004.45

.. , ..
, , () . , . , , . . PaaS-. , Mjolnirr.
: , , , , , Mjolnirr.

, , , , . : , , , . , , . , CloudSim.
. 1 , , , . 2 , CloudSim, CDOSim, TeachCloud, SPECI, DCSim. 3 : , . 4 PaaS-. PaaS- Mjolnirr. .
1.
[1] (. 1).
, ( , ..) ( , .). , . , .



&
. 1.
( , , ) (hyper-threading, , .). , , . , . Grid'5000 [2] Planet-Lab [3] .
, . . . , (, Microgrid [4]). . , ( ).
, [5]. HPL Linpack [6] NAS [6]. , , , / .. -
, , .
, . , . , , , . , , . . , GridSim [8], SimGrid [9], CloudSim [10] .
2.
, . GridSim, SimGrid CloudSim. -, CloudSim , .
, - .
, , , . , .
2.1. CloudSim
CloudSim , [10, 11]. GridSim, , -, .
. 2 CloudSim.
CloudSim
() (Cloudlet)


CPU RAM

,



CloudSim
. 2. CloudSim
CloudSim SimJava, , , (, , , ), . , , .
. . , , PaaS .
, . , :
, ;
;
. , CloudSim, . CloudSim, .
, CloudSim , .
2.2. CDOSim
CDOSim [12] (Cloud Deployment Options Simulator) , . CDOSim
, . CDOSim :
;
;
, ;
.
2.3. TeachCloud
TeachCloud [13] , . TeachCloud , , .
2.4. iCanCloud
iCanCloud [14]. iCanCloud . , . , iCanCloud .
2.5. SPECI
SPECI [15] (Simulation Program for Elastic Cloud Infrastructures) . SPECI , , . SPECI » .
2.6. DCSim
DCSim (Data Center Simulator) , IaaS- [16], . DCSim IaaS-.
3.
. , .
[17]:
, .. ;
, .. ;
, .. .
, , , .
. :
;
;
.
3.1.
, , [18].
, . , , . 3.
, » . , . , .
. 3.
3.2.
, , (. 4), [19].
. 4.

, , . , .
3.3.
, , .
.
(. 5). , . : , , , .
. 5.
(. 6). . . : , , , . , , .
. 6.
4. PaaS-
, PaaS-. ,
CloudSim , -
. 7. PaaS-
PaaS- (. 7):
1. Task , , . , , .
2. Container , . , .
3. WorkloadGenerator , . , .
4. Synchronizer , . , ,
5. Broker , . , , .
6. Controller , , .

JSON , -
, . :
datacenter ;
workload .
datacenter ( hosts), ( vms), ( brokers). hosts :
quantity ;
ram ;
storage ;
bandwidth ;
CPUs , (mips).
vms , :
quantity ;
cpu ;
size ;
ram ;
bandwidth ;
mips .
brokers , . PaaS- . com.model.wrapper.broker. TasksScheduler com.model.wrapper.broker.AppsScheduler (. 8).
package com.model.wrapper.broker; import java.util.List;
import com.model.wrapper.cloudlet.Task; import com.model.wrapper.vm.Container; import com.model.wrapper.App;
public interface TasksScheduler {
public void scheduleTasks(List<Task> tasks, List<Container> containers);
}
public interface AppsScheduler {
public void distributeApps(List<Task> tasks,
List<App> apps, List<Container> containers); public void redistributeAppsFor(Task task, List<Container> containers);
}
. 8. TaskScheduler AppsScheduler
workload , . :
quantity ;
length , ;
fileSize ;
outputSize ;
app , ;
subtasks , , -.
5. PaaS-
PaaS-, Mjolnirr [20].
. Mjolnirr , 11 , 4- 2 , 512 . : 2 Intel Xeon X5680 (6 , 3.33 GHz) 12 GB DDR3 RAM.
1 100 10 . . . , . 9.
1400

200

1
2
3
4
5
6
7
S

10

" -
. 9.
, Mjolnirr 5.2 %, , .

, . . , . , PaaS-. , .
, -, PaaS-.
( 14-07-00420).

