of dark energy with the Vera C. Rubin
Observatory Legacy Survey of Space and Time (LSST). Specifically,
I am working on the implementation of the cell-based (or “edgeless”)
image coaddition algorithm in the LSST Science Pipelines. Cell-based
coadds are substantially more computationally efficient when working
with large numbers of overlapping images and enable more robust
modeling of the point-spread function (PSF). The approach will be
tested and optimized for computational and science performance on
increasingly large datasets until it is validated for deployment
on survey-scale LSST data release processing.

### Recent Accomplishments
- Developed collection of testing analyses for cell-based coadds (in progress) [GitHub link](https://github.com/mirarenee/notebooks/blob/main/cell_coadds/cell_coadd_sandbox.ipynb)
Expand All @@ -30,3 +43,5 @@ Miranda Gorsuch is contributing to the development and testing of novel algorith
### Mentors
- Keith Bechtol (Univeristy of Wisconsin - Madison, advisor)
- Alex Drlica-Wagner (Fermilab)

### Traineeship dates
7 changes: 5 additions & 2 deletions _trainees/mmaroun.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,9 +14,12 @@ networks:
presentations:
---

### Biography
### Biography and Interests

### Project

### Project interests
### Recent Accomplishments

### Mentors

### Traineeship dates
10 changes: 7 additions & 3 deletions _trainees/nanoemc.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,9 +14,13 @@ networks:
presentations:
---

### Biography and Project Description
### Biography and Interests

I am a graduate student at the University of Wisconsin-Madison working on a PhD in High Energy Experimental Particle Physics with the CMS. I currently work on the Elastic Analysis Facility (EAF) hosted at Fermilab. It is part of the R&D effort to develop high performing analysis tools that we can use for HL-LHC and generally for other future experiments. Specifically I work on developing the Dask setup and system that integrates with Fermilab’s HTCondor system. I hope to improve the way we use Dask so that we can sustain and support many users in an easy to implement way. Other tasks I work on are benchmarking for the EAF, acting as a liaison between the EAF and IRIS-HEP, and I generally aid and support as a member of the EAF applications team.
I am a graduate student at the University of Wisconsin-Madison working on a PhD in High Energy Experimental Particle Physics with the CMS experiment.

### Project

I currently work on the Elastic Analysis Facility (EAF) hosted at Fermilab. It is part of the R&D effort to develop high performing analysis tools that we can use for HL-LHC and generally for other future experiments. Specifically I work on developing the Dask setup and system that integrates with Fermilab’s HTCondor system. I hope to improve the way we use Dask so that we can sustain and support many users in an easy to implement way. Other tasks I work on are benchmarking for the EAF, acting as a liaison between the EAF and IRIS-HEP, and I generally aid and support as a member of the EAF applications team.

### Recent Accomplishments
- [Demonstration Video of the EAF](https://indico.cern.ch/event/1291680/contributions/5581053/attachments/2714147/4713730/EAF_Demo_Video_with_audio.mp4) - Combination of the work described, produced for the AGC Demonstration Day
Expand All @@ -31,7 +35,7 @@ I am a graduate student at the University of Wisconsin-Madison working on a PhD
* Burt Holtzman (Fermilab>)
* Maria Acosta Flechas (Fermilab)

### Dates
### Traineeship dates

* 2023-01-24 to 2025-01-24

20 changes: 6 additions & 14 deletions _trainees/rpsimeon34.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,24 +10,12 @@ dates:
photo: /assets/images/trainees/Ryan-Simeon2024.jpeg
institution: University of Wisconsin - Madison
e-mail: [email protected]
website: https://www.physics.wisc.edu/directory/simeon-ryan-p/
networks:
- cms
project_goal: >
I am working on the set up and configuration of an analysis facility at UW-Madison. This work is being
done in collaboration with scientists at Fermilab and University of Nebraska/IRIS-HEP. Plans include installing
JupyterHub, as well as configuring distributed computing access to the UW-Madison both CPU and GPU resources at
the CMS Tier-2 and the Center for High Throughput Computing facility. I am investigating the use of HT-Condor
and Kubernetes. Once functional, I will be working on an LHC physics analysis that exploits the resources and motivates
further facility development.
mentors:
- <Sridhara Dasu (UW-Madison)
presentations:
current_status:
github-username: rpsimeon34
---

### Biography
### Biography and Interests

I am a second-year graduate student at the University of Wisconsin - Madison, working in the CMS group and advised by Professor Sridhara Dasu.

