Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed workflows”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

CMS distributed computing workflow experience

The vast majority of the CMS Computing capacity, which is organized in a tiered hierarchy, is located away from CERN. The 7 Tier-1 sites archive the LHC proton-proton collision data that is initially processed at CERN. These sites provide access to all recorded and simulated data for the Tier-2 sites, via wide-area network (WAN) transfers. All central data processing workflows are executed at the Tier-1 level, which contain re-reconstruction and skimming workflows of collision data as well as reprocessing of simulated data to adapt to changing detector conditions. This paper describes the operation of the CMS processing infrastructure at the Tier-1 level. The Tier-1 workflows are described in detail. The operational optimization of resource usage is described. In particular, the variation of different workflows during the data taking period of 2010, their efficiencies and latencies as well as their impact on the delivery of physics results is discussed and lessons are drawn from this experience. The simulation of proton-proton collisions for the CMS experiment is primarily carried out at the second tier of the CMS computing infrastructure. Half of the Tier-2 sites of CMS are reserved for central Monte Carlo (MC) production while the other half is available for user analysis. This paper summarizes the large throughput of the MC production operation during the data taking period of 2010 and discusses the latencies and efficiencies of the various types of MC production workflows. We present the operational procedures to optimize the usage of available resources and we the operational model of CMS for including opportunistic resources, such as the larger Tier-3 sites, into the central production operation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Modular performance prediction for scientific workflows using Machine Learning

Scientific workflows provide an opportunity for declarative computational experiment design in an intuitive and efficient way. A distributed workflow is typically executed on a variety of resources, and it uses a variety of computational algorithms or tools to achieve the desired outcomes. Such a variety imposes additional complexity in scheduling these workflows on large scale computers. As computation becomes more distributed, insights into expected workload that a workflow presents become critical for effective resource allocation. In this paper, we present a modular framework that leverages Machine Learning for creating precise performance predictions of a workflow. The central idea is to partition a workflow in such a way that makes the task of forecasting each atomic unit manageable and gives us a way to combine the individual predictions efficiently. We recognize a combination of an executable and a specific physical resource as a single module. This gives us a handle to characterize workload and machine power as a single unit of prediction. Overall, our modular technique of creating atomic modules and deployment of longest-path approach to estimate workflow performance, allows the framework to adapt to highly complex nested directed acyclic workflows and scale to new scenarios, since it does not make assumptions of underlying workflow structure. We present performance estimation results of independent workflow modules executed on the XSEDE SDSC Comet cluster using various Machine Learning algorithms. The results provide insights into the behavior and effectiveness of different algorithms in the context of scientific workflow performance prediction.

97 MATHEMATICS AND COMPUTING↗

Optimization of tomographic reconstruction workflows on geographically distributed resources

New technological advancements in synchrotron light sources enable data acquisitions at unprecedented levels. This emergent trend affects not only the size of the generated data but also the need for larger computational resources. Although beamline scientists and users have access to local computational resources, these are typically limited and can result in extended execution times. Applications that are based on iterative processing as in tomographic reconstruction methods require high-performance compute clusters for timely analysis of data. Here, time-sensitive analysis and processing of Advanced Photon Source data on geographically distributed resources are focused on. Two main challenges are considered: (i) modeling of the performance of tomographic reconstruction workflows and (ii) transparent execution of these workflows on distributed resources. For the former, three main stages are considered: (i) data transfer between storage and computational resources, (i) wait/queue time of reconstruction jobs at compute resources, and (iii) computation of reconstruction tasks. These performance models allow evaluation and estimation of the execution time of any given iterative tomographic reconstruction workflow that runs on geographically distributed resources. For the latter challenge, a workflow management system is built, which can automate the execution of workflows and minimize the user interaction with the underlying infrastructure. The system utilizes Globus to perform secure and efficient data transfer operations. The proposed models and the workflow management system are evaluated by using three high-performance computing and two storage resources, all of which are geographically distributed. Workflows were created with different computational requirements using two compute-intensive tomographic reconstruction algorithms. Experimental evaluation shows that the proposed models and system can be used for selecting the optimum resources, which in turn can provide up to 3.13× speedup (on experimented resources). Furthermore, the error rates of the models range between 2.1 and 23.3% (considering workflow execution times), where the accuracy of the model estimations increases with higher computational demands in reconstruction tasks.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Distributed Machine Learning Workflow with PanDA and iDDS in LHC ATLAS

