Presentation is loading. Please wait.

Presentation is loading. Please wait.

HeteroPar 2013 Optimization of a Cloud Resource Management Problem from a Consumer Perspective Rafaelli de C. Coutinho, Lucia M. A. Drummond and Yuri Frota.

Similar presentations


Presentation on theme: "HeteroPar 2013 Optimization of a Cloud Resource Management Problem from a Consumer Perspective Rafaelli de C. Coutinho, Lucia M. A. Drummond and Yuri Frota."— Presentation transcript:

1 HeteroPar 2013 Optimization of a Cloud Resource Management Problem from a Consumer Perspective Rafaelli de C. Coutinho, Lucia M. A. Drummond and Yuri Frota Fluminense Federal University – UFF Brazil Sponsorship:

2 Outline Motivation Definition and Formulation GRASP Results and Conclusions Outline o Motivation o Problem Definition and Mathematical Formulation o GRASP for Resource Selection in Cloud Environments o Preliminary Results and Conclusion Outline

3 Motivation Definition and Formulation GRASP Results and Conclusions Motivation Cloud Computing a large-scale distributed computing paradigm in which computing resources are available to consumers via Internet. Motivation

4 Outline Motivation Definition and Formulation GRASP Results and Conclusions Motivation Cloud Computing a large-scale distributed computing paradigm in which computing resources are available to consumers via Internet. It delivers infrastructure, platform and software as services by signing service-level agreements (SLAs) with consumers. Motivation

5 Outline Motivation Definition and Formulation GRASP Results and Conclusions Motivation  The provider should offer resource-economic services.  New pricing models based on the pay-as-you-go policy are necessary to address the highly variable demand for cloud resources. Motivation

6 Outline Motivation Definition and Formulation GRASP Results and Conclusions Motivation  The provider should offer resource-economic services.  New pricing models based on the pay-as-you-go policy are necessary to address the highly variable demand for cloud resources. For example, cloud service consumers might have an SLA with a cloud service provider concerning how much bandwidth, CPU, and memory the consumer can use at any given time throughout the day. Motivation

7 Outline Motivation Definition and Formulation GRASP Results and Conclusions Motivation  Application schedulers have different policies that vary according to the objective function.  Independently of the objective function, resource allocation problems in clouds are NP-hard problems. Motivation

8 Outline Motivation Definition and Formulation GRASP Results and Conclusions Motivation Cloud computing becomes more widespread and Computing resources grow More researches have been conducted on Resource Allocation Problem in Clouds. Motivation

9 Outline Motivation Definition and Formulation GRASP Results and Conclusions Definition and Formulation  A resource management problem that aims to reduce the payment cost and the execution time of the user application is defined.  An integer programming formulation and a heuristic based on Greedy Randomized Adaptive Search Procedure are proposed. Definition and Formulation

10 Outline Motivation Definition and Formulation GRASP Results and Conclusions Definition Problem  Infrastructure-as-a-Service (Iaas) model: cloud consumers request computing resources as processing power, disk storage, memory and architecture type, for a period of time and pays only what he/she uses. Cloud providers have a wide variety of virtual machines. Cloud consumers have to decide which packages should purchase in order to minimize a specific objective as execution time or payment cost. Definition and Formulation

11 Outline Motivation Definition and Formulation GRASP Results and Conclusions Definition Problem  Infrastructure-as-a-Service (Iaas) model: cloud consumers request computing resources as processing power, disk storage, memory and architecture type, for a period of time and pays only what he/she uses.  Cloud providers have a wide variety of package types. Cloud consumers have to decide which packages should purchase in order to minimize a specific objective as execution time or payment cost. Definition and Formulation

12 Outline Motivation Definition and Formulation GRASP Results and Conclusions Definition Problem  Infrastructure-as-a-Service (Iaas) model: cloud consumers request computing resources as processing power, disk storage, memory and architecture type, for a period of time and pays only what he/she uses.  Cloud providers have a wide variety of package types.  Cloud consumers have to decide which packages should purchase in order to minimize a specific objective as execution time or payment cost. Definition and Formulation

