Abstract. Emerging methods for enterprise systems analysis rely on the representation of work practices in the form of business process models. A standard for representing such models is the Business Process Modeling Notation (BPMN). BPMN models are mainly intended for communication and decision-making between domain analysts, but often they are also given as input to software development projects. Meanwhile, development methods for process-oriented systems rely on detailed process definitions that are executed by process engines. These process definitions refine BPMN models by adding data manipulation, application binding and other implementation details. A major standard for process implementation is the Business Process Execution Language for Web Services (BPEL4WS, or BPEL for short). Accordingly, a natural method for end-to-end development of process-oriented systems is to translate BPMN models to BPEL definitions for subsequent refinement. However, instrumenting this method is challenging because BPEL imposes far more syntactic restrictions than BPMN so as to ensure correctness. Existing techniques for translating BPMN to BPEL only work for limited classes of BPMN models. This paper proposes techniques that overcome these limitations. Beyond its direct relevance in the context of BPMN and BPEL, the techniques presented in this paper address issues that arise generally when translating from graphical/unstructured to textual/structured (i.e. more programming-like) languages. 1

Portable parallel benchmarks are widely used and highly effective for (a) the evaluation, analysis and procurement of high-performance computing (HPC) systems and (b) quantifying the potential benefits of porting applications for new hardware platforms. Yet, past techniques to synthetically parametrized hand-coded HPC benchmarks prove insufficient for today’s rapidly-evolving scientific codes particularly when subject to multi-scale science modeling or when utilizing domain-specific libraries. To address these problems, this work contributes novel methods to automatically generate highly portable and customizable communication benchmarks from HPC applications. We utilize ScalaTrace, a lossless, yet scalable, parallel application tracing framework to collect selected aspects of the run-time behavior of HPC applications, including communication operations and execution time, while abstracting