Abstract
Cloud gaming, wherein game graphics is rendered in the cloud and streamed back to the user as real-time video, expands the gaming market to billions of users who do not have gaming consoles or high-power graphics PCs. Companies like Nvidia, Amazon, Sony and Microsoft are investing in building cloud gaming platforms to tap this large unserved market. However, cloud gaming requires the user to have high-bandwidth and stable network connectivity – whereas a typical console game needs about 100–200 kbps, a cloud game demands minimum 10–20 Mbps. This makes the Internet Service Provider (ISP) a key player in ensuring the end-user’s good gaming experience.
In this paper we develop a method to detect Nvidia’s GeForce NOW cloud gaming sessions over their network infrastructure, and measure associated user experience. In particular, we envision ISPs taking advantage of our method to provision network capacity at the right time and in the right place to support growth in cloud gaming at the right experience level; as well as identify the role of contextual factors such as user setup (browser vs app) and connectivity type (wired vs wireless) in performance degradation. We first present a detailed anatomy of flow establishment and volumetric profiles of cloud gaming sessions over multiple platforms, followed by a method to detect gameplay and measure key experience aspects such as latency, frame rate and resolution via real-time analysis of network traffic. The insights and methods are also validated in the lab for XBox Cloud Gaming platform. We then implement and deploy our method in a campus network to capture gameplay behaviors and experience measures across various user setups and connectivity types which we believe are valuable for network operators.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We have obtained ethics clearance (UNSW Human Research Ethics Advisory Panel approval number HC211007) which allows us to analyze campus network traffic to infer application usage behaviors. Note that user identities remain anonymous – no attempt is made to extract or reveal any personal user information, and all results presented are aggregated across the campus.
References
Akbari, I., et al.: A look behind the curtain: traffic classification in an increasingly encrypted web. Proc. ACM Meas. Anal. Comput. Syst. (2021)
Amazon. Luna (2023). https://www.amazon.com/luna/landing-page. Accessed 12 Jan 2023
Bartolomeo, G., Cao, J., Su, X., Mohan, N.: Characterizing distributed mobile augmented reality applications at the edge. In: Proceedings of ACM CoNEXT, Paris, France (2023)
Bhuyan, S., Zhao, S., Ying, Z., Kandemir, M.T., Das, C.R.: End-to-end characterization of game streaming applications on mobile platforms. In: Proceedings of ACM Measurement and Analysis of Computing Systems (2022)
Business. Microsoft, Activision Blizzard and the Future of Gaming. The Economist (2022)
Cai, W., et al.: A survey on cloud gaming: future of computer games. IEEE Access 4, 7605–7620 (2016)
Carrascosa, M., Bellalta, B.: Cloud-gaming: analysis of google stadia traffic. Comput. Commun. 188, 99–116 (2022)
Chen, H., et al.: T-gaming: a cost-efficient cloud gaming system at scale. IEEE Trans. Parallel Distrib. Syst. 30(12), 2849–2865 (2019)
Chen, K.-T., Chang, Y.-C., Hsu, H.-J., Chen, D.-Y., Huang, C.-Y., Hsu, C.-H.: On the quality of service of cloud gaming systems. IEEE Trans. Multimedia 16(2), 480–495 (2014)
CloudFlare. What is SNI? How TLS Server Name Indication Works (2022). https://www.cloudflare.com/en-gb/learning/ssl/what-is-sni/. Accessed 12 Jan 2023
Di Domenico, A., Perna, G., Trevisan, M., Vassio, L., Giordano, D.: A network analysis on cloud gaming: stadia, GeForce now and PSNow. Network 1(3), 247–260 (2021)
Ghobaei-Arani, M., Khorsand, R., Ramezanpour, M.: An autonomous resource provisioning framework for massively multiplayer online games in cloud environment. J. Netw. Comput. Appl. 142, 76–97 (2019)
Gitnux. Cloud Gaming Services: A Look at the Latest Statistics (2023). https://blog.gitnux.com/cloud-gaming-services-statistics/#content. Accessed 26 June 2023
Graff, P., Marchal, X., Cholez, T., Mathieu, B., Festor, O.: Efficient identification of cloud gaming traffic at the edge. In: Proceedings of IEEE/IFIP Network Operations and Management Symposium (2023)
Han, Y., Guo, D., Cai, W., Wang, X., Leung, V.C.M.: Virtual machine placement optimization in mobile cloud gaming through QoE-oriented resource competition. IEEE Trans. Cloud Comput. 10(3), 2204–2218 (2022)
Illahi, G.K., Gemert, T.V., Siekkinen, M., Masala, E., Oulasvirta, A., Ylä-Jääski, A.: Cloud gaming with foveated video encoding. ACM Trans. Multimedia Comput. Commun. Appl. 16, 1 (2020)