13 Outline Motivation Definition and Formulation GRASP Results and Conclusions Modeling the problem … Definition and Formulation Package

14 Outline Motivation Definition and Formulation GRASP Results and Conclusions Modeling the problem … Definition and Formulation Package

15 Outline Motivation Definition and Formulation GRASP Results and Conclusions Modeling the problem … Definition and Formulation CloudProvider

16 Outline Motivation Definition and Formulation GRASP Results and Conclusions Modeling the problem … Definition and Formulation CloudProvider

17 Outline Motivation Definition and Formulation GRASP Results and Conclusions Modeling the problem … Definition and Formulation

18 Outline Motivation Definition and Formulation GRASP Results and Conclusions The Mathematical Formulation CC-IP Subject to: 1. The paid costs do not exceed the maximum recommended cost. Definition and Formulation

19 Outline Motivation Definition and Formulation GRASP Results and Conclusions The Mathematical Formulation CC-IP Subject to: 2. The storage and memory capacity should be sufficient large to meet the requirements in each period of time. Definition and Formulation

20 Outline Motivation Definition and Formulation GRASP Results and Conclusions The Mathematical Formulation CC-IP Subject to: 3. The processing power is at least large enough to satisfy the total demand. 4. The number of selected packages does not exceed the cloud providers limit. Definition and Formulation

21 Outline Motivation Definition and Formulation GRASP Results and Conclusions The Mathematical Formulation CC-IP Definition and Formulation

22 Outline Motivation Definition and Formulation GRASP Results and Conclusions The Mathematical Formulation CC-IP Subject to: 7. An order between the packages should be established and the symmetry should be eliminate. 8. The binary and integrality requirements of the variables. Definition and Formulation

23 Outline Motivation Definition and Formulation GRASP Results and Conclusions GRASP for Resource Selection in Cloud  Exact procedures have often proved incapable of finding solutions as they are extremely time-consuming. Heuristics and metaheuristics provide sub-optimal solutions in a reasonable time. GRASP

24 Outline Motivation Definition and Formulation GRASP Results and Conclusions GRASP for Resource Selection in Cloud  Exact procedures have often proved incapable of finding solutions as they are extremely time-consuming. Heuristics and metaheuristics provide sub-optimal solutions in a reasonable time.  A Greedy Randomized Adaptive Search Procedure (GRASP) is proposed. GRASP

25 Outline Motivation Definition and Formulation GRASP Results and Conclusions GRASP for Resource Selection in Cloud  Exact procedures have often proved incapable of finding solutions as they are extremely time-consuming. Heuristics and metaheuristics provide sub-optimal solutions in a reasonable time.  A Greedy Randomized Adaptive Search Procedure (GRASP) is proposed.  The proposed heuristic GraspCC is composed of two phases: a construction phase coCC a local search phase lsCC GRASP

26 Outline Motivation Definition and Formulation GRASP Results and Conclusions GraspCC GRASP

27 Outline Motivation Definition and Formulation GRASP Results and Conclusions GraspCC Adittional Definitions Define a cost function which will measure the quality of the solution: GRASP

28 Outline Motivation Definition and Formulation GRASP Results and Conclusions Construction Phase - coCC GRASP

29 Outline Motivation Definition and Formulation GRASP Results and Conclusions Local Search Phase - lsCC GRASP

30 Outline Motivation Definition and Formulation GRASP Results and Conclusions GraspCC  GraspCC consists to perform the coCC following by the lsCC until the maximum number of iterations without improvement in the best solution found is satisfied. GRASP

31 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results  Implemented in: C CPLEX 12.4, for mathematical formulation.  Executed on a computer with processor Intel Core i7 3.4Hz and 12Gb of RAM under Linux (Ubuntu 12.04) operating system.  Tested over a set of instances constructed from the requirements of real applications combined with the sets of virtual machine packages available in commercial clouds. Results and Conclusions

