Jose Saldana, Mirko Suznjevic, Invited talk "The problem of using a best-effort network for online games," 10th IEEE International Workshop on Networking Issues in Multimedia Entertainment NIME'14 – , held in conjunction with Consumer Communications and Networking Conference, CCNC 2014. Las Vegas, Nevada, USA – January 10, 2014.

1. THE PROBLEM OF USING A BESTEFFORT NETWORK FOR ONLINE GAMES
NIME WORKSHOP. IEEE CCNC 2014

2. Index
I.
Real-time services on
the Internet
II. The two problems
III. Adapting the network

3. Real-time services
-
Real-time services are being widely used:
VoIP, video conference, online gaming

4. Real-time services
In this presentation, we will focus on
online gaming. Why?

5. Popularity of online games

6. Popularity of online games

7. The approach
Client-server architecture
- Consistency of the game
- Cheating avoidance
- Billing

8. How does it look? (On the packet level)
40 50 60 70 80 90 100 110
bytes
0
10 20 30 40 50 60 70
ms
Unreal
Tournament
40 50 60 70 80 90 100 110
bytes
0
10 20 30 40 50 60 70
ms
Counter
Strike I
40 50 60 70 80 90 100 110
bytes
0
10 20 30 40 50 60 70
ms
Quake III

9. Index
I.
Real-time services on
the Internet
II. The two problems
III. Adapting the network

10. -
Still there are problem(s)
Efficiency problem: Sending very frequent
updates containing small volume of
information is not efficient
-
-
Not everybody has fast connections with
excess bandwidth
Quality problem: Using a best effort
network for a real-time service
-
Online games present very stringent realtime requirements: players are really
difficult to satisfy
We need to measure quality, in terms of
network parameters.

11. Efficiency problem
If I have to send a big file, I divide it
in chunks of 1500 bytes, and I may
have an efficiency of 97%
One IPv4/TCP packet 1500 bytes
η=1460/1500=97%
IPv4
TCP

12. Efficiency problem
But real-time services cannot wait.
They need information every (e.g.)
20 ms, so efficiency gets really bad
(33%):
One IPv4/UDP/RTP VoIP packet with two samples of 10 bytes
η=20/60=33%
IPv4
UDP
RTP
codec G.729a, 2 samples

13. Quality problem: QoE for games
S. Möller, S. Schmidt, and J. Beyer, “Gaming Taxonomy: An Overview of Concepts and Evaluation Methods for
Computer Gaming QoE,” in International Workshop on Quality of Multimedia Experience, QoMEX, 2013, pp. 1–6.

14. User impact
-
More experienced players demand more!
M. Suznjevic, L. Skorin-Kapov, and M. Matijasevic, “The Impact of User, System, and Context factors on Gaming QoE a Case
Study Involving MMORPGs”, The 12th Annual Workshop on Network and Systems Support for Games, 2013., p. 6

15. Context impact
-
Player’s surrounding s (other players) have an
impact on QoE scores!
M. Suznjevic, L. Skorin-Kapov, and M. Matijasevic, “The Impact of User, System, and Context factors on Gaming QoE a Case
Study Involving MMORPGs”, The 12th Annual Workshop on Network and Systems Support for Games, 2013., p. 6

16. System impact - genres
Difference between game genres
250
Tollerated one way delay [ms]
-
200
150
100
50
0
FPS
RPG
Game genre
RTS

17. Network side
- Major parameters: Delay, jitter, packet
loss
- Fixed networks
-
Stable
Very low packet loss and jitter
Delay is still an issue
- Mobile networks
-
Huge growing market
Worse network conditions
Variation of network parameters

18. How do we fight network delay?
- Client side prediction
- QoS-enabled home devices (traffic
prioritization)
- Traffic shaping
- Dedicated networks
- Geographical server distribution
- Optimization of server locations
- Lag compensation mechanisms
(moving server time into the past)

19. Index
I.
Real-time services on
the Internet
II. The two problems
III. Adapting the network

20. Adapting the network
- Different solutions in order to
enhance the network support
for online games
- Traffic optimization
- Traffic prioritization
- Latency reduction methods

21. Adapting the network
- Different solutions in order to
enhance the network support
for online games
- Traffic optimization
- Traffic prioritization
- Latency reduction methods

22. VoIP optimization
- Voice trunking between two
offices:

23. VoIP optimization
- Savings of 50% are achievable
.
.
.
Context
.
.
.
Context
IP Network
.
.
.
MUX+COMP
Real-Time
DEMUX+DECOMP
Tunneling-Compressing-Multiplexing
Real-Time
.
.
.

