Avatar improves scientific workflows using large language models
Avatar: Toward Autonomous End-to-End Orchestration of Scientific Workflows using LLMs
Distributed, Parallel, and Cluster ComputingMultiagent Systems
Summary
Scientific workflows are a way to automate complex research tasks, but traditional systems use fixed rules that don’t adapt well. The authors introduce Avatar, a new system that can control these workflows using either traditional rules or smart language models called LLMs. Avatar can switch between these methods using a shared framework, which lets it try smarter decision-making without breaking the usual process. Tests show that using LLMs with Avatar reduces wasted computing and speeds up GPU usage, suggesting smarter automation can make research computing more efficient.
scientific workflow managementorchestrationlarge language modelsLLM agentsactor-based architectureprovenance monitorrule-based controlcompute wastageGPU usageAcademy framework
Authors
Suman Raj, Hai Duc Nguyen, Haochen Pan, Ryan Chard, Kyle Chard, Ian Foster
Abstract
Scientific workflow management (WMSs) systems automate execution, yet orchestrate using fixed, hand-tuned rules. LLM agents promise more autonomous orchestration, but it remains unclear where to introduce agentic reasoning, how to bound its risk, and when it actually helps. We present Avatar, an actor-based architecture comprising an orchestrator, an executor, and a provenance monitor. Each actor's decision policy is pluggable (rule-based or LLM-backed) via a single adapter-validated action catalog, so conventional and agentic control run on the same core across different WMSs. We present an implementation using the Academy framework and evaluate Avatar across three workloads. We observe that Avatar's rule mode reproduces native execution, with a single unchanged core running all three. Moreover, LLM-backed Avatar reports a reduction of compute wastage by $55\%$ and cuts GPU-busy time by $40\%$. Overall, we envision Avatar as a step toward workflow systems that reason about their own orchestration rather than follow pre-fixed rules.