Skip to content

Commit fc077f7

Browse files
committed
Update to CDSS26 subject
1 parent 7005779 commit fc077f7

File tree

5 files changed

+27
-47
lines changed

5 files changed

+27
-47
lines changed

apply.md

Lines changed: 7 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -8,23 +8,23 @@ layout: page
88
To apply to the summer school, you must:
99

1010
- Be enrolled in a graduate program at an accredited university in a relevant field of study
11-
- Able to work onsite in Los Alamos for 10 weeks, sometime from mid-May through mid-August of 2025 (exact dates TBD)
11+
- Able to work onsite in Los Alamos for 10 weeks, sometime from mid-May through mid-August of 2026 (exact dates TBD)
1212
- **The School is open to non-U.S. citizens!**
1313

14-
## Desired Qualifications for 2025 CDSS
14+
## Desired Qualifications for 2026 CDSS
1515

16-
- **Computer Scientists:** Specializations in HPC with expertise in MPI, benchmarking, GPU programming, optimization, and performance analysis.
17-
- **Applied Mathematicians:** Focus on numerical methods, including experience with implicit solvers, preconditioners, and developing or using linear algebra libraries.
18-
- **Physicists:** Strong background in hydrodynamics, radiation physics, and shock physics with an interest in applying these areas to computational simulations.
16+
- **Computer Scientists:** Specializations in HPC with expertise in MPI, benchmarking, GPU programming, optimization, and performance analysis
17+
- **Applied Mathematicians:** Focus on numerical methods, including experience with time integration schemes for PDEs, hybrid methods
18+
- **Physicists:** Strong background in multimaterial hydrodynamics, detonation, combustion, and reactive flow models with an interest in applying these areas to computational simulations
1919

20-
Students will be chosen based both on their qualifications and on the alignment of their skills and interests with the focus area of the summer school. Broadly speaking, we will attempt to create a balanced team with an even distribution of students from the desired disciplines.
20+
Students will be chosen based both on their qualifications and on the alignment of their skills and interests with the focus area of the summer school. Broadly speaking, we will attempt to create a balanced team with an even distribution of students from the desired disciplines.
2121

2222
## Compensation
2323

2424
Los Alamos National Laboratory offers very competitive compensation:
2525

2626
* 10 week salary of $9-12K (based on education and experience)
2727

28-
## [Application Form](https://forms.gle/nYpjj8tyBh5AwMVf9)
28+
## [Application Form](https://forms.gle/kQHPpE3JHnvkexQk7)
2929

3030
Please use the form and include your C.V. and a cover letter describing your research interests.

assets/img/cdss2026_subject.png

345 KB
Loading

contact.md

Lines changed: 0 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -17,11 +17,4 @@ layout: page
1717

1818
[Email](mailto:[email protected])
1919

20-
## Administration
21-
22-
### Erika Maestas
23-
24-
(505) 665-2000
25-
26-
[Email](mailto:[email protected])
2720

curStaff.md

Lines changed: 1 addition & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -86,13 +86,6 @@ Maxim is a computational scientist in CCS-7 with a background in High-Performanc
8686
Maxim is an alumnus of the 2021 and 2022 summer school classes.
8787
</div>
8888

89-
#### Nirmal Prajapati
90-
91-
<!-- <div style="display:inline-block">
92-
<img src="/cdss/assets/img/nirmal.jpeg" style="margin-right:15px; width: 50%; float: left;"/> -->
93-
CCS-7 Applied Computer Science
94-
<!-- </div> -->
95-
9689
#### Philipp Edelmann
9790

9891
<div style="display:inline-block; margin-bottom: 50px;">
@@ -162,6 +155,7 @@ Al McPherson is a computer scientist and the former lead of the CCS-7 Co-Design
162155
- Karen Tsai
163156
- Li-Ta (Ollie) Lo
164157
- Marc Charest
158+
- Nirmal Prajapati
165159
- Patricia (Pat) Grubel
166160
- Patrick McCormick
167161
- Reid Priedhorsky

index.html

Lines changed: 19 additions & 26 deletions
Original file line numberDiff line numberDiff line change
@@ -12,44 +12,37 @@
1212
</div>
1313
<br/>
1414

15-
<h3 style="color:red;"> Applications are open until January <strike>10</strike><br> 19, 2025 </h3>
15+
<h3 style="color:red;"> Applications are open until January 10 2026 </h3>
1616

17-
<h1> 2025 Co-Design Summer School Focus: Model implosion dynamics at scale using FleCSI-HARD </h1>
17+
<h1> 2026 Co-Design Summer School Focus: Advancing scale-bridge simulations in high explosive applications using HARD at Scale </h1>
1818

1919
<br/>
2020

2121

2222
<div style="width:100%; text-align: center; display: table;">
23-
<img src="/cdss/assets/img/cdss25_implosion.png" style="display: block; margin-left: auto; margin-right: auto; width: 50%;">
24-
Pb Shell implosion test. Multiple Richtmyer-Meshkov instabilities occur during implosion.
23+
<img src="/cdss/assets/img/cdss2026_subject.png" style="display: block; margin-left: auto; margin-right: auto; width: 80%;">
24+
Source: Source: Houim, R. W. (2021). A simplified burn model for simulating explosive effects and afterburning. Shock Waves, 31, 851–875.
2525
</div>
2626
<br>
2727

