dynamic resource management considering the real behavior of aggregate traffic n.
Download
Skip this Video
Loading SlideShow in 5 Seconds..
Dynamic Resource Management Considering the Real Behavior of Aggregate Traffic PowerPoint Presentation
Download Presentation
Dynamic Resource Management Considering the Real Behavior of Aggregate Traffic

Loading in 2 Seconds...

play fullscreen
1 / 13

Dynamic Resource Management Considering the Real Behavior of Aggregate Traffic - PowerPoint PPT Presentation


  • 121 Views
  • Uploaded on

Dynamic Resource Management Considering the Real Behavior of Aggregate Traffic. Jos é R. Gallardo, Dimitrios Makrakis, and Marlenne Angulo IEEE Transaction on Multimedia, VOL. 3, NO. 2, June 2001. Outline. Motivation and Goal Dynamic traffic control using predictive techniques

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about 'Dynamic Resource Management Considering the Real Behavior of Aggregate Traffic' - sissy


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
dynamic resource management considering the real behavior of aggregate traffic

Dynamic Resource Management Considering the Real Behavior of Aggregate Traffic

José R. Gallardo, Dimitrios Makrakis, and Marlenne Angulo

IEEE Transaction on Multimedia, VOL. 3, NO. 2, June 2001

Chin-Kai Wu, CS, NTHU

outline
Outline
  • Motivation and Goal
  • Dynamic traffic control using predictive techniques
    • Differentiated Service
    • Model
    • Traffic policing and resource allocation algorithms
  • Simulation Parameters and Results

Chin-Kai Wu, CS, NTHU

motivation and goal
Motivation and Goal
  • Theory of effective bandwidth
    • The probability of buffer overflow in a stationary queueing system decreases exponentially with the buffer size as long as the “effective bandwidth” of the input traffic stream is smaller than the server capacity
  • Goal
    • Propose an alternative dynamic approach to resource allocation and management

Chin-Kai Wu, CS, NTHU

differentiated service
Differentiated Service

Chin-Kai Wu, CS, NTHU

model
Model

marker

Chin-Kai Wu, CS, NTHU

traffic policing
Traffic policing
  • Two different policing mechanisms that are based on the leaky bucket idea: conventional leaky bucketand adaptive leaky bucket assisted by traffic prediction.
  • The purpose is to optimize the resource utilization, which in turn intends to minimize the number of violations to the QoS requirements of the traffic.

Chin-Kai Wu, CS, NTHU

conventional leaky bucket clb

Packets arrive

Generate token

each constant time

NO

YES

Buffer full

YES

Pool full

Got Token

NO

NO

YES

Drop

Put

Drop

Send

Wait

Conventional Leaky Bucket (CLB)

Chin-Kai Wu, CS, NTHU

adaptive leaky bucket assisted by traffic prediction palb
Adaptive Leaky Bucket Assisted by Traffic Prediction (PALB)

Buffer occupancy above upper bond

YES

NO

Buffer occupancy below lower bond

Increase generation rate

NO

YES

Maintain generation rate

Decrease generation rate

Chin-Kai Wu, CS, NTHU

simulation parameters
Simulation Parameters

Chin-Kai Wu, CS, NTHU

results
Results

Chin-Kai Wu, CS, NTHU