24. VoIP optimization
- RFC 4170 (2005) deploys this,
also compressing the header
payload
payload
ECRTP
...
ECRTP
RTP
UDP
IP
PPP Mux
PPP
L2TP
VoIP
IP
One IPv4/UDP/RTP VoIP packet with two samples of 10 bytes
η=20/60=33%
40 to 6-8 bytes compression
Five IPv4/UDP/RTP VoIP packets with two samples of 10 bytes
η=20/60=33%
One IPv4 TCMTF Packet multiplexing five two sample packets
η=100/161=62%
saving

25. Optimization
- A trade-off appears:
- Efficiency problem: we are
improving it, since we are saving
bandwidth and reducing pps
- Quality problem: We are adding
an additional delay

26. Game traffic optimization
- Can we do the same with games?
- Is there any scenario where a
number of game traffic flows share
the same path?

27. Game optimization scenarios
- Between the network provider
and the game company
Home network with a
number of users (e.g.
Internet Café, access
shared by neighbours)
Individual DSL
xDSL router
DSLAM
ISP
aggregation
network
Network of
the service
provider
xDSL router
ISP network
DSLAM 2
LTE access
network
Application
server
BRAS
Serving
Gateway
Evolved NodeB
eNB
Internet
ISP
aggregation
network
Internet Router
(ISP)
Internet Router
(Service provider)
MUX/DEMUX

28. Game optimization scenarios
- The end user (e.g. internet café)
deploys the multiplexing
Home network with a
number of users (e.g.
Internet Café, access
shared by neighbours)
xDSL router
With multiplexer
embedded
DSLAM
Individual DSL
xDSL router
Without multiplexer
ISP network
Transport
network
BRAS
Internet Router
Internet

29. IETF proposal: TCM-TF
Tunneling, Compressing and Multiplexing Traffic Flows
draft-saldana-tsvwg-tcmtf-05
draft-suznjevic-tsvwg-mtd-tcmtf-02
payload
payload
RTP
payload
payload
RTP
UDP
TCP
UDP
UDP
IP
IP
IP
IP
ECRTP
Compression layer
No compr. / ROHC / IPHC / ECRTP
PPPMux
Multiplexing layer
PPPMux / Other
L2TP
Tunneling layer
GRE / L2TP
MPLS
IP
a) TCRTP
Network Protocol
IP
b) TCMTF

30. One IPv4/TCP packet 1500 bytes
η=1460/1500=97%
FPS optimization
Efficiency improvement IPv4, IPv6
One IPv4/UDP/RTP packet of VoIP with two samples of 10 bytes
η=20/60=33%
One IPv4/UDP server-to-client packet of Counter Strike with 9 players
One IPv6/TCP packet 1500 bytes
η=160/188=85%
η=1440/1500=96%
Four IPv4/UDP client-to-server packets of Counter Strike
η=61/89=68%
One IPv4/TCM packet multiplexing four client-to-server Counter Strike packets
saving
η=244/293=83%
One IPv6/UDP/RTP packet of VoIP with two samples of 10 bytes
η=20/80=25%
One IPv6/UDP server-to-client packet of Counter Strike with 9 players
η=160/208=77%
Three IPv6/UDP client-to-server packets of Counter Strike
η=61/109=56%
One IPv6/TCM packet multiplexing three client-to-server Counter Strike packets
η=183/246=74%
saving

31. FPS optimization
Bandwidth saving IPv4
IPv4 10 ms 5 players
IPv4 10 ms 20 players
60%
IPv4 reached
IPv4 theoretical
50%
40%
30%
20%
10%
0%
Quake 2
Unreal
Tournament
Counter Strike
1
Quake 3
Enemy
Territory
Counter Strike
2
Halo 2
Quake 4

32. MMORPG optimization
Can this be adapted to MMORPGs?
- Higher savings, due to the use of TCP (with
a lot of ACKs)
TCP ACKs without payload
Seven IPv4/TCP client-to-server packets of World of Warcraft. E[P]=20bytes
η=80/360=22%
One IPv4/TCMTF packet multiplexing seven client-to-server WoW packets
η=80/175=45%
saving

