ArticlePDF Available

Controlled Stochastic Petri Nets for Multimedia QoS Management

Authors:
Hermann de Meer at University of Passau
Hermann de Meer
Stefan Fischer at University of Lübeck
Stefan Fischer
Download full-text PDF
Read full-text
Download citation

Abstract and Figures

Most QoS management systems for multimedia applications just cope with technical issues of QoS guarantees. A QoS management system taking revenue issues and the possibly stochastic behavior of the environment into account seems to be superior to the existing ones. In this paper, we show how controller programs for such enhanced QoS management systems can be developed based on a new kind of Petri Nets, so-called Controlled Stochastic Petri Nets. We show how to numerically analyze such networks using a tool environment in order to obtain strategies for the QoS mangement system. 1 Introduction The design of distributed multimedia applications, such as systems for access to remote multimedia databases or teleconferencing, requires careful consideration of quality of service (QoS) issues, because the presentation quality of live media, especially video, requires relatively high utililisation of networking bandwidth and processing power in the end systems. For applications running in a shar...
Figures - uploaded by Stefan Fischer
Author content
All figure content in this area was uploaded by Stefan Fischer
Content may be subject to copyright.
Content uploaded by Stefan Fischer
Author content
All content in this area was uploaded by Stefan Fischer on Apr 27, 2013
Content may be subject to copyright.
resources
r1
waiting audio
active
audio
waiting video
r2
videos
active
compl.
succ.
4
4
p4
p5 p6
p7 p9p8
4
t1
aborted
i1 i2
p1 p2 p3
t2
t3 t4
t5
t6
reconfig.
resource
audio
compl.
succ.
admit audio
release
resources
severe QoS violation
audio abort
successful
audio completion
admit video
resource
reconfiguration
aborted video
successful
video completion
severe QoS violation
video abort
Firing probs.: 1
M000010 M000000 M010000 M020000
M100000 M110000
M200000
M000001
M001000M000100
M011000
M101000
Vanishing State
Markov State
Reconfiguration Edge
Transition
1
1
1
1
tau2
tau4 2*tau4
25*tau6*gamma
tau5
4*tau3*gamma
tau1
2*tau1
9*tau3*gamma
9*tau3*gamma
tau1
tau4
Region III
0
50
100
150
200
250
300
350
450
1e-05 0.0001 0.001 0.01
Time
gamma
audio reward = 1; video reward = 12
M000000->M000020
M100001->M100001
M010010->M010010
400
Region I
Region II
0
5
10
15
20
25
30
35
1e-06 1e-05 0.0001 0.001 0.01
Time
gamma
audio reward = 1; video reward = 14
M000000->M000020
M100001->M100001
M010010->M010010
0
5
10
15
20
25
30
35
40
45
1e-05 0.0001 0.001 0.01
Time
gamma
continuous audio reward = 0.1; video reward = 12
M000000->M020010
M000000->M000020
190
191
192
193
194
195
196
197
198
199
200
1e-05 0.0001 0.001 0.01
Reward
gamma
Performance for audio = 1; video = 12
M000000
... Diese Strategien werden nun an einen Agenten gegeben, so daß sein Verhalten entsprechend kontrolliert werden kann. Die Idee der Anwendung von COSTPN für QoS Management im allgemeinen wird in [3] beschrieben, die Anwendung speziell im Rahmen des Cooperative QoS Management kann in [5] detailliert nachgelesen werden. ...
View
Management of QoS with Software Agents
Article
Full-text available
  • Mar 1999
As distributed multimedia applications become more widely diffused, systems for QoS management are increasingly essential. In this paper we present an architecture for distributed QoS management based on software agents. We discuss the advantages of using agent technology in a problem inherently distributed and complex such as QoS management. Besides introducing several negotiation and coordination techniques among agents as fundamental properties for QoS management, it is argued in favor of decision making based on analytical and simulative modeling techniques. Furthermore, design and implementation issues are discussed in detail and our Java-based platform for an agent-based QoS management is introduced. Supported by italian CRUI and german DAAD under the Programma Vigoni 1 1 Introduction Recent technological developments in the field of high-speed networks and low-cost multimedia stations are pushing the development and diffusion of new types of distributed multimedia ...
View
PENELOPE: Dependability Evaluation and the Optimization of Performability.
Conference Paper
Full-text available
  • Jan 1997
  • Lect Notes Comput Sci
A new performance and performability modeling tool is introduced in this paper. PENELOPE is the first tool which incorporates evaluation and optimization algorithms. It is the result of a combination between the performability modeling concept and Markov decision theory. Different algorithms are adopted and included in the tool under the unifying paradigm of reconfigurability as the basis for adaptation and optimization. In addition to transient and steady-state performability measures, also transient and stationary control functions can be computed and graphically presented. Model specification and specification of transient or stationary control functions can be separately performed and deliberately combined with each other. Besides providing a new modeling paradigm, the tool supports model creation, experimentation, storage and presentation of results by means of an easily usable interface and an integrated model data base system.