1. Gustedt, J. Experimental methodologies for large-scale systems: a survey / J. Gustedt, E. Jeannot, M. Quinson // Parallel Process. Lett. World Scientific, 2009. Vol. 19.
P. 399-418.
2. Bolze, R. Grid'5000: A Large Scale And Highly Reconfigurable Experimental Grid Testbed / R. Bolze, F. Cappello, E. Caron, M. Dayde, F. Desprez et al. / / Int. J. High Perform. Comput. Appl. USA: Sage Publications, 2006. Vol. 20. P. 481-494.
3. Chun, B. Planetlab: an overlay testbed for broad-coverage services / B. Chun, D. Culler, T. Roscoe // ACM SIGCOMM. USA: ACM, 2003. Vol. 33. P. 3-12.
4. Song, H.J. The MicroGrid: a Scientific Tool for Modeling Computational Grids / H.J. Song // Proc. IEEE Supercomput. USA: IEEE, 2000. P. 4-10.
5. , .. /.. // . . . : , 2011. . 3. C. 105-110.
6. Endo, T. Linpack evaluation on a supercomputer with heterogeneous accelerators / T. Endo // Parallel & Distrib. Process. (IPDPS), 2010 IEEE Int. Symp. USA: IEEE, 2010. P. 1-8.
7. Bailey, D.H. NAS parallel benchmark results / D.H. Bailey / / Proc. Supercomput. '92.
USA: IEEE, 1992. P. 1-13.
8. Buyya, R. GridSim: a toolkit for the modeling and simulation of distributed resource management and scheduling for Grid computing / R. Buyya, M. Murshed / / Concurr. Comput. Pract. Exp. USA: Wiley, 2002. Vol. 14. 13-15. P. 1175-1220.
9. Quinson, M. SimGrid: a generic framework for large-scale distributed experiments / M. Quinson // 2009 IEEE Ninth Int. Conf. Peer-to-Peer Comput. USA: IEEE, 2009. P. 126-131.
10. Calheiros, R.N. CloudSim: A Novel Framework for Modeling and Simulation of Cloud Computing Infrastructures and Services./ R.N. Calheiros. Eprint: Australia, 2009. 9 p.
11. Buyya, R. Modeling and simulation of scalable Cloud computing environments and the CloudSim toolkit: Challenges and opportunities / R. Buyya, R. Ranjan, R.N. Calheiros // 2009 Int. Conf. High Perform. Comput. Simul. USA: IEEE, 2009. P. 1-11.
12. Fittkau, F. CDOSim: Simulating cloud deployment options for software migration support / F. Fittkau, S. Frey, W. Hasselbring // 2012 IEEE 6th Int. Work. Maint. Evol. Serv. Cloud-Based Syst. USA: IEEE, 2012. P. 37-46.
13. Jararweh, Y. TeachCloud: a cloud computing educational toolkit / Y. Jararweh et al. / / Int. J. Cloud Comput. 2012. InderScience Publ., 2012. Vol. 2. P. 237-257.
14. Nunez, A. iCanCloud: A Flexible and Scalable Cloud Infrastructure Simulator / A. Nunez // J. Grid Comput. 2012. Germany: Springer, 2012. Vol. 10. P. 185209.
15. Sriram, I. SPECI, a Simulation Tool Exploring Cloud-Scale Data Centres / I. Sriram // Lect. Notes Comput. Sci. 2009. Germany: Springer, 2009.- Vol. 5931. P. 381-392.
16. Keller, G. DCSim: A data centre simulation tool / G. Keller / / Integr. Netw. Manag. 2012. USA: IEEE, 2012. P. 1090-1091.
17. Li, W. Modeling for Dynamic Cloud Scheduling Via Migration of Virtual Machines / W. Li, J. Tordsson, E. Elmroth // 2011 IEEE Third Int. Conf. Cloud Comput. Technol. Sci. 2011. USA: IEEE, 2011. P. 163-171.
18. Maguluri, S.T. Stochastic models of load balancing and scheduling in cloud computing clusters / S.T. Maguluri, R. Srikant, L. Ying // INFOCOM, 2012 Proc. IEEE. USA: IEEE, 2012. P. 702-710.
19. Nidhi, K. Cloud Load Balancing Techniques: A Step Towards Green Computing / K. Nidhi, I. Chana // IJCSI Int. J. Comput. Sci. Issues. 2012. USA: Eprint, 2012. P. 238-246.
20. Savchenko, D. Mjolnirr: private PaaS as distributed computing evolution / D. Savchenko, G. Radchenko // MIPRO 2014. Proceedings of the 37th International Convention, 2014. USA: IEEE. P. 386-391.
, , - (, ), prohormihailov@ gmail .com
, ..-.., , - (, ), gleb.radchenko@susu.ru.
4 2014 .
Bulletin of the South Ural State University Series "Computational Mathematics and Software Engineering"
2014, vol. 3, no. 3, pp. 109123
MODELING AND PERFORMANCE EVALUATION OF CLOUD SYSTEMS
P.A. Mihailov, South Ural State University (Chelyabinsk, Russian Federation),
G.I. Radchenko, South Ural State University (Chelyabinsk, Russian Federation)
