arXiv subjects

Yadu Babuji

Publications and source records attributed to Yadu Babuji.

At least 19 recordsLinked to original sources

arXiv2026

ExaServe: Large-Scale LLM Serving on Exascale HPC Systems

Cloud-native LLM serving frameworks have made deployment routine in data centers, yet deploying them on leadership-class supercomputers remains an engineering challenge requiring scheduler integration, MPI launch, accelerator selection, node-local weight staging, and platform-specific patches. We present ExaServe, a pip-installable framework that transforms a declarative YAML specification into a reproducible large-scale LLM serving deployment. Using ExaServe, we deploy LLM serving on ALCF Aurora from 1 to 256 nodes (3072 vLLM replicas). Non-streaming inference scales nearly linearly to 256 nodes, reaching 27.1k requests/s (3.8M tokens/s). Token streaming scales differently: a centralized proxy plateaus at about 4.7k requests/s despite the model servers remaining within the service-level objective. We also identify an O(N^2) Ray Serve control-plane bottleneck that increases cluster bring-up to roughly 30 minutes at 256 nodes. ExaServe provides a practical, reproducible deployment path while exposing key barriers to future exascale LLM serving.

cs.DC↗

arXiv2026

Diamond Agent: Agentic Control of Federated HPC Resources as a Service

Efficiently aggregating and orchestrating computing power across heterogeneous clusters for HPC workflows faces four practical challenges: preserving workflow context across independently administered clusters, moving large datasets between sites, reasoning about site-specific environments and scheduler policies, and exploiting live queue and resource states for efficient task scheduling. To this end, we design Diamond Agent, an agentic system that enables intelligent execution of HPC workflows across heterogeneous clusters with typed skills as the interface. Diamond Agent provides an agent-facing workspace and skills that unify cross-site resource discovery, resource specification, data movement, task execution, and result retrieval. A centralized Diamond Agent instance can operate multiple supercomputers without being deployed separately on each login node. Diamond Agent translates high-level agent actions into valid site-specific executions, moves data through Globus Transfer, and uses live system capability and queue information to select feasible placements. Its event-driven continuation mechanism decouples agent actions from long-running batch jobs: persistent services monitor remote execution and resume the agent only when a result or decision-relevant event is available. We experiment with 27 hours of telemetry and 19 matched multi-site submission rounds comprising 83 jobs across four production supercomputers. Compared with a fixed-site baseline, Diamond Agent reduces the median additional completion time relative to the fastest observed placement from 42 seconds to 4 seconds, a 10.5x reduction.

cs.DC↗

arXiv2026

Scalable Agentic Reasoning for Designing Biologics Targeting Intrinsically Disordered Proteins

Intrinsically disordered proteins (IDPs) represent crucial therapeutic targets due to their significant role in disease -- approximately 80\% of cancer-related proteins contain long disordered regions -- but their lack of stable secondary/tertiary structures makes them "undruggable". While recent computational advances, such as diffusion models, can design high-affinity IDP binders, translating these to practical drug discovery requires autonomous systems capable of reasoning across complex conformational ensembles and orchestrating diverse computational tools at scale.To address this challenge, we designed and implemented StructBioReasoner, a scalable multi-agent system for designing biologics that can be used to target IDPs. StructBioReasoner employs a novel tournament-based reasoning framework where specialized agents compete to generate and refine therapeutic hypotheses, naturally distributing computational load for efficient exploration of the vast design space. Agents integrate domain knowledge with access to literature synthesis, AI-structure prediction, molecular simulations, and stability analysis, coordinating their execution on HPC infrastructure via an extensible federated agentic middleware, Academy. We benchmark StructBioReasoner across Der f 21 and NMNAT-2 and demonstrate that over 50\% of 787 designed and validated candidates for Der f 21 outperformed the human-designed reference binders from literature, in terms of improved binding free energy. For the more challenging NMNAT-2 protein, we identified three binding modes from 97,066 binders, including the well-studied NMNAT2:p53 interface. Thus, StructBioReasoner lays the groundwork for agentic reasoning systems for IDP therapeutic discovery on Exascale platforms.

q-bio.QM↗

arXiv2026

Empowering Scientific Workflows with Federated Agents

Agentic systems, in which diverse agents cooperate to tackle challenging problems, are exploding in popularity in the AI community. However, existing agentic frameworks take a relatively narrow view of agents, apply a centralized model, and target conversational, cloud-native applications (e.g., LLM-based AI chatbots). In contrast, scientific applications require myriad agents be deployed and managed across diverse cyberinfrastructure. Here we introduce Academy, a modular and extensible middleware designed to deploy autonomous agents across the federated research ecosystem, including HPC systems, experimental facilities, and data repositories. To meet the demands of scientific computing, Academy supports asynchronous execution, heterogeneous resources, high-throughput data flows, and dynamic resource availability. It provides abstractions for expressing stateful agents, managing inter-agent coordination, and integrating computation with experimental control. We present microbenchmark results that demonstrate high performance and scalability in HPC environments. To explore the breadth of applications that can be supported by agentic workflow designs, we also present case studies in materials discovery, astronomy, decentralized learning, and information extraction in which agents are deployed across diverse HPC systems.