Iqbal, H., Khalid, A., Shahzad, M.: Dissecting cloud gaming performance with DECAF. In: Proceedings of ACM Measurement and Analysis of Computing Systems (2021)
Kämäräinen, T., Siekkinen, M., Ylä-Jääski, A., Zhang, W., Hui, P.: A measurement study on achieving imperceptible latency in mobile cloud gaming. In: Proceedings of ACM MMSys, Taipei, Taiwan (2017)
Kinsta. What is a Content Management System (CMS)? (2022). https://kinsta.com/knowledgebase/content-management-system/. Accessed 12 Jan 2023
Ky, J., Graff, P., Mathieu, B., Cholez, T.: A hybrid P4/NFV architecture for cloud gaming traffic detection with unsupervised ML. In: Proceedings of IEEE Symposium on Computers and Communications, Los Alamitos, CA, USA (2023)
Ky, J.R., Mathieu, B., Lahmadi, A., Boutaba, R.: Assessing unsupervised machine learning solutions for anomaly detection in cloud gaming sessions. In: Proceedings of IEEE International Conference on Network and Service Management, Thessaloniki, Greece (2022)
Ky, J.R., Mathieu, B., Lahmadi, A., Boutaba, R.: ML models for detecting QoE degradation in low-latency applications: a cloud-gaming case study. IEEE Trans. Netw. Serv. Manag. (2023)
Li, Y., et al.: GAugur: quantifying performance interference of colocated games for improving resource utilization in cloud gaming. In: Proceedings of ACM HPDC, Phoenix, AZ, USA (2019)
Liu, S., et al.: AMIR: active multimodal interaction recognition from video and network traffic in connected environments. In: Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (2023)
Livingood, J.: Comcast Kicks Off Industry’s First Low Latency DOCSIS Field Trials (2023). https://corporate.comcast.com/stories/comcast-kicks-off-industrys-first-low-latency-docsis-field-trials. Accessed 26 June 2023
Madanapalli, S.C., Gharakheili, H.H., Sivaraman, V.: Know thy lag: in-network game detection and latency measurement. In: Proceedings of PAM (2022)
Madanapalli, S.C., Mathai, A., Gharakheili, H.H., Sivaraman, V.: Reclive: real-time classification and QoE inference of live video streaming services. In: Proceedings of IEEE/ACM IWQOS (2021)
Marchal, X., et al.: An analysis of cloud gaming platforms behaviour under synthetic network constraints and real cellular networks conditions. J. Netw. Syst. Manag. 31(2), 39 (2023)
Markets and Markets. Cloud Gaming Market by Offiering, Device Type, Solution, Game Type, Region - Global Forecast to 2024 (2019). https://bit.ly/3AyEjio. Accessed 27 Apr 2024
Microsoft. XBox Cloud Gaming (Beta) (2023). https://www.xbox.com/en-us/play. Accessed 12 Jan 2023
News. The Future of Video Games. The Economist (2023)
Nvidia. GeForce NOW (2023). https://www.nvidia.com/en-au/geforce-now/. Accessed 12 Jan 2023
Nvidia Support. How can i reduce lag or improve streaming quality when using geforce now? (2022). bit.ly/45TOmfR. Accessed 12 Dec 2022
Nvidia Support. WebRTC Browser Client (2022). bit.ly/3LhOPAj. Accessed 14 Dec 2022
Rietveld, J.: Microsoft and activision: the big questions that will decide whether the US\$68 billion deal goes ahead. The Conversation (2023)
Roy, S., Shapira, T., Shavitt, Y.: Fast and lean encrypted internet traffic classification. Comput. Commun. 186, 166–173 (2022)
Sabet, S.S., Schmidt, S., Zadtootaghaj, S., Griwodz, C., Möller, S.: Delay sensitivity classification of cloud gaming content. In: Proceedings of the 12th ACM International Workshop on Immersive Mixed and Virtual Environment Systems, Istanbul, Turkey (2020)
Schulzrinne, H., Casner, S., Frederick, R., Jacobson, V.: RTP: A Transport Protocol for Real-Time Applications. RFC 3550 (2003)
Sharma, T., Mangla, T., Gupta, A., Jiang, J., Feamster, N.: Estimating WebRTC video QoE metrics without using application headers (2023)
Shea, R., Liu, J., Ngai, E.C.-H., Cui, Y.: Cloud gaming: architecture and performance. IEEE Network 27(4), 16–21 (2013)
Slivar, I., Suznjevic, M., Skorin-Kapov, L.: Game categorization for deriving QoE-driven video encoding configuration strategies for cloud gaming. ACM Trans. Multimedia Comput. Commun. Appl. 14, 1–24 (2018)
Sony Interactive Entertainment. PlayStation Now (2023). https://www.playstation.com/en-us/ps-now/. Accessed 18 Apr 2023
Spang, B., Walsh, B., Huang, T.-Y., Rusnock, T., Lawrence, J., McKeown, N.: Buffer sizing and video QoE measurements at Netflix. In: Proceedings of the 2019 Workshop on Buffer Sizing (2020)
Wehner, N., Seufert, M., Schuler, J., Wassermann, S., Casas, P., Hossfeld, T.: Improving web QoE monitoring for encrypted network traffic through time series modeling. SIGMETRICS Perform. Eval. Rev. (2021)
