wiki:public/20121017

Project: High Performance Workflow

Team: Fernando Antonio Vanini, Luiz Eduardo Buzato, Leonardo Garcia Tampelini, Rodrigo Marchi

Research institution: UNICAMP - Brazil

Abstract:

In a typical business process automation environment, enterprise software operates in conjunction with a BPM suite. In such situations, the access to business data combined with process data cannot be done efficiently. Business data usually follows a carefully designed data model. By contrast, process data usually follows a less structured model in order to support the diversity and dynamicity of processes. Under these conditions, the access to process data in conjunction to business data can impact negatively the performance of a business automation system. This is more critical as the number of process instances or the volume of process data increases.

Overall structure:

  • KPM is the process execution engine that runs process instances based on Process Models created by a graphical process editor (KPE).
  • Process Interpreter is the part of KPM which performs all operations related to process instances.
  • Data Access Layer is the part of KPM responsible for instance data persistence.
  • Load Simulator generates the workload simulating the actions performed by users and applications that interact with the processes instances running under the KPM.

Planned activities:

  • Load Simulator construction: this module will be built based on KPS, a process simulator oriented to business performance analysis and optimization.
  • Data Access Layer adapting: the Data Access Layer module should be adapted in order to operate with HANA in memory database. Current version can operate with Oracle, SQLServer and Postgresql databases.
  • Simulations: for each simulation scenario, a simulation run should be done for each one of the databases.
  • Simulation results analysis.
Last modified 5 years ago Last modified on Aug 21, 2014 1:31:17 PM