Chapter 4 Simulation Results
4.4 Comparison of three resource allocation methods
We implemented multi-server, ANFIS-based DLP and DLP+SVMS with ANFIS methods. Experimental results show that the average access time (queuing time + CPU time) of the proposed ANFIS-based DLP+SVMS resource allocation method is 16.7% shorter than that of the ANFIS-based DLP method, as shown in Figure 16. In Figure 17, we show the VMS usages of the three resource allocation methods. The proposed ANFIS-based DLP+SVMS method has the smallest number of VMSs used among the three methods.
Figure 16: Average access time among three resource allocation methods.
25
Figure 17: Number of VMSs used among the three resource allocation methods.
26
Chapter 5 Conclusion
5.1 Concluding remarks
There are two phases in the proposed dynamic resource allocation method: load prediction phase and resource allocation phase. In the load prediction phase, we collected historical game data which includes CPU, memory and network loads from a popular MMOG, Lineage. We have designed and simulated an artificial neural network (ANN) and an adaptive neural fuzzy inference system (ANFIS) to predict an appropriate resource allocation policy to be executed in each game zone.
Experimental results show that in the load prediction phase, the mean square error and prediction time of the ANFIS-based load prediction scheme are lower than those of the ANN-based load prediction scheme. In the resource allocation phase, the average access time (execution time plus queuing time) of the proposed ANFIS-based deep-level partitioning (DLP) with secondary virtual machine servers (SVMSs) method is 16.7% shorter than that of the ANFIS-based DLP method. In addition, the proposed method has the smallest number of VMSs used among the three methods.
5.2 Future work
In our current design, we focused only on CPU, memory and network loads in a VMS. In the future, we will include the access time of storage devices in our load prediction. In addition, we will implement and evaluate our proposed load prediction methods and the proposed resource allocation policies in a real cloud computing environment.
27
References
[1] “Resource allocation,” [Online]. Available:
http://en.wikipedia.org/wiki/Resource_allocation.
[2] “Resource allocation for network communication,” [Online]. Available:
http://www.cs.nccu.edu.tw/~lien/VOIP/BBQresrc/.
[3] J. Muller, "On correctness of scalable multi-server state replication in online games," in Proceedings of the 5th ACM SIGCOMM Workshop on Network and System Support for Games, pp. 1-11, 2006.
[4] C. G. Dickey, "A fully distributed architecture for massively multiplayer online games,” in Proceedings of the 3rd ACM SIGCOMM Workshop on Network and System Support for Games, pp. 171, 2004.
[5] K.-W. Lee, B.-J. Ko, and S. Calo, "Adaptive server selection for large scale interactive online games," in Proceedings of Computer Networks, vol. 49, pp.
84-102, September 2005.
[6] B. Hariri, S. Shirmohammadi, M. R. Pakravan, and M. H. Alavi, “An adaptive latency mitigation scheme for massively multiuser virtual environments,” in Journal of Network and Computer Applications, 32(5), pp. 1049-1063, 2009.
[7] W. Kok Wai, "Resource allocation for massively multiplayer online games using fuzzy linear assignment technique," in Proceedings of the 5th IEEE on Consumer Communications and Networking Conference, pp. 1035-1039, January 2008.
[8] R. Prodan and V. Nae, "Prediction-based real-time resource provisioning for massively multiplayer online games," in Proceedings of Future Generation Computer Systems, vol. 25, pp. 785-793, July 2009.
28
[9] L. D. Brice, "Robust resource allocation in a massive multiplayer online gaming environment," in Proceedings of the 4th International Conference on Foundations of Digital Games, pp 232-239, 2009.
[10] J. Slegers, I. Mitrani, and N. Thomas, "Evaluating the optimal server allocation policy for clusters with on/off sources," in Proceedings of Performance Evaluation, vol. 66, pp. 453-467, August 2009.
[11] W. Streitberger and T. Eymann, "A simulation of an economic, self-organizing resource allocation approach for application layer networks," in Proceedings of Computer Networks, vol. 53, pp. 1760-1770, July 2009.