Expand All @@ -36,7 +24,7 @@ I am a second-year graduate student at the University of Wisconsin - Madison, wo
I am contributing to the Analysis Facilities effort within USCMS computing working with Fermilab scientists Dr. Lindsey Gray and Dr. Nick Smith. I am developing tools that make it easier for physicists to harness large amounts of computing power and big data for scientific analyses. In particular, I am working on a JupyterLab interface and built-in Python package that presents users with an environment pre-configured for modern Python-based data analysis and high throughput computing at our particular Analysis Facility.
The work involves integrating with the CMS Tier-2 at UW-Madison and the Center for High Throughput Computing (CHTC) resources. This includes developing a [repository of example and tutorial code](https://github.com/rpsimeon34/wisc-af-examples) that guides users from first interaction with contemporary High Energy Physics packages to performing data analysis on a computing cluster. I am also making contributions to various packages deployed at Analysis Facilities, including improving remote data access with [fsspec-xrootd](https://github.com/CoffeaTeam/fsspec-xrootd) and developing documentation for [coffea](https://github.com/CoffeaTeam/coffea).

### Recent Work
### Accomplishments

- Created a [repository of annotated examples](https://github.com/rpsimeon34/wisc-af-examples) for using the Analysis Facility
- Created the [cowtools](https://github.com/rpsimeon34/cowtools) package for off-the-shelf distributed computing tools for our particular Analysis Facility
Expand All @@ -45,3 +33,7 @@ The work involves integrating with the CMS Tier-2 at UW-Madison and the Center f
* Sridhara Dasu (University of Wisconsin - Madison)
* Nick Smith (Fermi National Accelerator Laboratory)
* Lindsey Gray (Fermi National Accelerator Laboratory)

### Traineeship dates


14 changes: 9 additions & 5 deletions _trainees/skkwan.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,15 +14,19 @@ networks:
presentations:
---

### Biography
### Biography and Interests
I am an an alumni of the TAC-HEP Progam. I participated in 2022-2024. I got my PhD from Princeton Universityin CMS experiment group advised by Prof. Isobel Ojalvo. Formerly I was an undergraduate at Caltech, and was supported by the NSF GRFP during part of my graduate program.

### Project interests

I am very interested in High Level Synthesis and FPGAs. My research includes a search for new physics (exotic Higgs decays to pseudoscalars to two b-quarks and two tau leptons), and developing improved high-granularity algorithms for reconstructing electrons and photons in the Level-1 Trigger of the CMS Experiment for the Phase 2 High-Luminosity LHC upgrade.

As a summer project I developed a GELU (Gaussian Error Linear Unit) activation function for a Transformer on an FPGA using HLS. This involved creating high-level descriptions of the key components in C/C++ and using HLS tools to synthesize them into HDL code. The GELU function and Transformer’s self-attention mechanism serve as examples of how to approach this complex task. Properly designing and optimizing these components for FPGA resources and performance is crucial for an efficient implementation. I compared different methods for implementation: one using look up tables and one using DSPs and compared the resource utilization for each method.
### Project

As my TAC-HEP project I developed a GELU (Gaussian Error Linear Unit) activation function for a Transformer on an FPGA using HLS. This involved creating high-level descriptions of the key components in C/C++ and using HLS tools to synthesize them into HDL code. The GELU function and Transformer’s self-attention mechanism serve as examples of how to approach this complex task. Properly designing and optimizing these components for FPGA resources and performance is crucial for an efficient implementation. I compared different methods for implementation: one using look up tables and one using DSPs and compared the resource utilization for each method.

### Recent Accomplishments

### Mentors
* Jennifer Ngadiuba (Fermilab)
* Adrian Pol (Princeton, IRIS-HEP)
* Adrian Pol (Princeton, IRIS-HEP), Isobel Ojalvo (Princeton)

### Traineeship dates
2 changes: 2 additions & 0 deletions _trainees/twnelson0.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,8 @@ networks:
presentations:
---

### Biography and Interests

### Project
I am currently working on porting the doublet/triplet seed building code and seed matching code from the CMSSW e/gamma reconstruction algorithm to run on GPUs using the ALPAKA library. The e/gamma pixel matching algorithm was chosen as it comprises 5%-10% of both HLT and offline reconstruction time. The goal of the project is to have a working implementation of the GPU compatible code integrated in CMSSW by the end of this year. Over the past year I have performed efficiency studies on the reconstructed electrons. The efficiency is the ratio of the number of reconstructed electrons using the new algorithm to the number of reconstructed electrons using the old algorithm. These studies also compared the kinematic distributions of the reconstructed electrons and looked at the timing of the different algorithms. These studies have been used to validate the physics performance of the new algorithm. In March of 2024 I participated in the 15th Patatrack hackathon and wrote an initial implementation of an ALPAKA compatible endcap propagator. This function is designed to extrapolate a track that is compatible with an electron whose trajectory is towards one of the two ECAL endcaps. I then integrated this function into the electron seed producer.

Expand Down

0 comments on commit 5e07082

Please sign in to comment.