Cloud gaming has become increasingly more popular in the academia and the industry, evident by the large numbers of related research papers and startup companies. Some public cloud gaming services have attracted hundreds of thousands subscribers, demonstrating the initial success of cloud gaming services. Pushing the cloud gaming services forward, however, faces various challenges, which open up many research opportunities. In this paper, we share our views on the future cloud gaming research, and point out several research problems spanning over a wide spectrum of different directions: including distributed systems, video codecs, virtualization, human-computer interaction, quality of experience, resource allocation, and dynamic adaptation. Solving these research problems will allow service providers to offer high-quality cloud gaming services yet remain profitable, which in turn results in even more successful cloud gaming eco-environment. In addition, we believe there will be many more novel ideas to capitalize the abundant and elastic cloud resources for better gaming experience, and we will see these ideas and associated challenges in the years to come.

1. Cloud Gaming Onward:
Research Opportunities and
Outlook
Kuan-Ta Chen, Chung-Ying Huang,
and Cheng-Hsin Hsu

2. Cloud Gaming is Hot
Cloud gaming is expected to lead the future
growth of computer games: 9 times in 6 years
[CGR]
[CGR] http://www.cgconfusa.com/report/documents/Content-5minCloudGamingReportHighlights.pdf
T5-Labs

3. Cloud Gaming Onward / Kuan-Ta Chen 4
Limitations of Existing Services
OnLive demands for 5 Mbps for reasonable quality
OnLive dictates a server rendering/processing
latency of 100+ ms, and partially copes with it by
setting up 5 data centers (CA, VA, TX, IL, GA)
Only people who live in 1000 mile radius from a data center can play
the game
and more…
We, researchers, have tons of ideas to improve
cloud gaming services, but all existing cloud gaming
systems are proprietary and closed

4. Cloud Gaming Onward / Kuan-Ta Chen 5
Solutions: GamingAnywhere is the first
OPEN cloud gaming platform for
researchers, service providers, and users

5. Cloud Gaming Onward / Kuan-Ta Chen 6
GA Has Lower Response Delay
Low response delay

6. Cloud Gaming Onward / Kuan-Ta Chen 7
Support of Android Devices
Implement three proof-of-concept controllers,
designed for
Nintendo 64
Nintendo DS
Limbo
Nintendo 64 Controller Limbo Controller

7. Cloud Gaming Onward / Kuan-Ta Chen 8
Demo

8. • 25k+ visitors, 70k+ downloads, and
400+ forum discussions since April 2013
http://gaminganywhere.org/

9. Cloud Gaming Onward / Kuan-Ta Chen 10
Research Opportunities
Game Integration
Video Codec
Virtualization
User Interface
QoE Measurement and Modeling
Server Selection
Parameter Adaptation
Resource Scheduling

11. Cloud Gaming Onward / Kuan-Ta Chen 12
Game Integration
Two facets
When/how game screens are captured
How game inputs are replayed
De facto approach
Use API hooking to intercept events
related to screen rendering
Use API hooking to emulate user
inputs
Goal
A general framework to bridge
cloud gaming middleware and games

12. Cloud Gaming Onward / Kuan-Ta Chen 13
Video Codec
H.264 is commonly used, but not necessarily the most
efficient codec for gaming
Network bandwidth is always scarce (e.g., continuous &
stable 5 Mbps avail BW cannot be assumed)
Considering the huge variety in graphics style…

13. Video Codec (cont.)
E.g., [8] proposed a layered coding approach
that can reduce bandwidth usage significantly
[8] S.-P. Chuah and N.-M. Cheung, “Layered coding for mobile cloud gaming,” in
Proceedings of ACM MMVE 2014.

14. Cloud Gaming Onward / Kuan-Ta Chen 15
Virtualization
GPU virtualization is not as mature as virtualization of
other resources (e.g., CPU, memory)
 Scalability problems
 financial difficulty of service providers
Mediated GPU pass-through technology might be the
solution

15. Cloud Gaming Onward / Kuan-Ta Chen 16
User Interface Mapping
?

16. Cloud Gaming Onward / Kuan-Ta Chen 17
QoE Measurement and Modeling
Challenges
Low cost continuous QoE assessment
Large parameter space (network delay/bandwidth/packet
loss, codec, game, display, control, and so on)
Common approaches not sufficient
Absolute Categorial Rating (ACR)
SSCQE, DSCQE
Possible solutions
Paired comparison [14] (not continuous and still costly)
Physiological approach [15]

17. Facial EMG Measurement
The corrugator
supercilii muscle
Negative emotions
The amount of annoyance
caused by latency
[15] Y.-T. Lee, K.-T. Chen, H.-I. Su, and C.-L. Lei, “Are all games equally cloud-gaming-friendly? An electromyographic approach,” in Proceedings of
the IEEE/ACM Annual Workshop on Network and Systems Support for Games (NetGames’12), October 2012, pp. 3:1–3:6.

18. Cloud Gaming Onward / Kuan-Ta Chen 19
Quantifying QoE Degradation due to
Latency in Cloud Gaming

19. Cloud Gaming Onward / Kuan-Ta Chen 20
Server Selection
The closest server may not be available due to
Overloaded
Not powerful enough for particular games
Not installed with particular games
Also, we need to consider
Multi-user games where multiple players participates in a
game session simultaneously
Online games that both (player <-> cloud gaming server
delay) and (cloud gaming server <->
online game server delay) need to
be considered

20. Cloud Gaming Onward / Kuan-Ta Chen 21
Parameter Adaptation for
Real-time Graphics Streaming
Control-theoretic algorithms are required to adapt to
changing network conditions
Huge parameter space
Video codec
 Layer design: MDC, SVC, among others
 Various parameters associated with video codec
Network delivery
 Network condition measurement
 Path selection (overlay?)
 Multi-path delivery
Playout
 Error concealment
 Playout buffering
And lots more…

21. Cloud Gaming Onward / Kuan-Ta Chen 22
Resource Scheduling
Different games have different requirements on
CPU, GPU, memory, and disks
Car racing games are disk read intensive
Realistic shooting games are normally GPU intensive
RPG games are normally memory intensive
Such requirements may change over time
 CPU, GPU, memory,
and disk co-scheduling

22. Cloud Gaming Onward / Kuan-Ta Chen 23
To Make Cloud Gaming Scalable…
In terms of # cloud-enabled games
Game integration framework
User interface mapping
In terms of # concurrent players
GPU virtualization
Resource scheduling
In terms of market adoption (UX)
User interface, Video codec
QoE assessment and management (including parameter
adaptation and server selection)

23. Cloud Gaming Onward / Kuan-Ta Chen 24
Conclusion
Cloud gaming shares similar fundamental
problems with many interesting applications
Screencasting
Mobile smart lens
Tele medicine
Immersive remote
communications
Thus, cloud gaming can
be seen a rewarding entrance to fundamental
multimedia system challenges!

24. Cloud Gaming Onward / Kuan-Ta Chen 25
Kuan-Ta Chen
Academia Sinica
cloud gaming
rocks!
Thank You!
http://www.iis.sinica.edu.tw/~swc