如何在雲端遊戲廠商以及玩家的遊戲品質體驗之間找到一個平衡點 非常複雜,進而導致最佳化 雲端遊戲體驗變成了一件不容易的工作。 我們解決了這項挑戰並且研究一個最佳化問提來 最大化雲端遊戲廠商 的利益並且同時讓玩家有足夠的遊戲品質體驗。我們測量並且得到了 遊戲品質體驗以及效能的數學模型,接著我們將問題轉換為數學式並 得出最佳解,但最佳 解需要指數倍的運算時間,所以我們發展一個有 效率的啟發式演算法。我們也為了封閉式的 雲端遊戲環境提出了另一 項方程式以及演算法,在這個情境下利益將不被考量,我們會最大化 玩家的遊戲品質體驗。我們利用現存的虛擬化技術實作出了一個系統 雛形以及小型實驗環境來驗證 我們演算法的效率以及實用性,我們的 實驗指引了雲端遊戲廠商如何創造他們自己的盈利環境。 接著我們延 伸我們的實驗到擁有實際數據的模擬器上,此實驗說明了: (i)此演算 法接近最佳解, (ii) 能夠有兩萬個主機以及四萬個玩家,(iii) 比現行 的虛擬機器配置演算法效能高出許多,例如: 擁有3.5倍的利益。在解 決了虛擬機器配置問題之後,我們對最新的顯示卡進行測量,進而回 答一 問提: 是否現行的顯示卡已經足夠供給雲端遊戲了呢?。與以前 的研究不同的地方,我們得到了 許多違背過去常識的結果。其一,最 新的顯示卡虛擬化技術可能會使分享式顯示卡的效能比專用的 顯示卡 虛擬化還好,其二,越多的工作轉換不一定會導致幀數下降。總的來 說,我們得知了最新的 顯示卡虛擬化技術已經足以分享給多個需要大 量顯示卡效能的雲端遊戲玩家。最後,我們發現使用 最新顯示卡的主 機的瓶頸可能會轉為處理器,而必須將影像的編碼從處理器移植到專 用的編解碼晶片 上來得到較好的遊戲品質體驗。

Optimizing cloud gaming experience is no easy task due to the complex tradeoff between gamer Quality of Experience (QoE) and provider net profit. We tackle the challenge and study an optimization problem to maximize the cloud gaming provider’s total profit while achieving just-good-enough QoE. Moreover, we conduct expeirments using a modern GPU and a cloud gaming platform to answer the following question: Are modern GPUs ready for cloud gaming? For the optimization problem, We conduct measurement studies to derive the QoE and performance models. We formulate and optimally solve the problem. The optimization problem has exponential running time, and we develop an efficient heuristic algorithm. We also present an alternative for- mulation and algorithms for closed cloud gaming services with dedicated in- frastructures, where the profit is not a concern and overall gaming QoE needs to be maximized. We present a prototype system and testbed using off-the- shelf virtualization software, to demonstrate the practicality and efficiency of our algorithms. Our experience on realizing the testbed sheds some lights on how cloud gaming providers may build up their own profitable services. Moreover, we conduct extensive trace-driven simulations to evaluate our pro- posed algorithms. The simulation results show that the proposed heuristic algorithms: (i) produce close-to-optimal solutions, (ii) scale to large cloud gaming services with 20000 servers and 40000 gamers, and (iii) outperform the state-of-the-art placement heuristic, e.g., by up to 3.5 times in terms of net profits. For the measurement study of modern GPU, the observations are different from earlier studies, our measurement results reveal several findings that are counter to common beliefs. First, with the latest GPU virtualization technique, shared GPUs may run faster than dedicated GPUs. Second, more context switches not necessarily lead to lower FPS (frame-per-second). In summary, we conclude that modern GPUs are powerful enough and can be shared by multiple GPU-intensive cloud games. Last, we present some sug- gestions for future cloud gaming platforms, e.g., the latest GPU servers may be CPU-bounded, which require the platforms to offload the video encoding from CPUs to dedicated codec chips for good gaming experience. 

[1] F. Bari, R. Boutaba, R. Esteves, M. Podlesny, G. Rabbani, Q. Zhang, F. Zhani, and L. Granville. Data center network virtualization: A survey. IEEE Communications Surveys & Tutorials, 15(2):909 – 928, 2012. Accepted to appear.
[2] K. Chen, Y. Chang, P. Tseng, C. Huang, and C. Lei. Measuring the latency of cloud gaming systems. In Proc. of ACM International Conference on Multimedia (MM’11), pages 1269–1272, Scottsdale, AZ, November 2011.
