ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
WORKSHOP PROGRAM
2
TimePlanRoom NumberSpeakerAffliliation
3
Thu 4/25Remote participants can join the zoom room for day one here.2405 Siebel Center
4
8:30-9:00Coffee & Bakery items
5
9:00-9:15IntroductionSanjay KaleUIUC
6
Session: ApplicationsChair: Maya Taylor
7
9:15-9:45SpECTRE: a task-based high-order code for relativistic astrophysicsNils Deppe and Lawrence KidderCornell
8
9:45-10:15(NAMD for Biomolecular Simulations: status and updates)David HardyUIUC
9
10:15-10:45Enzo-E / Cello: current status and recent updatesJames Bordner, Mike NormanUCSD
10
10:45-11:00Coffee Break
11
Session: PythonChair: Ritvik Rao
12
11:00-11:20CharmTyles: Large-scale interactive Charm++ with PythonAditya BhosaleUIUC
13
11:20-11:40Migratable User-Level Threads in PythonZane FinkUIUC
14
11:40-12:00Implementing the Ray Core API on Charm4PyAditya BhosaleUIUC
15
12:00-12:45Lunch
16
Session: Communication and Load balancingChair: Aditya Bhosale
17
12:45-13:15High-Performance Communication Support for Asynchronous Many-Task SystemsJiakun Yan, et alUIUC
18
13:15-13:45Supporting HPE Cassini Slingshot-11 interconnect via CXI extensions to Charm++/Converse OFI communication layerEric BohmUIUC
19
13:45-14:15Evaluating the Capability of Large Language Models for Parallel and High Performance Code GenerationDaniel NicholsUniversity of Maryland
20
14:15-14:45A Communication and Memory-Aware Model for Load BalancingJonathan LifflanderSandia
21
Session: Programming modelsChair: Zane Fink
22
14:45-15:00Coffee Break
23
15:00-15:30Chplx: an Asynchronous Many Task Runtime Foundation to ChapelShreyas Atre, et alLouisiana State University
24
15:30-16:00Evaluating HPX and Kokkos on RISC-V using an Astrophysics Application Octo-TigerPatrick Diehl, et alLouisiana State University
25
16:00-16:30Scaling Mixed-Precision Matrix Computations on Heterogeneous Architectures with PaRSECQinglei Cao and George BosilcaSaint Louis University
26
16:30-17:00Evaluating Performance Portability of GPU Programming ModelsJoy KitsonUniversity of Maryland
27
17:00-17:30Composing Distributed Programs Through Task and Kernel FusionRohan YadavStanford
28
17:30-18:00Quinoa: Massively parallel high-order multi-material hydrodynamicsAditya PandareLANL
29
19:00-21:00Dinner (At KoFusion)
30
Fri 4/26Remote participants can join the zoom room for day two talks here.
Join the zoom room for the day two tutorial here.
31
8:30-9:00Coffee & Bakery items
32
9:00-10:00Keynote - Arachne: An Open-Source Framework for Interactive Massive-Scale Graph AnalyticsDavid BaderNJIT
33
34
Session: Charm++ featuresChair: Kavitha Chandrasekar
35
10:00-10:20Only Pay for What You Need: Registration-Based Queueing in Asynchronous Many-Task SystemsZane FinkUIUC
36
10:20-10:35Coffee break
37
Session: Applications/RuntimesChair: Ritvik Rao
38
10:35-11:05The STAPL View and an Asynchronous Parallel Non-dominated Sorting for Distributed MemoryFrancisco Coral Sabido, et alTAMU, UIUC
39
11:05-11:35Embracing the pain of irregular parallelism on HPC and AWS-HPCRoger Pearce and Trevor SteilLLNL
40
11:35-12:05Computational Cosmology with ChaNGa on Leadership Class MachinesThomas QuinnUniversity of Washington
41
12:05-12:35Leveraging GPUs for Large-Scale Astrophysical Simulations With ChaNGaSpencer Wallace and Thomas QuinnUniversity of Washington
42
12:35-13:20Lunch
43
44
Potpourri (mixed topics talks)Chair: Aditya Bhosale3102 Siebel Center
45
13:20-13:40CkIO: Parallel file input for over-decomposed task-based systemsMathew Jacob and Maya TaylorUIUC
46
13:40-14:00Diffusion-based distributed load balancer in Charm++Kavitha ChandrasekarUIUC
47
14:00-14:30
Common infrastructure for dynamic load balancing: across multiple programming systems based on over-decompistions
Sanjay KaleUIUC
48
14:30-15:00Charm++ Advanced Users group discussion.
49
15:00-15:15Coffee break
50
51
Session: Charm++ TutorialTime subject to change3102 SC
52
15:15-17:15TutorialSanjay KaleUIUC
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100