View
A Class of Generalized Stochastic Petri Nets for the Performance Evaluation of Multiprocessor Systems
Article
Full-text available
  • May 1984
Generalized Stochastic Petri Nets (GSPNs) are presented and are applied to the performance evaluation of multiprocessor systems. GSPNs are derived from standard Petri nets by partitioning the set of transitions into two subsets comprising timed and immediate transitions. An exponentially distributed random firing time is associated with each timed transition, whereas immediate transitions fire in zero time. It is shown that GSPNs are equivalent to continuous-time stochastic processes, and solution methods for the derivation of the steady state probability distribution are presented. Examples of application of GSPN models to the performance evaluation of multiprocessor systems show the usefulness and the effectiveness of this modeling tool.
View
Controlled stochastic Petri nets
Conference Paper
Full-text available
  • Nov 1997
A new framework for the extension of stochastic Petri nets (SPNs) is introduced. SPNs are extended by elements providing means for a dynamic optimization of performability measures. A new type of transition is defined, offering a feature for specification of controlled switching, called reconfiguration, from one marking of a SPN to another marking. Optional reconfiguration transitions are evaluated in order to optimize a specified reward or cost function. The result of an analysis is provided in the output of a numerical computation, in the form of a graphical presentation of an optimal, marking dependent control strategy and the resulting performability measure when applying the optimal strategy. The extended SPNs are called COSTPNs (Controlled Stochastic Petri Nets). COSTPNs are mapped on EMRMs (Extended Markov Reward Models) for a numerical analysis. Computational analysis is possible with algorithms adopted from Markov decision theory, including transient and stationary optimization. The scope of the paper is to introduce the new control structure for SPNs and to present an algorithm for the mapping of COSTPNs on EMRMs
View
Cooperative QoS Management for Multimedia Applications
Article
Full-text available
  • Jul 1998
Quality of Service (QoS) management becomes more and more important, especially in networks where many applications are competing for a limited number of resources. As these applications become more complex (consider e.g. multiparty multimedia applications), the number of options for QoS management increases, leading to more complex decision processes. In this paper, we propose an approach for cooperative QoS management, where application-oriented QoS agents are distributed throughout the network and the end systems, communicating with each other. This distributed management system tries to guarantee the QoS level negotiated with the users, at the same time optimizing resource usage. The advantages of distributing the management process are (i) an easier and more precise localization of the cause of QoS problems, (ii) better knowledge of local situations, (iii) a lower complexity for a single QoS agent and (iv) an increase in possible actions. We describe management procedures for QoS ...
View
Guarded repair of dependable systems
Article
  • Jun 1994
  • THEOR COMPUT SCI
Imperfect coverage and nonnegligible reconfiguration delay are known to have a deleterious effect on the dependability and the performance of a multiprocessor system. In particular, increasing the number of processor elements does not always increase dependability. An obvious reason for this is that the total failure rate increases, generally, linearly with the number of components in the system. It is also a well-known fact that the performance gain due to parallelism mostly turns out to be sublinear with the number of processors. It is therefore important to optimize the degree of parallelism in system design. A related issue is that by deferring repair, it is sometimes possible to improve system dependability. In this case decisions have to be made dynamically as to when to repair and when not to repair.
View
  • Jan 1997
  • C Aurrecoechea
  • A Campbell
  • L Hauw
C. Aurrecoechea, A. Campbell, and L. Hauw. A Survey of QOS Architectures. Multimedia Systems Journal, Special Issue on QoS Architectures, 1997. To appear.
PENELOPE: dePENdability EvaLuation and the Optimization of PErformability
  • Jun 1997
  • Lect Notes Comput Sci
  • 19-32
  • H De Meer
  • H Sevcikova
H. de Meer and H. Sevcikova. PENELOPE: dePENdability EvaLuation and the Optimization of PErformability. In 9th Int. Conf. on Computer Performance Evaluation -Modelling Techniques and Tools, St. Malo, France, Lecture Notes in Computer Science 1245, pages 19{32. Springer, June 1997.