[3] M. Chen, H. Zhang, Y. Su, X. Wang, G. Jiang, and K. Yoshihira. Effective VM sizing in virtualized data centers. In Proc. of International Symposium on Integrated Network Management (IM’11), pages 594–601, Dublin, Ireland, May 2011.
[4] P. Chen and M. Zark. Perceptual view inconsistency: An objective evaluation frame- work for online game quality of experience (QoE). In Proc. of the Annual Workshop on Network and Systems Support for Games (NetGames’11), pages 2:1–2:6, Ottawa, Canada, October 2011.
[5] X. Cheng, S. Su, Z. Zhang, H. Wang, F. Yang, Y. Luo, and J. Wang. Virtual net- work embedding through topology-aware node ranking. ACM SIGCOMM Computer Communication Review, 41(2):38–47, April 2011.
[6] N. Chowdhury, M. Rahman, and R. Boutaba. Virtual network embedding with coor- dinated node and link mapping. In Proc. of IEEE INFOCOM 2009, pages 783–791, Rio de Janeiro, Brazil, April 2009.
[7] C.Clark,K.Fraser,S.Hand,J.Hansen,E.Jul,C.Limpach,I.Pratt,andA.Warfield. Live migration of virtual machine. Symposium on Networked Systems Design and Implementation, 2:273–286, 2005.
[8] M. Claypool. Motion and scene complexity for streaming video games. In Proc. of the International Conference on Foundations of Digital Games (FDG’09), pages 34–41, Port Canaveral, FL, April 2009.
[9] Cloud gaming adoption is accelerating . . . and fast! http://www.nttcom.tv/2012/07/09/ cloud-gaming-adoption-is-acceleratingand-fast/.
[10] IBM ILOG CPLEX optimizer. http://www-01.ibm.com/software/ integration/optimization/cplex-optimizer/.
[11] DigSites.http://digsitesvalue.net/s/onlive.com.
[12] Distribution and monetization strategies to increase revenues from cloud
gaming. http://www.cgconfusa.com/report/documents/ Content-5minCloudGamingReportHighlights.pdf.
[13] M. Dowty and J. Sugerman. GPU virtualization on VMware’s hosted I/O architec- ture. ACM SIGOPS Operating Systems Review, 43(3):73–82, July 2009.
[14] DummyNet.http://info.iet.unipi.it/ ̃luigi/dummynet/.
[15] T. Duong, X. Li, R. Goh, X. Tang, and W. Cai. QoS-aware revenue-cost optimiza- tion for latency-sensitive services in IaaS clouds. In Proc. of IEEE/ACM 16th In- ternational Symposium on Distributed Simulation and Real Time Applications (DS- RT’12), Dublin, Ireland, October 2012.
[16] T. Ferreto, M. Netto, R. Calheiros, and C. Rose. Server consolidation with mi- gration control for virtualized data centers. Future Generation Computer Systems, 27(8):1027–1034, October 2011.
[17] K. Gummadi, S. Saroiu, and S. Gribble. King: Estimating latency between arbitrary Internet end hosts. In Proc. of ACM SIGCOMM Internet Measurement Workshop (IMW’02), pages 5–18, Boston, MA, November 2002.
[18] H. Hong, D. Chen, C. Huang, K. Chen, and C. Hsu. QoS-aware virtual machine placement for cloud games. In Proc. of ACM Annual Workshop on Network and Systems Support for Games (NetGames’13), Denver, CO, December 2013.
[19] C. Huang, K. Chen, D. Chen, H. Hsu, and C. Hsu. GamingAnywhere: The first open source cloud gaming system. ACM Transactions on Multimedia Computing, Communications, and Applications, 10(1s):10:1–10:25, January 2014.
[20] C. Huang, C. Hsu, Y. Chang, and K. Chen. Gaminganywhere: An open cloud gaming system. In Proc. of the ACM Multimedia Systems Conference (MMSys’13), pages 36–47, Oslo, Norway, February 2013.
[21] ip2c.http://firestats.cc/wiki/ip2c.
[22] IsoHunt.http://isohunt.com/.
[23] M. Kutner, C. Nachtsheim, J. Neter, and W. Li. Applied linear statistical models. McGraw-Hill Higher Education, 5th edition, 2004.
[24] Y.LeeandK.Chen.IsserverconsolidationbeneficialtoMMORPG?acasestudyof World of Warcraft. In Proc. of IEEE International Conference on Cloud Computing (CLOUD’10), pages 435 – 442, Miami, FL, February 2010.
