Hypergraph-Based Data Reduced Scheduling Policy for Data-Intensive Workflow in Clouds

Hu, Zhigang; Li, Jia; Zheng, Meiguang; Zhang, Xinxin; Kang, Hui; Tao, Yong; Jiao, Yang

doi:10.1007/978-981-10-6388-6_28

Cited by 1 publication

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dependence on XML, SOAP, and WSDL makes adapting BPEL to control systems modeling difficult and inefficient. Research has studied the directed hypergraphs based representation for workflows [38], scheduling [6], and resource allocation [39]. However, such representations limit their focus to business processes rather than scientific workflows [40].…”

Section: Business Process Workflowsmentioning

confidence: 99%

“…Enabling communication and coordination across existing workflows to compose a complex workflow is currently not a trivial undertaking due to incompatibility across workflow languages and the lack of an orchestrator that spans multiple scientific workflow frameworks. Furthermore, a workflow composed of several workflows can be more flexibly represented by a directed hypergraph (DHG) than a DG or a DAG [6]. However, such a dynamic representation is hindered by workflow definitions that tightly couple how the data flows between the services and the control of the services that compose the workflow.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Software-Defined Workflows for Distributed Interoperable Closed-Loop Neuromodulation Control Systems

Kathiravelu

Sarikhani

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Closed-loop neuromodulation control systems facilitate regulating abnormal physiological processes by recording neurophysiological activities and modifying those activities through feedback loops. Designing such systems will require closed-loop workflows with interoperable modules. Workflow frameworks enable standard modular designs, offering reproducible automated pipelines. But workflow languages and standards limit their aim to executions represented by directed acyclic graphs (DAGs). DAGs require a pre-defined start and end execution step with no cycles, thus preventing the researchers from using the standard workflow languages as-is for neuromodulation control systems which consist of closed-loops. Testing and prototyping closed-loop neuromodulation control systems require in-silico simulations before being integrated into in-vivo experimental setups. In this paper, we present NEXUS, our workflow orchestration framework for distributed analytics systems. NEXUS proposes a Software-Defined Workflows approach, inspired by Software-Defined Networking (SDN), which separates the data flows across the service instances from the control flows. NEXUS enables creating interoperable workflows with closed loops by dynamically defining the workflows in a logically centralized approach, from microservices representing each execution step. The centralized orchestrator of NEXUS facilitates dynamically composing scientific workflows from the service nodes with minimal restrictions. NEXUS further supports composing and managing complex workflows from existing workflows via its orchestrator by representing the workflows as directed hypergraphs (DHGs) rather than DAGs. We build sample workflows for a few closed-loop neuromodulation control systems with NEXUS. We illustrate a seamless execution of neuromodulation control systems by supporting loops in a workflow as the use case of NEXUS. Our evaluations highlight the feasibility, flexibility, performance, and scalability of NEXUS in modeling and executing workflows with closed-loops.

show abstract

Section: Business Process Workflowsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%