During usage of industrial grid and cloud systems, there are issues related to the changes in the structure and algorithms of distributed computing systems and how these changes will affect the system performance. The article describes the main approaches to the experimental study of methodologies for cloud systems. The strengths and weaknesses of approaches of natural modeling, simulation, benchmarking and simulation of cloud systems are evaluated. A brief review of systems simulation is provided. As the result of the analysis we present the design and implementation of a prototype of own system for simulation of private cloud PaaS-systems. We describe the implementation of the system, as well as the test results of the developed models on the example of Mjolnirr cloud platform.
: distributed computing systems, modeling, simulation, cloud computing, cloud, Mjolnirr.
References
1. Gustedt J., Jeannot E., Quinson M. Experimental methodologies for large-scale systems: a survey // Parallel Process. Lett. World Scientific. 2009. Vol. 19. P. 399-418.
2. Bolze R., Cappello F., Caron E., Dayde M., Desprez, F. et al. Grid'5000: A Large Scale And Highly Reconfigurable Experimental Grid Testbed / R. Bolze, F. Cappello, E. Caron, M. Dayde, F. Desprez et al. // Int. J. High Perform. Comput. Appl. USA: Sage Publications. 2006. Vol. 20. P. 481-494.
3. Chun B., Culler D., Roscoe T. Planetlab: an overlay testbed for broad-coverage services // ACM SIGCOMM. USA: ACM. 2003. Vol. 33. P. 3-12.
4. Song H.J. The MicroGrid: a Scientific Tool for Modeling Computational Grids Proc. IEEE Supercomput. USA: IEEE. 2000. P. 4-10.
5. Korsukov A.S. Instrumentalnyye sredstva polunaturnogo modelirovaniya raspredelennyh vychislitelnyh system [Tools for seminatural simulation of distributed computing systems] // Sovremennyye tekhnologii. Sistemnyy analiz. Modelirovaniye. [Modern technology. System analysis. Modeling]. Russia: Irkutsk State University of Railway Transport. 2011. Vol. 3. P. 105-110.
6. Endo T. Linpack evaluation on a supercomputer with heterogeneous accelerators / / Parallel & Distrib. Process. (IPDPS), 2010 IEEE Int. Symp. USA: IEEE, 2010. P. 1-8.
7. Bailey D.H. NAS parallel benchmark results // Proc. Supercomput. '92. USA: IEEE, 1992. P. 1-13.
8. Buyya R., Murshed M. GridSim: a toolkit for the modeling and simulation of distributed resource management and scheduling for Grid computing / / Concurr. Comput. Pract. Exp. USA: Wiley. 2002. Vol. 14. 13-15. P. 1175-1220.
9. Quinson M. SimGrid: a generic framework for large-scale distributed experiments / / 2009 IEEE Ninth Int. Conf. Peer-to-Peer Comput. USA: IEEE. 2009. P. 126 131.
10. Calheiros R.N. CloudSim : A Novel Framework for Modeling and Simulation of Cloud Computing Infrastructures and Services. Eprint: Australia, 2009. 9 p.
11. Buyya R., Ranjan R., Calheiros R.N. Modeling and simulation of scalable Cloud computing environments and the CloudSim toolkit: Challenges and opportunities // 2009 Int. Conf. High Perform. Comput. Simul. USA: IEEE, 2009. P. 1-11.
12. Fittkau F., Frey S., Hasselbring W. CDOSim: Simulating cloud deployment options for software migration support / / 2012 IEEE 6th Int. Work. Maint. Evol. Serv. Cloud-Based Syst. USA: IEEE. 2012. P. 37-46.
13. Jararweh Y. et al. TeachCloud: a cloud computing educational toolkit // Int. J. Cloud Comput. 2012. InderScience Publ. 2012. Vol. 2. P. 237-257.
14. Nunez A. iCanCloud: A Flexible and Scalable Cloud Infrastructure Simulator // J. Grid Comput. 2012. Germany: Springer. 2012. Vol. 10. P. 185-209.
15. Sriram I. SPECI, a Simulation Tool Exploring Cloud-Scale Data Centres // Lect. Notes Comput. Sci. 2009. Germany: Springer. 2009. Vol. 5931. P. 381-392.
16. Keller G. DCSim: A data centre simulation tool // Integr. Netw. Manag. 2012. USA: IEEE. 2012. P. 1090-1091.
17. Li W., Tordsson J., Elmroth E. Modeling for Dynamic Cloud Scheduling Via Migration of Virtual Machines // 2011 IEEE Third Int. Conf. Cloud Comput. Technol. Sci. 2011. USA: IEEE, 2011. P. 163-171.
18. Maguluri S.T., Srikant R., Ying L. Stochastic models of load balancing and scheduling in cloud computing clusters // INFOCOM, 2012 Proc. USA: IEEE. 2012. P. 702-710.
19. Nidhi K. Chana I. Cloud Load Balancing Techniques : A Step Towards Green Computing // IJCSI Int. J. Comput. Sci. Issues. 2012. USA: Eprint. 2012. P. 238-246.
20. Savchenko D., Radchenko G. Mjolnirr: private PaaS as distributed computing evolution // MIPRO 2014. Proceedings of the 37th International Convention, 2014. USA: IEEE. P. 386-391.
Received 4 August 2014-