RECORD
Cloud Services Enable Efficient AI-Guided Simulation Workflows across Heterogeneous Resources
Distributed, Parallel, and Cluster Computing · Submitted 15 Mar 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim We show that these methods can achieve performance parity with systems that rely on direct connection between resources.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Applications that fuse machine learning and simulation can benefit from the use of multiple computing resources, with, for example, simulation codes running on highly parallel supercomputers and AI training and inference tasks on specialized accelerators. Here, we present our experiences deploying two AI-guided simulation workflows across such heterogeneous systems. A unique aspect of our approach is our use of cloud-hosted management services to manage challenging aspects of cross-resource authentication and authorization, function-as-a-service (FaaS) function invocation, and data transfer. We show that these methods can achieve performance parity with systems that rely on direct connection between resources. We achieve parity by integrating the FaaS system and data transfer capabilities with a system that passes data by reference among managers and workers, and a user-configurable steering algorithm to hide data transfer latencies. We anticipate that this ease of use can enable routine use of heterogeneous resources in computational science.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Applications that fuse machine learning and simulation can benefit from the use of multiple computing resources, with, for example, simulation codes running on highly parallel supercomputers and AI training and inference tasks on specialized accelerators. | background 1.00 | 0.06 | 0.04 | hedged possibility 0.86 |
| s1 | Here, we present our experiences deploying two AI-guided simulation workflows across such heterogeneous systems. | contribution 1.00 | 0.92 | 0.05 | asserted 0.77 |
| s2 | A unique aspect of our approach is our use of cloud-hosted management services to manage challenging aspects of cross-resource authentication and authorization, function-as-a-service (FaaS) function invocation, and data transfer. | contribution 0.76 | 0.89 | 0.09 | asserted 0.98 |
| s3 | We show that these methods can achieve performance parity with systems that rely on direct connection between resources. | result 1.00 | 0.93 | 0.47 | asserted 0.96 |
| s4 | We achieve parity by integrating the FaaS system and data transfer capabilities with a system that passes data by reference among managers and workers, and a user-configurable steering algorithm to hide data transfer latencies. | method 0.98 | 0.94 | 0.17 | asserted 1.00 |
| s5 | We anticipate that this ease of use can enable routine use of heterogeneous resources in computational science. | interpretation 0.76 | 0.08 | 0.33 | hedged possibility 0.55 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Applications that fuse machine learning and simulation can benefit from the use of multiple computing resources, with, for example, simulation codes running on highly parallel supercomputers and AI training and inference tasks on specialized accelerators. Here, we present our experiences deploying two AI-guided simulation workflows across such heterogeneous systems. A unique aspect of our approach is our use of cloud-hosted management services to manage challenging aspects of cross-resource authentication and authorization, function-as-a-service (FaaS) function invocation, and data transfer. We show that these methods can achieve performance parity with systems that rely on direct connection between resources. We achieve parity by integrating the FaaS system and data transfer capabilities with a system that passes data by reference among managers and workers, and a user-configurable steering algorithm to hide data transfer latencies. We anticipate that this ease of use can enable routine use of heterogeneous resources in computational science.
Human review
How the literature uses it
3 arXiv papers cite this in 4 sentences
What it cites
Uses data or values from 1 paper
-
This application seeks to identify molecules within a candidate set (here, 1115321 molecules from the MOSES dataset [ref] , as collected in the nCov-Group Data Repository [ref] ) that have desirable properties.
Targeting SARS-CoV-2 with AI- and HPC-enabled Lead Generation: A First Data Release · Application: Molecular Design · 1.00
Cites as background 1 paper
-
Following the lexicon of Jha et al. [ref] , there are at least “ML-in-HPC” applications where AI-based software are used inside conventional HPC applications (e.g., surrogates for expensive computational routines [ref] ) and “ML-out-HPC” where machine learning controls the execution of the application (e.g., steering a workflow via active learning [ref] ).
AI-coupled HPC Workflows · AI-Integrated Workflows · 0.82
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.