32 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results These instances use requirements of five real applications: an algorithm for the Quadratic Assignment Problem that can be solved by a branch-and-bound algorithm; three applications related to manipulation of biologic sequences: RAXML, ModelGenerator and Segemehl; a typical analysis user job for the CMS experiment running at the LHC/CERN grid. They use packages of two commercial clouds: Amazon EC2, with package groups of high performance computing (HPC) clusters Google Cloud Platform, with packages from the Google Compute Engine Results and Conclusions

33 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results  Accomplished experiments: 1. An evaluation of the CC-IP mathematical formulation 2. An evaluation of the GraspCC heuristic Results and Conclusions

34 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results  Accomplished experiments: 1. An evaluation of the CC-IP mathematical formulation 2. An evaluation of the GraspCC heuristic Results and Conclusions

35 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results  Accomplished experiments: 1. An evaluation of the CC-IP mathematical formulation 2. An evaluation of the GraspCC heuristic Results and Conclusions The objectives of the cost function were normalized due to their distinct range values. - They share the same minimum and maximum values (0 and 1). - The payment cost was divided by the most expensive package cost times the maximum time informed by the cloud consumer. - The execution time was divided by the maximum time informed by the cloud consumer.

36 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Results and Conclusions

37 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Results and Conclusions For example, nug24-cbb with packages of Amazon (i) 3 time units for execution time and $46.20 for payment cost. Packages set: 8 packages of type 2 (ii) 2 time units for execution time and $84.00 for payment cost. Packages set: 20 packages of type 2

38 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Comparison of the GraspCC with the CC-IP Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution Value Total Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00 Results and Conclusions

39 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Comparison of the GraspCC with the CC-IP Results and Conclusions Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution ValueTotal Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00

40 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Comparison of the GraspCC with the CC-IP Results and Conclusions Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution ValueTotal Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00

41 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Comparison of the GraspCC with the CC-IP Results and Conclusions Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution ValueTotal Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00

42 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Comparison of the GraspCC with the CC-IP Results and Conclusions Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution ValueTotal Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00

43 Outline Motivation Definition and Formulation GRASP Results and Conclusions Preliminary Results Comparison of the GraspCC with the CC-IP Results and Conclusions Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution ValueTotal Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00

44 Outline Motivation Definition and Formulation GRASP Results and Conclusions Instances GraspCCCC-IP Function cost Solution value Total Time Gap (%) Function cost Solution Value Total Time Time value Payment Time Value Payme nt nug22-sbb_am120.045614.50.010.00000.045614.50.59 nug28-sbb_am720.0128140.80.070.00000.0128140.827.23 nug22-cbb_am120.0519111.80.030.00000.0519111.80.77 nug28-cbb_am720.0123136.90.070.00000.0123136.921.20 nug30-cbb_am840.2652231034.73.630.00000.2652231034.7118.42 segemehl_am4 0.1688 116.80.070.0000 0.1688 116.80.26 cms-1500_am240.16254410.40.530.00000.16254410.432.83 nug22-sbb_go120.049714.95.360.00660.049714.977.72 nug28-sbb_go720.0655687.4133.660.00000.0655687.435278.98 nug22-cbb_go120.1137218.617.970.00000.1137218.6246.79 nug28-cbb_go720.0541570.760.000.08590.0552570.686400.00 nug30-cbb_go840.8502811573.22987.790.00000.8502811573.25085.60 segemehl_go40.5038326.252.400.0000 0.5038 326.223.52 cms-1500_go240.479613573.553.270.0000 0.4797 13573.986400.00 Preliminary Results Comparison of the GraspCC with the CC-IP Results and Conclusions GraspCC presented an outstanding improvement of the execution time, in average 99% less than the execution time of CC-IP.

45 HeteroPar 2013 Optimization of a Cloud Resource Management Problem from a Consumer Perspective Rafaelli de C. Coutinho, Lucia M. A. Drummond and Yuri Frota Fluminense Federal University – UFF Brazil Sponsorship:

46 Outline Motivation Definition and Formulation GRASP Results and Conclusions Local Search Phase - lsCC GRASP


Download ppt "HeteroPar 2013 Optimization of a Cloud Resource Management Problem from a Consumer Perspective Rafaelli de C. Coutinho, Lucia M. A. Drummond and Yuri Frota."

Similar presentations


Ads by Google