A Dynamically Adaptive Hybrid Algorithm for Scheduling Lightpaths
A Dynamically Adaptive Hybrid Algorithm for Scheduling Lightpaths in Lambda-Grids Neena R. Kaushik and Silvia M. Figueira Santa Clara University Presented by Yang-suk Kee
Outline n n Background Lightpath Scheduling Experiments Conclusion
Background n Advance reservation • Guarantees that resources will be available at a pre- determined time to participate in the execution of a Grid application n Lambda Grid • Lightpath n n A wavelength data channel liking multiple optical segments n A single wavelength through the path n Different wavelength with wavelength conversion Data-intensive grid application n Multiple paths
Problem Definition n Advance reservation of multiple lightpaths • How to schedule multiple lightpaths requested by advance reservation over optical network with no converter? A Source B D C Destination
Lightpath Scheduling n Skeleton • Step 1: determines edge-disjoint paths • Step 2: determines wavelengths for paths n Edge-disjoint path • nth edge-disjoint path A path that does not share any of its edges with the previous n-1 edge-disjoint paths st edge-disjoint path is the shortest path n 1 • Algorithm n Repeat Dijkstra’s shortest path algorithm by removing the edges that were part of the shortest paths. n
Lightpath Scheduling Schemes Spreading • Balances the wavelength assignment throughout the available edge n -disjoint paths Packing • Concentrates the wavelength assignment in the shortest path first n followed by the alternate edge-disjoint paths A->C (4) A B A D B D C C Spreading Packing
Wavelength-Balancing Algorithm Begin For i = 1 to number of wavelengths For j = 1 to number of edge-disjoint paths If wavelength[i] is available for all segments in edge-disjoint path[j] Allocate wavelength[i] for all segments in edge-disjoint path[j] If all requests of user are satisfied Print success Else Print the number of requests satisfied and number denied End
Wavelength-Concentrating Algorithm Begin For i = 1 to number of edge-disjoint paths For j = 1 to number of wavelengths If wavelength[j] is available for all segments in edge-disjoint path[i] Allocate wavelength[j] for all segments in edge-disjoint path[i] If all requests of user are satisfied Print success Else Print the number of requests satisfied and number denied End
Scheduling Comparison A->C (4), B->D (4), A->B (1), B->D (2) Four wavelengths per link A B (x) D A B D (x) C Balancing C Concentrating
Blocking Probability Requests for 1 -hop lightpaths in a 32 -node ring: uniform requests Requests for 8 -hop lightpaths in a 32 -node ring with chords: constant requests.
Lessons n From the experiments using simple topologies • Partial mesh/ring with spike/ring with chord n n Balancing algorithm has similar or better blocking probability than concentrating one in most cases Concentrating algorithm is better • With short hop (1 -hop) n n Shorter time slots provide lower blocking probability => Hybrid algorithm of balancing and concentrating
Hybrid Algorithm of Balancing and Concentrating n n n begin firstpass = 1; while (firstpass is not equal to 3) • for i = 1 to number of wavelengths n for j = 1 to number of edge-disjoint paths n if first pass is equal to 1 and edge-disjoint path has more than x-hops n Continue n … check if it satisfies balance algorithm end (for edge-disjoint paths loop) • end (for number of wavelengths loop) • Increment firstpass by 1 n n end (while loop) return the number of requests satisfied end
Experimental Setup n Metric of quality • Blocking probability (denial rate) n Simulation-based • FONTS (Flexible Optical Network Traffic Simulator) Generates on-demand advance reservation requests n Uses Stochastic models • LRSS (Lightpath Request Scheduling Simulator) n Simulates scheduling algorithms n Takes as input the network topology and a trace of requests for ligthpaths (from FONTS) n
FONTS (Flexible Optical Network Traffic Simulator) n n n Request arrival time • Poisson distribution Source node/Destination node • Constant/uniform/arbitrary probability Size of data to transfer (Tera bytes) • Constant/uniform/arbitrary probability/heavy-tailed Number of lightpaths requested • Constant/uniform/heavy-tailed Advance reservation start time • Poisson distribution Number of time slots • Constant/variable
Simple Topology 16 20 17 18 24 Requests for 1 -hop lightpaths in a 32 -node ring: uniform requests All schemes except balancing are identical Requests for 8 -hop lightpaths in a 32 -node ring with chords: constant requests.
National Lambda Rail: Constant Requests
Conclusion & Discussion n Conclusion • Provides simulation tools for lightpath scheduling • Quality of scheduling algorithms depends on traffic characteristics • Hybrid scheduling achieved lower blocking probability n Discussion • Blocking probability represents network utilization properly? • This hybrid algorithm is close to balancing algorithm except the case of ring topology. • Need more study about the degree of nodes, length of hops, and network utilization
- Slides: 17