cs.MA↗

arXiv2026

LSHBloom: Memory-efficient, Extreme-scale Document Deduplication

Contemporary large language model (LLM) training pipelines require the assembly of internet-scale databases full of text data from a variety of sources (e.g., web, academic, and publishers). Preprocessing these datasets via deduplication -- detecting and eliminating additional instances of the same content -- is a major focus for assembling and curating training datasets for LLMs. Unrestrained, duplicates in the training dataset increase training costs and lead to undesirable properties such as memorization in trained models or cheating on evaluation. Unfortunately, contemporary approaches to document-level deduplication are either unreliable at accurately identifying duplicate documents or extremely expensive in terms of both runtime and memory. We propose LSHBloom, an extension to MinhashLSH, which replaces the expensive LSHIndex with lightweight Bloom filters. LSHBloom demonstrates the same state-of-the-art deduplication performance as MinhashLSH, with only a marginal increase in false positives (near zero in our experiments), while boasting competitive runtime (12$\times$ faster than MinhashLSH on peS2o) and, crucially, using 18$\times$ less disk space than MinhashLSH (as measured on peS2o). Based on extrapolation, we show that this advantage in space and runtime remains even at the extreme scale of several billion documents. LSHBloom allows practitioners to access the deduplication quality of MinHashLSH at scales that are normally only tractable for less sophisticated, heuristic solutions. As a result, LSHBloom promises to enable scaling high-quality document deduplication to internet-scale text datasets.

cs.LG↗

arXiv2025

AERO: An autonomous platform for continuous research

The COVID-19 pandemic highlighted the need for new data infrastructure, as epidemiologists and public health workers raced to harness rapidly evolving data, analytics, and infrastructure in support of cross-sector investigations. To meet this need, we developed AERO, an automated research and data sharing platform for continuous, distributed, and multi-disciplinary collaboration. In this paper, we describe the AERO design and how it supports the automatic ingestion, validation, and transformation of monitored data into a form suitable for analysis; the automated execution of analyses on this data; and the sharing of data among different entities. We also describe how our AERO implementation leverages capabilities provided by the Globus platform and GitHub for automation, distributed execution, data sharing, and authentication. We present results obtained with an instance of AERO running two public health surveillance applications and demonstrate benchmarking results with a synthetic application, all of which are publicly available for testing.

cs.CE↗

arXiv2025

AdaParse: An Adaptive Parallel PDF Parsing and Resource Scaling Engine

Language models for scientific tasks are trained on text from scientific publications, most distributed as PDFs that require parsing. PDF parsing approaches range from inexpensive heuristics (for simple documents) to computationally intensive ML-driven systems (for complex or degraded ones). The choice of the "best" parser for a particular document depends on its computational cost and the accuracy of its output. To address these issues, we introduce an Adaptive Parallel PDF Parsing and Resource Scaling Engine (AdaParse), a data-driven strategy for assigning an appropriate parser to each document. We enlist scientists to select preferred parser outputs and incorporate this information through direct preference optimization (DPO) into AdaParse, thereby aligning its selection process with human judgment. AdaParse then incorporates hardware requirements and predicted accuracy of each parser to orchestrate computational resources efficiently for large-scale parsing campaigns. We demonstrate that AdaParse, when compared to state-of-the-art parsers, improves throughput by $17\times$ while still achieving comparable accuracy (0.2 percent better) on a benchmark set of 1000 scientific documents. AdaParse's combination of high accuracy and parallel scalability makes it feasible to parse large-scale scientific document corpora to support the development of high-quality, trillion-token-scale text datasets. The implementation is available at https://github.com/7shoe/AdaParse/

cs.IR↗

arXiv2024

Parsl+CWL: Towards Combining the Python and CWL Ecosystems

The Common Workflow Language (CWL) is a widely adopted language for defining and sharing computational workflows. It is designed to be independent of the execution engine on which workflows are executed. In this paper, we describe our experiences integrating CWL with Parsl, a Python-based parallel programming library designed to manage execution of workflows across diverse computing environments. We propose a new method that converts CWL CommandLineTool definitions into Parsl apps, enabling Parsl scripts to easily import and use tools represented in CWL. We describe a Parsl runner that is capable of executing a CWL CommandLineTool directly. We also describe a proof-of-concept extension to support inline Python in a CWL workflow definition, enabling seamless use in the Python ecosystem of Parsl. We demonstrate the benefits of this integration by presenting example CWL CommandLineTool definitions that show how they can be used in Parsl, and comparing performance of executing an image processing workflow using the Parsl integration and other CWL runners.