Machine Learning (ML) has become one of the important tools for High Energy Physics analysis. As the size of the dataset increases at the Large Hadron Collider (LHC), and at the same time the search spaces become bigger and bigger in order to exploit the physics potentials, more and more computing resources are required for processing these ML tasks. In addition, complex advanced ML workflows are developed in which one task may depend on the results of previous tasks. How to make use of vast distributed CPUs/GPUs in WLCG for these big complex ML tasks has become a popular research area. In this paper, we present our efforts enabling the execution of distributed ML workflows on the Production and Distributed Analysis (PanDA) system and intelligent Data Delivery Service (iDDS). First, we describe how PanDA and iDDS deal with large-scale ML workflows, including the implementation to process workloads on diverse and geographically distributed computing resources. Next, we report real-world use cases, such as HyperParameter Optimization, Monte Carlo Toy confidence limits calculation, and Active Learning. Finally, we conclude with future plans.

97 MATHEMATICS AND COMPUTING↗

Software and computing for Run 3 of the ATLAS experiment at the LHC

The ATLAS experiment has developed extensive software and distributed computing systems for Run 3 of the LHC. These systems are described in detail, including software infrastructure and workflows, distributed data and workload management, database infrastructure, and validation. The use of these systems to prepare the data for physics analysis and assess its quality are described, along with the software tools used for data analysis itself. An outlook for the development of these projects towards Run 4 is also provided.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Towards Cross-Facility Workflows Orchestration through Distributed Automation

Modern science relies on end-to-end workflows that incorporate experimental instruments and utilize edge, cloud, or high-performance computing and storage resources. These components are geographically dispersed across various user facilities and interconnected through high-speed networks. In this paper, we present Zambeze, an automated distributed framework designed to facilitate this new class of cross-facility workflows. Utilizing swarm intelligence principles, Zambeze orchestrates science campaigns by managing distributed autonomous agents. These agents can offer a suite of services, including computing, storage, and data management. We demonstrate the feasibility of Zambeze through a real-world application involving electron microscopy, enhanced with Artificial Intelligence capabilities.

Skluzacek, Tyler↗

Automated Integration of Continental-Scale Observations in Near-Real Time for Simulation and Analysis of Biosphere–Atmosphere Interactions

The National Ecological Observatory Network (NEON) is a continental-scale observatory with sites across the US collecting standardized ecological observations that will operate for multiple decades. To maximize the utility of NEON data, we envision edge computing systems that gather, calibrate, aggregate, and ingest measurements in an integrated fashion. Edge systems will employ machine learning methods to cross-calibrate, gap-fill and provision data in near-real time to the NEON Data Portal and to High Performance Computing (HPC) systems, running ensembles of Earth system models (ESMs) that assimilate the data. For the first time gridded EC data products and response functions promise to offset pervasive observational biases through evaluating, benchmarking, optimizing parameters, and training new machine learning parameterizations within ESMs all at the same model-grid scale. Leveraging open-source software for EC data analysis, we are already building software infrastructure for integration of near-real time data streams into the International Land Model Benchmarking (ILAMB) package for use by the wider research community. We will present a perspective on the design and integration of end-to-end infrastructure for data acquisition, edge computing, HPC simulation, analysis, and validation, where Artificial Intelligence (AI) approaches are used throughout the distributed workflow to improve accuracy and computational performance.

Durden, David J.↗

CMS distributed data analysis with CRAB3

The CMS Remote Analysis Builder (CRAB) is a distributed workflow management tool which facilitates analysis tasks by isolating users from the technical details of the Grid infrastructure. Throughout LHC Run 1, CRAB has been successfully employed by an average of 350 distinct users each week executing about 200,000 jobs per day.CRAB has been significantly upgraded in order to face the new challenges posed by LHC Run 2. Components of the new system include 1) a lightweight client, 2) a central primary server which communicates with the clients through a REST interface, 3) secondary servers which manage user analysis tasks and submit jobs to the CMS resource provisioning system, and 4) a central service to asynchronously move user data from temporary storage in the execution site to the desired storage location. Furthermore, the new system improves the robustness, scalability and sustainability of the service.Here we provide an overview of the new system, operation, and user support, report on its current status, and identify lessons learned from the commissioning phase and production roll-out.