28-
Predictions of the implosion dynamics and the level of shell distortion induced by hydrodynamic instabilities such as Richtmyer-Meshkov are essential to model contemporary inertial confinement fusion (ICF) experiments.
29-
Radiation also plays a key role in these phenomena.
30-
Therefore, detailed radiation hydrodynamics simulations are crucial to understand these phenomena properly.
31-
In the Co-Design Summer School (CDSS) 2025, we propose using the FleCSI Hydrodynamics and Radiative Diffusion (FleCSI-HARD) framework to model various implosion dynamics and simulate the instabilities associated with these phenomena.
32-
FleCSI-HARD is an open-source radiation hydrodynamics code that was developed from FleCSI-SP by the students of CDSS 2024.
33-
This code implements a basic flux limited diffusion (FLD) model and a multigrid solver. CDSS 2025 will focus on updating the FleCSI-HARD code and testing it at scale. Details are as follows.
34-
- More complex radiation hydrodynamics schemes will be implemented for the current code with the addition of different possible closures: radiation energy-pressure relation called Pn, variable Eddington factor, Larsen-FLD, etc.
35-
- A better set of solvers will be implemented through the use of FleCSolve, a FleCSI-based code supporting Algebraic Multigrid methods.
36-
- Variable opacity and a time-stepping/integration scheme will be added for better performance and accuracy.
37-
- The code will then be tested with verification and validation problem setups such as the <a href="https://github.com/lanl/ExactPack"> ExactPack project </a>. The school will model large-scale three-dimensional single and double shell implosion problems with multi-mode instabilities that can be used in ICF studies.
38-
39-
The increasing complexity of supercomputers, in both number of nodes and on-node hybridization, forces us to rethink our approach to high performance computing(HPC).
40-
Task-based parallelism provides a promising path forward. CDSS 2025 proposes to use the FleCSI framework as a base for the simulations.
41-
FleCSI is a compile-time configurable framework from LANL.
42-
It supports the development of multiphysics applications and introduces a functional programming model that can use multiple backends such as Legion, MPI, and HPX.
43-
FleCSI proposes different topologies that are then extended via a specialization, allowing it to target specific problems. LANL’s new, next-generation exascale supercomputer, Venado, combines the new NVIDIA architectures, Grace and Hopper. CDSS 2025 aims to run simulations at scale on Venado.
4428

45-
<div style="width:100%; text-align: center; display: table;">
46-
<img src="/cdss/assets/img/cdss25_kh.png" style="display: block; margin-top:5%; margin-left: auto; margin-right: auto; width: 50%;">
47-
Representation of mass density for Kelvin-Helmholtz instability with strong radiation (performed with 10242 resolution on NV V100 GPU) Radiation + Hydrodynamics Simulation using <a href="https://github.com/lanl/hard" >HARD code </a>.
48-
</div>
29+
Many natural and engineered systems, from astrophysical supernovae to nuclear detonations and high-explosive experiments, involve rapid, intense energy release that strongly couples to both hydrodynamic and radiative processes. To understand these phenomena correctly, the study of high-explosive (HE) models is essential. Furthermore, real-world applications span an extraordinary range of spatial and temporal scales, from the submicron chemistry of reaction zones, to the mesoscale dynamics of heterogeneous media, to the macroscale propagation of shock and detonation waves. Many existing works lack the machinery to bridge these scales in a consistent manner. In this work, we propose to incorporate HE physics and multimaterial interface capabilities into HARD, a radiation hydrodynamics code based on the FleCSI framework, to overcome these challenges.
4930

50-
<br>
31+
<br/>
32+
33+
The new framework will accommodate multiple HE-burn models, including programmed burn and reactive burn, by employing an explicit multimaterial formulation. Material interfaces will be captured with a level-set method, while a ghost-fluid approach will enforce the correct pressure and velocity conditions across discontinuities. Leveraging the Singularity-EOS library, each constituent (explosive, product gases, air, etc.) can be described by its own equation of state, preserving numerical stability even in the presence of strong shocks. In parallel, we will augment HARD’s existing hydrodynamic solvers with the additional physics modules required for these capabilities.
34+
35+
<br/>
36+
37+
Validation will be performed on canonical test problems such as shock-tube and Sedov-blast configurations, as well as on experimentally anchored benchmarks including one-dimensional detonation-tube and two-dimensional cylinder-expansion tests. The ExactPack library will be used to compare simulation results against analytical solutions, providing a thorough quantitative assessment of accuracy and robustness.
38+
39+
<br/>
40+
41+
The increasing complexity of supercomputers, in both number of nodes and on-node hybridization, forces us to rethink our approach to high-performance computing (HPC). Task-based parallelism provides a promising path forward. CDSS 2026 proposes to use the FleCSI framework, with HARD, as the basis for the simulations. FleCSI is a compile-time configurable framework from LANL. It supports the development of multiphysics applications and introduces a functional programming model that can use multiple backends such as Legion, MPI, and HPX. FleCSI provides different topologies that are then extended via specialization, allowing it to target specific problems. CDSS 2026 aims to run simulations at scale on LANL’s world-class supercomputers.
42+
43+
<br/>
5144

52-
LA-UR-24-31298
45+
LA-UR-25-29849
5346

5447

5548
<h1> What is the Co-Design Summer School? </h1>

0 commit comments

Comments
 (0)