Xu, X., Claypool, M.: Measurement of cloud-based game streaming system response to competing TCP cubic or TCP BBR flows. In: Proceedings of ACM Internet Measurement Conference, Nice, France (2022)
Zhang, X., et al.: Improving cloud gaming experience through mobile edge computing. IEEE Wirel. Commun. 26(4), 178–183 (2019)
Acknowledgements
We thank our shepherd Alessandro Finamore and the four anonymous reviewers for their insightful feedback. Funding for this project is provided by the Australian Government’s Cooperative Research Centres Projects (CRC-P) Grant CRCPXIV000099.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
A Flow Profiles of GeForce NOW Cloud Game Sessions Across User Setups
As complimentary to the flow profiles for cloud game sessions via desktop console application given in Fig. 3, the ones via browser and mobile console applications are shown in Fig. 12. Comparing the three figures, they all have different usage of service flows for platform administration and game session management purposes. Sessions via both mobile and desktop console applications have five gameplay flows each with a unique functionality, whereas those via browsers only have two gameplay flows, one for management, and one for combined user input and game media streaming.
B Network Traffic Characteristics of XBox Cloud Gaming Sessions
We have conducted similar lab experiments on another cloud gaming platform Microsoft’s XBox Cloud Gaming that is currently available in Australia. We discuss flow profiles of gameplay sessions on XBox Cloud Gaming platform and the applicability of our developed methods.
Figure 13 shows the usage of service flows during cloud gaming sessions on XBox Cloud Gaming platform accessed via three supported user setups including XBox hardware console (Fig. 13(a)), PC browser (Fig. 13(b)), and mobile browser (Fig. 13(c)), respectively. Compared to the evolution of service flows in Nvidia GeForce NOW we discussed in Sect. 3, we observe similar insights. First of all, the purposes of flows are also categorized into platform administration, platform management, and gameplay; second, prior to each gameplay session, platform management flows (annotated as “regional-node”) are started to check current network connectivity and select appropriate cloud server; third, RTP flows that are destinated to a certain range of port numbers (e.g., UDP|1040 to UDP|1190) on the cloud server are used for gaming media and user input. Similar to GeForce NOW, the services being accessed in platform sessions of XBox Cloud Gaming vary across user setup. As shown by example service names in Fig. 13(a), 13(b) and 13(c), sessions accessed by XBox hardware console, PC browser and mobile browser uses different graphic services namely xgpuconsole, xgpuweb and xgpu, respectively.
Apart from specific domain name (e.g., XBox Cloud Gaming uses xboxlive.com as its gameplay domain while GeForce NOW uses nvidiagrid.net) and range of service port numbers used on the cloud server that are different between XBox and GFN, we observed that XBox uses a single RTP flow for both streaming media and user input even on its native hardware console, while GFN uses separate RTP flows each only carry one type of traffic for sessions from console application.
It is not surprising to observe the above commonalities as cloud gaming platforms are built on similar technological paradigms and communication protocols. Therefore, our methods in detecting cloud gaming session, identifying user setup, and measuring gameplay QoE metrics are evidently applicable to XBox Cloud Gaming and other platforms with similar underlying structures.
C Charts Illustrating Classification of Gameplay Flows Using Heuristically Simplified Criteria
Figures 14 and 15 visually show the gameplay session flow classification processes for GeForce NOW and Xbox Cloud Gaming derived from our generic process shown in Fig. 8 with heuristically simplified criteria obtained from our training process on ground-truth traffic traces.
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Lyu, M., Madanapalli, S.C., Vishwanath, A., Sivaraman, V. (2024). Network Anatomy and Real-Time Measurement of Nvidia GeForce NOW Cloud Gaming. In: Richter, P., Bajpai, V., Carisimo, E. (eds) Passive and Active Measurement. PAM 2024. Lecture Notes in Computer Science, vol 14537. Springer, Cham. https://doi.org/10.1007/978-3-031-56249-5_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-56249-5_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-56248-8
Online ISBN: 978-3-031-56249-5
eBook Packages: Computer ScienceComputer Science (R0)