97 MATHEMATICS AND COMPUTING↗

Scientific Data Management Beyond Traditional Computing Boundaries

Scientific data management is undergoing a fundamental transformation driven by the convergence of artificial intelligence (AI)/machine learning workflows, distributed computing and storage environments, and exponential data growth. Here, we analyze how these developments address current limitations while enabling new capabilities for cross-facility collaboration and AI-driven research.

Widener, Patrick [Oak Ridge National Laboratory (O↗

ASCEM Data Brower (ASCEMDB) v0.8

Data management tool designed for the Advanced Simulation Capability for Environmental Management (ASCEM) framework. Distinguishing features of this gateway include: (1) handling of complex geometry data, (2) advance selection mechanism, (3) state of art rendering of spatiotemporal data records, and (4) seamless integration with a distributed workflow engine.

Romosan, Alexandru↗

An Executable Choreography Framework for Dynamic Service-Oriented Architectures

Interoperability and loose coupling requirements are pushing the next generation of distributed applications towards more decentralized and more dynamic interaction schemes, which the classic requestJresponse communication paradigm can hardly accommodate. Hence, sound foundations and mechanisms for the establishment of unmticiptitteb peer-to-peer interactions across organizational boundaries are of significant importance to upcoming middleware platforms. The Executable Choreography Framework (ECF) is a middleware-level framework that targets dynamic and decentralized service compositions. The ECF combines transparent context propagation with aspect-oriented software composition techniques to dynamically refine the default control and data flow of service invocations. The framework provides a ground for experimentation with dynamic and distributed workflows, and a base to assess their safety and applicability when depioyed across organizational boundaries.

Akkawi, Faisal↗

Provenance in Data Interoperability for Multi-Sensor Intercomparison

As our inventory of Earth science data sets grows, the ability to compare, merge and fuse multiple datasets grows in importance. This requires a deeper data interoperability than we have now. Efforts such as Open Geospatial Consortium and OPeNDAP (Open-source Project for a Network Data Access Protocol) have broken down format barriers to interoperability; the next challenge is the semantic aspects of the data. Consider the issues when satellite data are merged, cross-calibrated, validated, inter-compared and fused. We must match up data sets that are related, yet different in significant ways: the phenomenon being measured, measurement technique, location in space-time or quality of the measurements. If subtle distinctions between similar measurements are not clear to the user, results can be meaningless or lead to an incorrect interpretation of the data. Most of these distinctions trace to how the data came to be: sensors, processing and quality assessment. For example, monthly averages of satellite-based aerosol measurements often show significant discrepancies, which might be due to differences in spatio- temporal aggregation, sampling issues, sensor biases, algorithm differences or calibration issues. Provenance information must be captured in a semantic framework that allows data inter-use tools to incorporate it and aid in the intervention of comparison or merged products. Semantic web technology allows us to encode our knowledge of measurement characteristics, phenomena measured, space-time representation, and data quality attributes in a well-structured, machine-readable ontology and rulesets. An analysis tool can use this knowledge to show users the provenance-related distrintions between two variables, advising on options for further data processing and analysis. An additional problem for workflows distributed across heterogeneous systems is retrieval and transport of provenance. Provenance may be either embedded within the data payload, or transmitted from server to client in an out-of-band mechanism. The out of band mechanism is more flexible in the richness of provenance information that can be accomodated, but it relies on a persistent framework and can be difficult for legacy clients to use. We are prototyping the embedded model, incorporating provenance within metadata objects in the data payload. Thus, it always remains with the data. The downside is a limit to the size of provenance metadata that we can include, an issue that will eventually need resolution to encompass the richness of provenance information required for daata intercomparison and merging.

Lynnes, Chris↗

iDDS: intelligent distributed dispatch and scheduling for workflow orchestration

The intelligent distributed dispatch and scheduling (iDDS) service is a versatile workflow orchestration system designed for large-scale, distributed scientific computing. iDDS extends traditional workload and data management by integrating data-aware execution, conditional logic, and programmable workflows, enabling automation of complex and dynamic processing pipelines. Originally developed for the ATLAS experiment at the large hadron collider, iDDS has evolved into an experiment-agnostic platform that supports both template-driven workflows and a Function-as-a-Task model for Python-based orchestration. This paper presents the architecture and core components of iDDS, highlighting its scalability, modular message-driven design, and integration with systems such as PanDA and Rucio. We demonstrate its versatility through real-world use cases: fine-grained tape resource optimization for ATLAS, orchestration of large Directed Acyclic Graph (DAG) workflows for the Rubin Observatory, distributed hyperparameter optimization for machine learning applications, active learning for physics analyses, and AI-assisted detector design at the electron–ion collider. By unifying workload scheduling, data movement, and adaptive decision-making, iDDS reduces operational overhead and enables reproducible, high-throughput workflows across heterogeneous infrastructures. We conclude with current challenges and future directions, including interactive, cloud-native, and serverless workflow support.

