KSU CCM Workshop

May 19-20, 2003

Small Picture of Nichols

Location: KSU CIS Department -- Nichols Hall Room 122

(KSU Campus Map -- note: Ramada appears on map at corner of 17th and Anderson)

Monday Morning: CCM and Cadena

8:00 - 8:30 Coffee, light fruit and snacks
8:30 - 8:45: Greeting and Goals

Speaker: John/Doug

8:45 - 9:30: Component development and system configuration in Cadena and OpenCCM (.ppt)

Speaker: John/Jesse
Goals: Present the CIAO team with an overview of how Cadena currently supports component development and configuration using the Cadena GUI and the OpenCCM infrastructure. Example driven presentation using the simple Boeing OEP scenarios (e.g. BasicSP, ModalSP). Expose enough of the details concerning IDL, code generation, and configuration to lead into the following discussions on reconciling the Boeing OEP to CCM.
Supporting artifacts:

Cadena example repository
BasicSP artifacts from example repository
Cadena tutorial
Cadena component development tutorial

9:30 - 10:00: Reconciling the Boeing OEP with CCM -- component interface issues (.ppt)

Speaker: John
Goals: Present issues that arise when trying to capture the Boeing OEP components (and associated RT properties) at the component interface level. Make some progress on agreeing on how to resolve these so that we can maintain a common repository of OEP scenarios in CCM, and provide guidance to Boeing team for evolving toward CCM.
Supporting artifacts:

Boeing OEP code for BM_PushPullComponentImpl.cpp
From OEP to CCM en route CAD (.pdf ) (Venkatesh Ranganath)
E-mail exchanges with Boeing people

--- Break ---

10:30 - 11:00 Reconciling the Boeing OEP with CCM -- configuration issues (.ppt)

Speaker: Venkatesh
Goals: Present issues that arise when trying describe the configuration of Boeing OEP scenarios (and other DRE systems) using the standard CCM configuration formats. In particular, convention CCM does not have direct mechanisms for describing various RT and QoS attributes. An important goal is to move toward a proposal for a common configuration format that CIAO and Cadena can agree on.
Supporting artifacts:

(see Boeing OEP scenario examples above)
From OEP to CCM en route CAD (.pdf ) (Venkatesh Raganath)

11:00 - 11:45 Integrating the RT Event Channel with CCM

Speaker: Masaaki/Gurdip
Goals: Summarize the structure of CCM components and containers, propose mechanism for including RT-event channel, report on prototype experiments.
Supporting artifacts:

OpenCCM Demo3 example code

--- Lunch: Aggieville ---

Monday Afternoon: CCM and CIAO

1:15 - 2:00 CIDL Overview

Speaker: Jeff Parsons
Goals: purpose of CIDL, example-driven presentation

2:00 - 2:30 Component Development in CIAO

Speaker: Bala
Goals: brief summary of how components from BasicSP are developed with CIAO, help KSU people understand how to use the current version of Ciao

2:30 - 3:00 Packaging and Deployment in CIAO

Speaker: Kitty
Goals: reminder about CCM packaging and deployment, current implement and CIAO and future plans. Highlighting of issues that Cadena will need to address to support full CCM deployment facilities.

--- Break ---

3:30 - 4:00 QoS issues in CIAO

Speaker: Nanbor
Goals: example-driven presentation of how QoS facilities are being incorporated in CIAO and what the Cadena needs to be aware of to support these.

4:00 - 4:30 Configuring attributes across the life cycle -- an example (.ppt)

Speaker: Chris
Goals: an example that Nanbor and I have evolved that illustrates how configuring different attributes at different points in the system lifecycle (packaging, assembly, deployment) allows both functional and real-time properties to be checked and customized, and can result in a "best fit" balance between utilization and feasibility.

4:30 - 5:00 QoS issues in RT middleware

Speaker: Gurdip

5:00 - 6:00 Working Groups

--- Break ---

6:30 - 8:00 Dinner: Little Apple Brewing Company

Tuesday Morning: Modeling and Model-Checking DRE Systems

8:00 - 8:30 Coffee, light fruit and snacks
8:30 - 8:45: KSU Modeling Overview -- Strategic Directions (.ppt for 1st and last talks in session)

Speaker: Matt
Goals: Explain to CIAO Team our approach to model-checking and where our work is headed over next several years.

8:45 - 9:15: Overview of Bogor (.ppt)

Speaker: Robby
Goals: gentle introduction to model-checking and Bogor -- our extensible model-checker targeted to OO software
Supporting artifacts:

9:15 - 10:00: Model-checking Boeing OEP Scenarios in Bogor

Speaker: Matt
Goals: how we are currently model-checking Boeing OEP designs and the RT event channel in Bogor
Supporting artifacts:

Model-checking Middleware-based Event-driven Real-time Embedded Software (paper) (note: this paper is the most appropriate one to read if you want to understand our approach to model-checking systems built on top of the ACE/Tao event-channel)
Space Reductions for Model Checking Quasi-Cyclic Systems (paper)

--- Break ---

10:30 - 11:00 GME Overview and Demo

Speaker: Tao
Goals: explain the rationale and functionality of GME to KSU team
Supporting artifacts:

GME web-site

11:00 - 11:45 Overview of CoSMIC

Speaker: Andy
Goals: present an overview of MDA in CoSMIC
Supporting artifacts:

CoSMIC: An MDA Generative Tool for Distributed Real-time and Embedded Component Middleware and Applications

--- Lunch: Aggieville ---

Tuesday Afternoon: Open Issues and Planning

1:15 - 1:45 Proposal for a correlation framework for Cadena

Speaker: Georg
Goals: example-driven presentation of how correlation specification and generation can be incorporated into CCM and Cadena

1:45 - 3:00 Discussion and Planning

Slides containing notes of discussion (ppt)

--- Break ---

3:30 - 5:30 Working Groups

--- Break ---

6:00 - 7:30 Dinner: Hunam Mongolian Barbeque
8:00 – 10:00 The Matrix II – Seth Childs Cinema