1 / 48

Analysis of cooperation in multi-organization Scheduling

Analysis of cooperation in multi-organization Scheduling. Pierre-François Dutot (Grenoble University) Krzysztof Rzadca (Polish-Japanese school, Warsaw) Fanny Pascual (LIP6, Paris) Denis Trystram (Grenoble University and INRIA) NCST, CIRM, may 13, 2008. Goal.

jersey
Download Presentation

Analysis of cooperation in multi-organization Scheduling

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. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Analysis of cooperation in multi-organization Scheduling Pierre-François Dutot (Grenoble University) Krzysztof Rzadca (Polish-Japanese school, Warsaw) Fanny Pascual (LIP6, Paris) Denis Trystram (Grenoble University and INRIA) NCST, CIRM, may 13, 2008

  2. Goal The evolution of high-performance execution platforms leads to distributed entities (organizations) which have their own « local » rules. Our goal is to investigate the possibility of cooperation for a better use of a global computing system made from a collection of autonomous clusters. Work partially supported by the Coregrid Network of Excellence of the EC.

  3. Main result We show in this work that it is always possible to produce an efficient collaborative solution (i.e. with a theoretical performance guarantee) that respects the organizations’ selfish objectives. A new algorithm with theoretical worst case analysis plus experiments for a case study (each cluster has the same objective: makespan).

  4. Target platforms … … … … … … Our view of computational grids is a collection of independent clusters belonging to distinct organizations.

  5. Computational model(one cluster) Independent applications are submitted locally on a cluster. The are represented by a precedence task graph. An application is a parallel rigid job. Let us remind briefly the model (close to what presented Andrei Tchernykh this morning)… See Feitelson for more details and classification.

  6. Cluster Local queue of submitted jobs … J3 J2 J1 … …

  7. Job

  8. overhead Computational area Rigid jobs: the number of processors is fixed.

  9. Runtime pi #of required processors qi

  10. Runtime pi #of required processors qi high jobs (those which require more than m/2 processors) and low jobs (the others).

  11. Scheduling rigid jobs:Packing algorithms Scheduling independent rigid jobs may be solved as a 2D packing Problem (strip packing). List algorithm (off-line). m

  12. Cluster n organizations. … J3 J2 J1 … … Organization k m processors (identical for the sake of presentation) k

  13. … … … … … …

  14. … … … … … …

  15. n organizations (here, n=3, red 1, green 2 and blue 3) Cmax(Ok) organization Cmax(Mk) cluster Each organization k aims at minimizing « its » own makespan max(Ci,k) We want also that the global makespan is minimized

  16. n organizations (here, n=3, red 1, green 2 and blue 3) Cmax(O3)= Cmax(M1) Cmax(Ok) organization Cmax(Mk) cluster Cmax(O2) Each organization k aims at minimizing « its » own makespan max(Ci,k) We want also that the global makespan (max over all local ones) is minimized

  17. Problem statement MOSP: minimization of the « global » makespan under the constraint that no local schedule is increased. Consequence: taking the restricted instance n=1 (one organization) and m=2 with sequential jobs, the problem is the classical 2 machines problem which is NP-hard. Thus, MOSP is NP-hard.

  18. Multi-organizations Motivation: A non-cooperative solution is that all the organizations compute their local jobs (« my job first » policy). However, such a solution is arbitrarly far from the global optimal (it grows to infinity with the number of organizations n). See next example with n=3 for jobs of unit length. O1 O1 O2 O2 O3 O3 with cooperation (optimal) no cooperation

  19. Preliminary results Single organization scheduling (packing). Resource constraint list algorithm (Garey-Graham 1975). • List-scheduling: (2-1/m) approximation ratio • HF: Highest First schedules (sort the jobs by decreasing number of required processors). Same theoretical guaranty but better from the practical point of view.

  20. Analysis of HF (single cluster) Proposition. All HF schedules have the same structure which consists in two consecutive zones of high (I) and low (II) utilization. Proof. (2 steps) By contracdiction, no high job appears after zone (II) starts low utilization zone (II) high utilization zone (I) (more than 50% of processors are busy)

  21. Collaboration between clusters can substantially improve the results. Other more sophisticated algorithms than the simple load balancing are possible: matching certain types of jobs may lead to bilaterally profitable solutions. no cooperation with cooperation O1 O1 O2 O2

  22. If we can not worsen any local makespan, the global optimum can not be reached. local globally optimal 2 O1 1 O1 1 1 1 2 O2 O2 2 2 2 2

  23. If we can not worsen any local makespan, the global optimum can not be reached. local globally optimal 2 O1 1 O1 1 1 1 2 O2 O2 2 2 2 2 O1 2 1 1 best solution that does not increase Cmax(O1) O2 2 2

  24. If we can not worsen any local makespan, the global optimum can not be reached. Lower bound on approximation ratio greater than 3/2. 2 O1 1 O1 1 1 1 2 O2 O2 2 2 2 2 O1 2 1 1 best solution that does not increase Cmax(O1) O2 2 2

  25. Multi-Organization Load-Balancing 1 Each cluster is running local jobs with Highest First LB = max (pmax,W/nm) 2. Unschedule all jobs that finish after 3LB. 3. Divide them into 2 sets (Ljobs and Hjobs) 4. Sort each set according to the Highest first order 5. Schedule the jobs of Hjobs backwards from 3LBon all possible clusters 6. Then, fill the gaps with Ljobs in a greedy manner

  26. Hjob Ljob let consider a cluster whose last job finishes before 3LB 3LB

  27. Hjob Ljob 3LB

  28. Hjob Ljob 3LB

  29. Hjob Ljob 3LB

  30. Ljob 3LB

  31. Ljob 3LB

  32. Notice that the centralized scheduling mechanism isnot work stealing, moreover, the idea is to changeas minimum as we can the local schedules.

  33. Feasibility (insight) Zone (I) Zone (II) Zone (I) 3LB

  34. Sketch of analysis Proof by contradiction: let us assume that it is not feasible, and call x the first job that does not fit in a cluster. Case 1: x is a small job. Global surface argument Case 2: x is a high job. Much more complicated, see the paper for technical details

  35. 3-approximation (by construction)This bound is tight Local HF schedules

  36. 3-approximation (by construction)This bound is tight Optimal (global) schedule

  37. 3-approximation (by construction)This bound is tight Multi-organization load-balancing

  38. We add an extra load-balancing procedure Improvement O1 O2 Local schedules O3 O4 O5 O1 O2 Multi-org LB O3 O4 O5 O1 O2 O3 Compact O4 O5 O1 O2 O3 load balance O4 O5

  39. Some experiments

  40. Conclusion • We proved in this work that cooperation may help for a better global performance (for the makespan). • We designed a 3-approximation algorithm. • It can be extended to any size of organizations (with an extra hypothesis).

  41. Conclusion • We proved in this work that cooperation may help for a better global performance (for the makespan). • We designed a 3-approximation algorithm. • It can be extended to any size of organizations (with an extra hypothesis). • Based on this, it remains a lot of interesting open problems, including the study of the problem for different local policies or objectives (Daniel Cordeiro)

  42. Thanks for attentionDo you have any questions?

  43. Using Game Theory? We propose here a standard approach using Combinatorial Optimization. Cooperative Game Theory may also be usefull, but it assumes that players (organizations) can communicate and form coalitions. The members of the coalitions split the sum of their playoff after the end of the game. We assume here a centralized mechanism and no communication between organizations.

More Related