97 MATHEMATICS AND COMPUTING↗

SWARM: Reimagining scientific workflow management systems in a distributed world

Modern scientific workflows process massive amounts of data from diverse instruments and sensors, leveraging geographically distributed, heterogeneous compute and storage resources—from leadership-class systems to edge devices—connected by high-performance networks. The diversity of resources introduces challenges in harnessing their full potential, with resilience issues arising across applications, system software, networks, storage, and hardware. Today, workflow management systems (WMS) coordinate the execution of computation and data management tasks across target resources. However, WMS’s centralized nature makes them vulnerable to faults and scalability issues that may result in failures of entire computational campaigns. In conclusion, this paper introduces a novel agentic framework for workflow management, fully distributing and decentralizing the WMS functions and modeling them as swarm intelligence agents infused with advanced artificial intelligence solutions and traditional distributed computing algorithms that can make coordinated decisions in the presence of failures of the underlying cyberinfrastructure.

Swarm intelligence↗

Building a Wide-Area File Transfer Performance Predictor: An Empirical Study

Wide-area data transfer is central to geographically distributed scientific workflows. Faster delivery of data is important for these workflows. Predictability is equally (or even more) important. With the goal of providing a reasonably accurate estimate of data transfer time to improve resource allocation & scheduling for workflows and enable end-to-end data transfer optimization, we apply machine learning methods to develop predictive models for data transfer times over a variety of wide area networks. To build and evaluate these models, we use 201,388 transfers, involving 759 million files totaling 9 PB transferred, over 115 heavily used source-destination pairs (“edges”) between 135 unique endpoints. We evaluate the models for different retraining frequencies and different window size of history data. In the best case, the resulting models have a median prediction error of ≤21% for 50% of the edges, and ≤32% for 75% of the edges. We present a detailed analysis of these results that provides insights into the cause of some of the high errors. We envision that the performance predictor will be informative for scheduling geo-distributed workflows. The insights also suggest obvious directions for both further analysis and transfer service optimization.

Liu, Zhengchun↗

Privacy by Design in Distributed Edge Systems: Innovating Secure Workflows for Smart Cities

The proliferation of distributed edge systems, such as those in smart cities, healthcare, and industrial IoT, offers unprecedented opportunities for data processing closer to its source, thereby reducing latency and enhancing efficiency. However, these systems also present significant privacy challenges due to the handling of sensitive data from multiple sources. This article explores the critical need for designing privacy-preserving workflows in distributed edge systems to ensure data security while maximizing the potential of edge computing. By examining the challenges, technological advancements, and potential of privacy-by-design approaches, we highlight the importance of integrating advanced privacy-preserving techniques like federated learning, differential privacy, homomorphic encryption, secure multi-party computation, and zero-knowledge proofs. These innovations are crucial for enhancing data security, regulatory compliance, and public trust in smart city applications, ultimately leading to safer and more efficient urban environments.

Kotevska, Olivera↗

Object Proxy Patterns for Accelerating Distributed Applications

Workflow and serverless frameworks have empowered new approaches to distributed application design by abstracting compute resources. However, their typically limited or one-size-fits-all support for advanced data flow patterns leaves optimization to the application programmer—optimization that becomes more difficult as data become larger. The transparent object proxy, which provides wide-area references that can resolve to data regardless of location, has been demonstrated as an effective low-level building block in such situations. Here we propose three high-level proxy-based programming patterns—distributed futures, streaming, and ownership—that make the power of the proxy pattern usable for more complex and dynamic distributed program structures. We motivate these patterns via careful review of application requirements and describe implementations of each pattern. As a result, we evaluate our implementations through a suite of benchmarks and by applying them in three meaningful scientific applications, in which we demonstrate substantial improvements in runtime, throughput, and memory usage.

Distributed Computing↗