Title :
A Highly Concurrent Process Virtual Machine Based on Event-driven Process Execution Model
Author :
Dongyao Wu ; Jun Wei ; Chushu Gao ; Wenshen Dou
Author_Institution :
Inst. of Software, Beijing, China
Abstract :
Existing orchestration and choreography process engines only serve for dedicate process languages and majority of them show rough performance under even moderate workloads which cause them incapable for practical use. To solve the problems, the Event-driven Process Execution Model (EPEM) is presented as the common process model and the formalization of EPEM is presented to guarantee the correctness and efficiency for process transformation. A Process Virtual Machine - Once PVM is implemented based on EPEM and extended to support execution of WS-BPEL. For evaluation, the performance of Once PVM is compared with existing orchestration engines including Active BPEL, Apache ODE and Oracle BPEL Process Manager. The experimental results indicate that Once PVM shows better scalability than other tested platforms especially under high workloads.
Keywords :
concurrency control; virtual machines; Active BPEL; Apache ODE; EPEM; Once PVM; Oracle BPEL process manager; Process Virtual Machine; choreography process engines; event-driven process execution model; highly concurrent process virtual machine; orchestration engines; orchestration process engines; process languages; Connectors; Context; Engines; Message systems; Runtime; Semantics; Virtual machining; BPEL; Business process modeling; event-driven; process virtual machine; workflow;
Conference_Titel :
e-Business Engineering (ICEBE), 2012 IEEE Ninth International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-2601-8
DOI :
10.1109/ICEBE.2012.20