Impact of Middleware Design on the Communication Performance

The precise architectural approach of a communication middleware technology is of outmost importance in areas where a certain degree of control over the underlying platform is needed, e.g., for fine-grain management of temporal behavior or concurrency management. The design of the middleware libraries and, especially, the resulting execution environment that these provide affect the control level of the application code over the platform resources. In this paper, we describe and analyze the two main design approaches of communication middleware implementations, that offer two different execution environments to applications: direct execution and virtual machine intermediation. We exemplify this idea with a selected set of technologies that does not intend to be exhaustive, but simply representative of both possible middleware architectural design approaches. The goal is to show the impact of these architectures on the resulting communication performance, that we show as overhead, i.e., the percentage of time taken by both architectures with respect to a fixed remote operation with a fixed processing time.

[1]  Marisol García-Valls,et al.  Improving service time with a multicore aware middleware , 2017, SAC.

[2]  Aniruddha S. Gokhale,et al.  DOORS: towards high-performance fault tolerant CORBA , 2000, Proceedings DOA'00. International Symposium on Distributed Objects and Applications.

[3]  Marisol García-Valls,et al.  Adaptive middleware design for CPS: Considerations on the OS, resource managers, and the network run-time , 2015, ARM@Middleware.

[4]  Chenyang Lu,et al.  Reconfigurable Real-Time Middleware for Distributed Cyber-Physical Systems with Aperiodic Events , 2008, 2008 The 28th International Conference on Distributed Computing Systems.

[5]  Chenyang Lu,et al.  Real-Time Performance and Middleware for Multiprocessor and Multicore Linux Platforms , 2009, 2009 15th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications.

[6]  Marisol García-Valls,et al.  An Architecture of a Quality of Service Resource Manager Middleware for Flexible Embedded Multimedia Systems , 2002, SEM.

[7]  Tommaso Cucinotta,et al.  Challenges in real-time virtualization and predictable cloud computing , 2014, J. Syst. Archit..

[8]  Stefano Russo,et al.  Performance assessment of OMG compliant data distribution middleware , 2008, 2008 IEEE International Symposium on Parallel and Distributed Processing.

[9]  Marisol Garcia Valls,et al.  iLAND: An Enhanced Middleware for Real-Time Reconfiguration of Service Oriented Distributed Real-Time Systems , 2013 .

[10]  Marc Mercuri,et al.  Windows Communication Foundation Unleashed (WCF) (Unleashed) , 2007 .

[11]  Marisol García-Valls,et al.  Analyzing point-to-point DDS communication over desktop virtualization software , 2017, Comput. Stand. Interfaces.

[12]  José M. Troya,et al.  Development of distributed real-time simulators based on CORBA , 2007, Simul. Model. Pract. Theory.

[13]  Marisol García-Valls,et al.  Using DDS middleware in distributed partitioned systems , 2018, SIGBED.

[14]  Tobias Schüle,et al.  Work Stealing Strategies for Parallel Stream Processing in Soft Real-Time Systems , 2012, ARCS.