33. MMORPG optimization
TCMTF Bandwidth Saving, TCP/IPv4, World of Warcraft
tm
50%
45%
40%
35%
BS
30%
25%
20%
100 players
15%
50 players
10%
20 players
5%
10 players
0%
10
20
30
40
50
60
period (ms)
70
80
90
100

34. THANK YOU!
jsaldana@unizar.es
Mirko.Suznjevic@fer.hr

35. Extra slides

36. The approach
The network is the part that I do
not control
- Big amount of data in my
hard disk (size of WoW folder)
- Small packets (interactivity
requires frequent updates)

37. The two problems
- Efficiency problem
- If I use IPv6 (the next version of
the protocol, with a huge number
of different addresses), the
efficiency only drops to 96%
One IPv6/TCP packet 1500 bytes
η=1440/1500=96%
IPv6
TCP

38. The two problems
- Efficiency problem
- But for real-time services it
becomes even worse: only one
byte out of four is useful
information!
One IPv6/UDP/RTP packet of VoIP with two samples of 10 bytes
η=20/80=25%
IPv6
UDP
RTP
codec G.729a, 2 samples

39. System impact - protocols
- TCP is sometimes used for games
(usually MMORPGs)
-
Delayed acknowledgments
Nagle algorithm
A lot of “pure” ACK packets
Protocol
MMORPGs
TCP
World of Warcraft, Lineage I/II, Guild Wars,
Ragnarok Online, Anarchy Online, Mabinogi
UDP
EverQuest, City of Heroes, Star Wars Galaxies,
Ultima Online, Asherons Call, Final Fantasy XI
TCP/UDP
Dark Age of Camelot

40. VoIP optimization
- We merge the packets and make
them share the header:
1 flow
2 flows
3 flows

41. Game optimization scenarios
- The game company places a
multiplexer in the ISP network
Home network with a
number of users (e.g.
Internet Café, access
shared by neighbours)
DSLAM
Individual DSL
Network of
the service
provider
ISP
aggregation
network
xDSL router
ISP network
Application 1
server
Internet
xDSL router
BRAS
Internet Router
(ISP)
LTE access
network
Serving
ISP Gateway
Evolved NodeB aggregation
network MUX/DEMUX of the Service
eNB
Provider, hosted by the ISP
Application 2
server
Internet Router
(Service provider)
MUX/DEMUX

42. VoIP optimization
- RFC 4170 savings (50%):
TCMTF Bandwidth Saving, RTP/UDP/IPv4 voice G.729a, 2 samples per packet
60%
50%
40%
30%
20%
10%
0%
20 19
18 17
16 15
14 13
12 11
10
Number of calls
70%
9
8
7
6
5
85%
prob. of reduced header
4
3
2
100%
1

43. FPS optimization
Significant savings (Counter Strike)
Bandwidth Saving
35%
30%
25%
BS
20%
15%
20 players
10%
15 players
10 players
5%
5 players
0%
5
10
15
20
25
30
period (ms)
35
40
45
50

44. Shooting around the corner
- The same problems we can find
in VoIP.
- Instead of having two users
talking at the same time, we
have the “shooting around the
corner” problem.

45. VoIP
Jack
Wang
Question
Are you there?
time
Network delay scheme
Answer

46. Online games
“Shooting around the corner”
Jack
Wang
Wang

47. Online games
“Shooting around the corner”
Wang: Dead
Jack
Wang
Wang

48. Online games
“Shooting around the corner”
Wang: Dead
Jack
¡¡¡ #%$& !!!
Wang

49. Online games
“Shooting around the corner”
Jack
Game server
Wang
Position 1
Position 2
time
Shot
Wang DEAD
Network delay scheme

50. Game traffic optimization
A period is defined, and all the packets
arrived are compressed and
multiplexed
Native
traffic
...
Multiplexed
traffic . . .
PE
PE
PE
PE
...
...

51. Multiplexing
Bandwidth saving IPv6
IPv6 10 ms 5 players
IPv6 10 ms 20 players
60%
IPv6 reached
IPv6 theoretical
50%
40%
30%
20%
10%
0%
Quake 2
Unreal
Tournament
Counter Strike
1
Quake 3
Enemy
Territory
Counter Strike
2
Halo 2
Quake 4