[12] R. Stanojevic and R. Shorten, "Load balancing vs. distributed rate limiting: an unifying framework for cloud control," in Proceedings of the IEEE International Conference on Communications, pp. 1-6, August 2009.
[13] E. Caron, F. Desprez, D. Loureiro, and A. Muresan, "Cloud computing resource management through a grid middleware: a case study with DIET and Eucalyptus,"
in Proceedings of the IEEE International Conference on Cloud Computing, pp.
151-154, September 2009.
[14] C. Yang, W. Tianyu, and L. Jianxin, "An efficient resource management system for on-line virtual cluster provision," in Proceedings of the IEEE International Conference on Cloud Computing, pp. 72-79, December 2009.
[15] Y. T. Lee and K. T. Chen, “Is server consolidation beneficial to MMORPG? a case study of World of Warcraft,” in Proceedings of the IEEE 3rd International Conference on Cloud Computing (CLOUD), pp 435-442, August 2010.
[16] C. C. Lee, “Design and implement dynamic load balancing mechanism for MMOG,” [Online]. Available: http://www.cs.npue.edu.tw.
[17] S. Zhang, S Zhang, X. Chen and X. Huo, “Cloud computing research and development trend,” in Proceedings of the Second International Conference on
29
Future Networks, pp 39-97, March 2010.
[18] “Hadoop and Map-Reduce,” [Online.] Available:
http://dotnetmis91.blogspot.com/2010/04/vs-hadoop-mapreduce.html.
[19] D. T. Ahmed and S. Shirmohammadi “Uniform and non-uniform zoning for load balancing in virtual environments,” in Proceedings of 5th International Conference on Embedded and Multimedia Computing (EMC), pp. 1-6, August 2010.
[20] D. T. Ahmed and S. Shirmohammadi, “An auxiliary area of interest management for synchronization and load regulation in zonal P2P MMOGs,” in Proceedings of IEEE International Workshop on Haptic Audio visual Environments and Games, pp. 36-41, October 2009.
[21] “Lineage,” [Online.] Available: http://tw.beanfun.com/lineage/index.aspx.
[22] “World of Warcraft,” [Online.] Available:
http://www.wowtaiwan.com.tw/main_index.asp.
[23] M.R. Head, A. Kochut, C. Schulz and H. Shaikh, “Virtual hypervisor: enabling fair and economical resource partitioning in cloud environments,” in Proceedings of Network Operations and Management Symposium (NOMS), pp 104-111, June 2010.
[24] D. T. Ahmed and S. Shirmohammadi, “A microcell oriented load balancing model for collaborative virtual environments,” in Proceedings of IEEE Conference on Virtual Environments, Human-Computer Interfaces and Measurement Systems, pp. 86-91, August 2008.
[25] J. Wang and Z. Yue, “A finding less-loaded server algorithm based on MMOG and analysis,” in Proceedings of IEEE Conference on Intelligent Computation Technology and Automation (ICICTA), pp. 96-99, July 2010.
[26] C. Bezerra, J. Comba and C. Geyer, “A fine granularity load balancing technique for MMOG servers using a KD-tree to partition the space,” in Proceedings of
30
VIII Brazilian Symposium on Games and Digital Entertainment (SBGAMES), pp.
17-26, June 2009.
[27] V. Nae, A. Iosup and R. Prodan, “Dynamic resource provisioning in massively multiplayer online games,” in the IEEE Transactions on Parallel and Distributed Systems, pp. 380-395, January 2010.
[28] A. K. Jain, M. Jianchang and K. M. Mohiuddin, “Artificial neural network: a tutorial,” in the IEEE Computer, pp. 31-44, August 2002.
[29] “MATLAB,” [Online.] Available: http://www.mathworks.com/.
[30] “CloudSim,” [Online.] Available: http://www.cloudbus.org/cloudsim/.
[31] J. S. R. Jang, “ANFIS: adaptive-network-based fuzzy inference system,” in the IEEE Transactions on Man and Cybernetics Systems, pp. 665-685, May 1993.