[25] Y. Lee, K. Chen, H. Su, and C. Lei. Are all games equally cloud-gaming-friendly? an electromyographic approach. In Proc. of the ACM SIGCHI International Con- ference on Advances in Computer Entertainment Technology (ACE’05), pages 117– 124, October 2012.
[26] Y. Li, W. Li, and C. Jiang. A survey of virtual machine system: Current technology and future trends. In Proc. of International Symposium on Electronic Commerce and Security (ISECS’10), pages 332–336, Guangzhou, China, July 2010.
[27] libtorrent.http://www.rasterbar.com/products/libtorrent/.
[28] J. Lin, C. Chen, and J. Chang. QoS-aware data replication for data-intensive ap- plications in cloud computing systems. IEEE Transactions on Cloud Computing, 1(1):101–115, July-December 2013.
[29] C. Mark and C. Kajal. Latency and player actions in online games. Communications of the ACM, 49(11):40–45, November 2006.
[30] M. Marzolla, O. Babaoglu, and F. Panzieri. Server consolidation in clouds through gossiping. In Proc. of International Symposium on World of Wireless, Mobile and Multimedia Networks (WoWMoM’11), pages 1–6, Lucca, Italy, June 2011.
[31] X.Meng,V.Pappas,andL.Zhang.Improvingthescalabilityofdatacenternetworks with traffic-aware virtual machine placement. In Proc. of IEEE INFOCOM 2010, pages 1–9, San Diego, CA, March 2010.
[32] S. Nanda and T. Chiueh. A survey of virtualization technologies. Technical report, February 2005. www.ecsl.cs.sunysb.edu/tr/TR179.pdf.
[33] R. Nathuji, A. Kansal, and A. Ghaffarkhah. Q-clouds: Managing performance in- terference effects for QoS-aware clouds. In Proc. of the European Conference on Computer systems (EuroSys’10), pages 237–250, Paris, France, April 2010.
[34] OnLivelaunchesnewcompanytoavoidbankruptcy.http://techland.time. com/2012/08/20/onlive.
[35] S.Osman,D.Subhraveti,G.Su,andJ.Nieh.Thedesignandimplementationofzap: a system for migrating computing environments. ACM SIGOPS Operating Systems Review, 36(SI):361–376, December 2002.
[36] R. Rose. Survey of system virtualization techniques. Technical report, March 2004. http://www.robertwrose.com/vita/rose-virtualization.pdf.
[37] P. Ross. Cloud computing’s killer app: Gaming. IEEE Spectrum, 46(3):14, March 2009.
[38] R. Shea and J. Liu. On GPU pass-through performance for cloud gaming: Exper- iments and analysis. In Proc. of ACM Annual Workshop on Network and Systems Support for Games (NetGames’13), Denver, CO, December 2013.
[39] S. Shi, K. Nahrstedt, and R. Campbell. Distortion over latency: Novel metric for measuring interactive performance in remote rendering systems. In Proc. of IEEE International Conference on Multimedia and Expo (ICME’11), pages 1–6, Barcelona, Spain, July 2011.
[40] B. Speitkamp and M. Bichler. A mathematical programming approach for server consolidation problems in virtualized data centers. IEEE Transactions on Services Computing, 3(4):266–278, December 2010.
[41] War of Warcraft avatar history dataset. http://mmnet.iis.sinica.edu. tw/dl/wowah/.
[42] D. Wu, Z. Xue, and J. He. iCloudAccess: Cost-effective streaming of video games from the cloud with low latency. IEEE Transactions on Circuits and Systems for Video Technology, January 2014. Accepted to appear.
[43] M. Yu, Y. Yi, J. Rexford, and M. Chiang. Rethinking virtual network embedding: Substrate support for path splitting and migration. ACM SIGCOMM Computer Com- munication Review, 38(2):17–29, April 2008.
[44] S. Zaman and D. Grosu. A combinatorial auction-based mechanism for dynamic VM provisioning and allocation in clouds. IEEE Transactions on Cloud Computing, 1(2):129–141, July-December 2013.
[45] C. Zhang, Z. Qi, J. Yao, M. Yu, and H. Guan. vGASA: Adaptive scheduling algo- rithm of virtualized GPU resource in cloud gaming. IEEE Transactions on Parallel and Distributed Systems, November 2013. Accepted to appear.
[46] Q. Zhu and T. Tung. A performance interference model for managing consolidated workloads in QoS-aware clouds. In IEEE International Conference on Cloud Com- puting (CLOUD’12), pages 170–179, Honolulu, HI, June 2012.