cs.DC↗

arXiv2024

Workflows Community Summit 2024: Future Trends and Challenges in Scientific Workflows

The Workflows Community Summit gathered 111 participants from 18 countries to discuss emerging trends and challenges in scientific workflows, focusing on six key areas: time-sensitive workflows, AI-HPC convergence, multi-facility workflows, heterogeneous HPC environments, user experience, and FAIR computational workflows. The integration of AI and exascale computing has revolutionized scientific workflows, enabling higher-fidelity models and complex, time-sensitive processes, while introducing challenges in managing heterogeneous environments and multi-facility data dependencies. The rise of large language models is driving computational demands to zettaflop scales, necessitating modular, adaptable systems and cloud-service models to optimize resource utilization and ensure reproducibility. Multi-facility workflows present challenges in data movement, curation, and overcoming institutional silos, while diverse hardware architectures require integrating workflow considerations into early system design and developing standardized resource management tools. The summit emphasized improving user experience in workflow systems and ensuring FAIR workflows to enhance collaboration and accelerate scientific discovery. Key recommendations include developing standardized metrics for time-sensitive workflows, creating frameworks for cloud-HPC integration, implementing distributed-by-design workflow modeling, establishing multi-facility authentication protocols, and accelerating AI integration in HPC workflow management. The summit also called for comprehensive workflow benchmarks, workflow-specific UX principles, and a FAIR workflow maturity model, highlighting the need for continued collaboration in addressing the complex challenges posed by the convergence of AI, HPC, and multi-facility research environments.

cs.DC↗

arXiv2024

Flight: A FaaS-Based Framework for Complex and Hierarchical Federated Learning

Federated Learning (FL) is a decentralized machine learning paradigm where models are trained on distributed devices and are aggregated at a central server. Existing FL frameworks assume simple two-tier network topologies where end devices are directly connected to the aggregation server. While this is a practical mental model, it does not exploit the inherent topology of real-world distributed systems like the Internet-of-Things. We present Flight, a novel FL framework that supports complex hierarchical multi-tier topologies, asynchronous aggregation, and decouples the control plane from the data plane. We compare the performance of Flight against Flower, a state-of-the-art FL framework. Our results show that Flight scales beyond Flower, supporting up to 2048 simultaneous devices, and reduces FL makespan across several models. Finally, we show that Flight's hierarchical FL model can reduce communication overheads by more than 60%.

cs.LG↗

arXiv2024

Employing Artificial Intelligence to Steer Exascale Workflows with Colmena

Computational workflows are a common class of application on supercomputers, yet the loosely coupled and heterogeneous nature of workflows often fails to take full advantage of their capabilities. We created Colmena to leverage the massive parallelism of a supercomputer by using Artificial Intelligence (AI) to learn from and adapt a workflow as it executes. Colmena allows scientists to define how their application should respond to events (e.g., task completion) as a series of cooperative agents. In this paper, we describe the design of Colmena, the challenges we overcame while deploying applications on exascale systems, and the science workflows we have enhanced through interweaving AI. The scaling challenges we discuss include developing steering strategies that maximize node utilization, introducing data fabrics that reduce communication overhead of data-intensive tasks, and implementing workflow tasks that cache costly operations between invocations. These innovations coupled with a variety of application patterns accessible through our agent-based steering model have enabled science advances in chemistry, biophysics, and materials science using different types of AI. Our vision is that Colmena will spur creative solutions that harness AI across many domains of scientific computing.

cs.DC↗

arXiv2024

GreenFaaS: Maximizing Energy Efficiency of HPC Workloads with FaaS

Application energy efficiency can be improved by executing each application component on the compute element that consumes the least energy while also satisfying time constraints. In principle, the function as a service (FaaS) paradigm should simplify such optimizations by abstracting away compute location, but existing FaaS systems do not provide for user transparency over application energy consumption or task placement. Here we present GreenFaaS, a novel open source framework that bridges this gap between energy-efficient applications and FaaS platforms. GreenFaaS can be deployed by end users or providers across systems to monitor energy use, provide task-specific feedback, and schedule tasks in an energy-aware manner. We demonstrate that intelligent placement of tasks can both reduce energy consumption and improve performance. For a synthetic workload, GreenFaaS reduces the energy-delay product by 45% compared to alternatives. Furthermore, running a molecular design application through GreenFaaS can reduce energy consumption by 21% and runtime by 63% by better matching tasks with machines.

cs.DC↗

arXiv2024

UniFaaS: Programming across Distributed Cyberinfrastructure with Federated Function Serving

