Research at Mlardalen University and Department of Computer Science and Electrics

1
Research at Mälardalen Universityand Department
of Computer Science and Electrics

• Ivica Crnkovic
• Mälardalen University
• Department of Computer Science and Electronics,
• Mälardalen Real-time Research Centre
• Sweden
• http//www.idt.mdh.se/icc

2
Mälardalen University (MdH)
Mälardalen University, Vasteras (Västerås)
Prof. in Software Engineering http//www.idt.mdh.
se/icc ivica.crnkovic_at_mdh.se
Department of Computer Engineering
Real-Time Systems Design Lab Computer
Architecture Lab Computer Science Lab
Software Engineering Lab
3
Mälardalen University

• Placed in Västerås and Eskilstuna
• 15 000 students
• 400 intl. magister students
• Research
• 12 M
• 60 professors
• 250 PhD students

4
IDE och other MdH-depts

• Engineering
• Department of Computer Science and Electronics,
  IDE
• Innovation, Design and Product Development, IDP
• Public Technology
• Mathematics Physics, IMa
• Biology Chemistry, IBK
• Humanities, Social sciences and Caring sciences

5
IDE facts and figures

• Staff 100
• 10 Professors
• 20 Researchers/Senior Lecturers
• 60 PhD-students (20 of which are Industrial
  PhD-students)
• 10 Lecturers
• 10 AdminEngineering
• IDE merge of two depts. in Jan 2005
• Computer Science and Engineering (MRTC)
• Electrical Engineering (ISS)
• Strong focus on research gt 60

6
MRTCs 3 legs
Int'l scientific community
Students
Basic and applied research
Undergraduate Graduate Continued Education
Industrial co-op Commercialisation of research
Industry
Society
7
Education_at_IDE

• Netcenter
• ABB, Ericsson, m.fl.
• Co-ordination and marketing

• Engineering in CSE (civing)
• Engineering in Robotics (civing)
• Computer Science
• Mechatronics
• Computer Interaction Computer games
• Intl MSc (within MIMA)
• Computer Sci. with AI
• Computer Sci. with Software Engineering
• Electronics with Biomedical Engineering
• Real Time Systems
• Robotics

External courses
Under- graduate (including MSc)
Graduate education
Volume and organisation 30 supervisors 10 Dr
2005 60 PhD-stud. 10 lic 2005
8
MRTC ISS activities
Groups
Lab
Meka- tronic
Real Time
ISS
medical
IDE
Architecture for
Medical
Tech.
RT
-
System
MRTC
Tech.
Software
-
Computer science
engineering
Volvo
S80
AI
Program- analysis
6
9
MRTC Scientific basis
10
MRTC cooperation
BTH (C.Wohlin)
11
Model of cooperation
Industry
Theses interns
Research stays
Industry stays
Eval. of research results
Courses
12
Approach and research questions

• Questions
• Process for
• Component development
• System construction from components
• Integration, configuration management
• Distributed development

Life cycle processes (development/maintenance/evol
ution)
Existing systems ((or rthird party s.)
Product requirements
componentisation
Model based framework For Component-based
software development
Environment
validation

• Research questions
• Generic component model
• Composition - integration
• adaptation
• Non functional requirements
• Integration models
• Static analysis
• Scheduling
• Resource utilization
• Reliability, safety
• Approximate analysis
• Dynamic Analysis
• System modelling
• Simulation
• Test and monitoring
• Synthesis
• Models
• Realisation
• Existing subsystem

• Platform - questions
• Component-middleware
• Optimized platform
• Virtual network
• Diagnosis and error management

Componentmodel
System Models
Flexible platform
Plattformsegenskaper
Analysis models
Analysis models
Analysis models
Runtime environment
Network Nodes
Component database
13
Some of the Activities
Composition and RT theories
Predictable CBD
Modeling RevEng
Processes
14
Example Research project SaveSaveCCM - a
component model for safety-critical real-time
systems
15
SaveCCM

• Goal develop a component model that is suitable
  for RT embedded systems
• Steps
• Spcify Component model
• Build a prototype and build some systems using
  SaveCCM
• Provide different analysis to achieve predictable
  behavior
• Tradeoff
• Freedom vs. Restrictions
• SaveCCM restrictive, but expressive power focused
  on domain specific needs, e.g.,
• Mode Changes
• Static Configuration
• Control Feedback

16
Architectural Elements

• Components
• In and Output ports, (1) data only, (2)
  triggering, (3) triggering and data (for all
  elements, not only components)
• Basic units of encapsulated behavior
• Execution model read input, execute, write output

Input port, data and triggering
17
Architectural Elements

• Switches, special type of component
• Conditionally changes the componentinterconnectio
  n structure
• Static or dynamic configuration of conditions

P
I
D
Switch name Inports INPPport,
INIIport, INDDport, PSetport, ISetport,
DSetport Outports OUTPPport,
OUTIIport, OUTDDport Switching P
INP-gtOUTP I INI-gtOUTI D
IND-gtOUTD
OUTP
INP
OUTI
INI
IND
OUTD
18
Architectural Elements

• Assemblies, special type of component
• Consists of components and switches
• Naming of sub-systems, hiding internal structure

19
Static Configuration (Off-line)
P-Mode
1 0 0
20
Development Method Overview
Fully Automated Compile-Time Step
Task Allocation
Attribute Assignment
Model transformation

Analysis
Real
-
time model

Compile
-


Time


t

Glue Code Generation

Real
-
Time

Synthesis
Analysis

Target Compiler

RTOS

RTOS

21
Allocating components to real-time tasks

• Today one-to-one allocation is commonly used
• Not efficient in terms of cpu-overhead and stack
  usage
• However, highly analyzable
• How can the mapping between components and tasks
  be analyzable and efficient?
• Infeasible to calculate due to the many different
  possible mappings in a large system
• Limitations
• Only pipe-and-filter architectures
• No advanced real-time constraints

22
Current state