Modern scientific applications are increasingly decomposable into individual functions that may be deployed across distributed and diverse cyberinfrastructure such as supercomputers, clouds, and accelerators. Such applications call for new approaches to programming, distributed execution, and function-level management. We present UniFaaS, a parallel programming framework that relies on a federated function-as-a-service (FaaS) model to enable composition of distributed, scalable, and high-performance scientific workflows, and to support fine-grained function-level management. UniFaaS provides a unified programming interface to compose dynamic task graphs with transparent wide-area data management. UniFaaS exploits an observe-predict-decide approach to efficiently map workflow tasks to target heterogeneous and dynamic resources. We propose a dynamic heterogeneity-aware scheduling algorithm that employs a delay mechanism and a re-scheduling mechanism to accommodate dynamic resource capacity. Our experiments show that UniFaaS can efficiently execute workflows across computing resources with minimal scheduling overhead. We show that UniFaaS can improve the performance of a real-world drug screening workflow by as much as 22.99% when employing an additional 19.48% of resources and a montage workflow by 54.41% when employing an additional 47.83% of resources across multiple distributed clusters, in contrast to using a single cluster

cs.DC↗

arXiv2023

The Changing Role of RSEs over the Lifetime of Parsl

This position paper describes the Parsl open source research software project and its various phases over seven years. It defines four types of research software engineers (RSEs) who have been important to the project in those phases; we believe this is also applicable to other research software projects.

cs.SE↗

arXiv2023

Developing Distributed High-performance Computing Capabilities of an Open Science Platform for Robust Epidemic Analysis

COVID-19 had an unprecedented impact on scientific collaboration. The pandemic and its broad response from the scientific community has forged new relationships among domain experts, mathematical modelers, and scientific computing specialists. Computationally, however, it also revealed critical gaps in the ability of researchers to exploit advanced computing systems. These challenging areas include gaining access to scalable computing systems, porting models and workflows to new systems, sharing data of varying sizes, and producing results that can be reproduced and validated by others. Informed by our team's work in supporting public health decision makers during the COVID-19 pandemic and by the identified capability gaps in applying high-performance computing (HPC) to the modeling of complex social systems, we present the goals, requirements, and initial implementation of OSPREY, an open science platform for robust epidemic analysis. The prototype implementation demonstrates an integrated, algorithm-driven HPC workflow architecture, coordinating tasks across federated HPC resources, with robust, secure and automated access to each of the resources. We demonstrate scalable and fault-tolerant task execution, an asynchronous API to support fast time-to-solution algorithms, an inclusive, multi-language approach, and efficient wide-area data management. The example OSPREY code is made available on a public repository.

cs.DC↗

arXiv2023

Cloud Services Enable Efficient AI-Guided Simulation Workflows across Heterogeneous Resources

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.

cs.DC↗

arXiv2022

funcX: Federated Function as a Service for Science

funcX is a distributed function as a service (FaaS) platform that enables flexible, scalable, and high performance remote function execution. Unlike centralized FaaS systems, funcX decouples the cloud-hosted management functionality from the edge-hosted execution functionality. funcX's endpoint software can be deployed, by users or administrators, on arbitrary laptops, clouds, clusters, and supercomputers, in effect turning them into function serving systems. funcX's cloud-hosted service provides a single location for registering, sharing, and managing both functions and endpoints. It allows for transparent, secure, and reliable function execution across the federated ecosystem of endpoints--enabling users to route functions to endpoints based on specific needs. funcX uses containers (e.g., Docker, Singularity, and Shifter) to provide common execution environments across endpoints. funcX implements various container management strategies to execute functions with high performance and efficiency on diverse funcX endpoints. funcX also integrates with an in-memory data store and Globus for managing data that may span endpoints. We motivate the need for funcX, present our prototype design and implementation, and demonstrate, via experiments on two supercomputers, that funcX can scale to more than 130 000 concurrent workers. We show that funcX's container warming-aware routing algorithm can reduce the completion time for 3000 functions by up to 61% compared to a randomized algorithm and the in-memory data store can speed up data transfers by up to 3x compared to a shared file system.

cs.DC↗

arXiv2022

Extended Abstract: Productive Parallel Programming with Parsl

Parsl is a parallel programming library for Python that aims to make it easy to specify parallelism in programs and to realize that parallelism on arbitrary parallel and distributed computing systems. Parsl relies on developers annotating Python functions-wrapping either Python or external applications-to indicate that these functions may be executed concurrently. Developers can then link together functions via the exchange of data. Parsl establishes a dynamic dependency graph and sends tasks for execution on connected resources when dependencies are resolved. Parsl's runtime system enables different compute resources to be used, from laptops to supercomputers, without modification to the Parsl program.

cs.DC↗