SLIDE 1
Institute of Computer Science Chair of Communication Networks
- Prof. Dr.-Ing. P. Tran-Gia
%
Florian Wamser 2 2
Improving User QoE and Network Usage of YouTube in Mobile Broadband - - PowerPoint PPT Presentation
Improving User QoE and Network Usage of YouTube in Mobile Broadband - - PowerPoint PPT Presentation
Institute of Computer Science Chair of Communication Networks Prof. Dr.-Ing. P. Tran-Gia Dynamic Bandwidth Allocation for Multiple Network Connections: Improving User QoE and Network Usage of YouTube in Mobile Broadband Florian Wamser, Thomas
SLIDE 2
SLIDE 3
%
Florian Wamser 3 3
20 40 60 100 200 300 400 time [s] d a t a [ k B / s ]
10 20 30 40 50 60 5 10 15 time [s] buffered playtime [s]
Actual progress
Impact on Application Quality
bulk data (downloads) YouTube
Throughput
Desired progress Actual progress progress
Influence on YouTube
- Content unaware networks
- Fair share with respect to QoS (throughput)
- Bulk data download performance: good
- YouTube quality: bad
SLIDE 4
%
Florian Wamser 4 4
Application-Aware Networking
Tasks and objectives
- Integrating application needs’ in network resource management
- Add or re-allocate resources on demand
1.
Application and network monitoring
- Collects information with high correlation to QoE
- Example: YouTube monitor (YoMo), browsing monitor, etc.
2.
Decision entity
- Evaluates the information and decides about appropriate
resource management action
3.
Dynamic resource management
- Enforces resource management actions
- Example:
resource allocation, scheduling, traffic prioritization, access technology selection, … Dynamic Resource Management Application and Network Monitoring Decision Entity
SLIDE 5
%
Florian Wamser 5 5
Resource Management: Dual Connectivity of Devices
More than just one transmission
technology is available at current mobile devices
Wi-Fi Communications Cellular Communications
SLIDE 6
%
Florian Wamser 6 6
Framework for Intelligent Bandwidth Aggregation
Virtual access network (VAN) to aggregate
multiple networks into single IP pipe
Technical implementation: TCP/IP over UDP
tunneling (mobile IP-like approach)
Features of Intel‘s OTT VAN
Configurable bandwidth aggregation for multiple networks (TCP) packet reordering (re-sequencing)
Missing features
Smart algorithms for dynamic offloading Application specific guidelines
SLIDE 7
%
Florian Wamser 7 7
Intel’s OTT VAN Testbed: Hardware and Software
Core Network
Internet
Wi-Fi Access Network Mobile Access Network
Application TCP/UDP IP UDP UDP IP (Wi-Fi) IP (3G) Application TCP/UDP IP Server
Tunnel endpoint Implements re-sequencing
buffer
Enforces resource
management Client
Virtual network device provides tunneling
functionality
Access technologies
- Wi-Fi communications (limited to max. 2 Mbps)
- Mobile communications (limited to 4 Mbps)
SLIDE 8
%
Florian Wamser 8 8
Resource Management Algorithms
Adjust offload ratio between Wi-Fi and 3G cellular traffic, based on a
required throughput
Always use Wi-Fi and dynamically add 3G
If current throughput < required throughput
Increase 3G bandwidth
If current throughput > required throughput
Decrease 3G bandwidth
SLIDE 9
%
Florian Wamser 9 9
Resource Management Algorithms for
Algorithm 1: Static Offloading Based on Video
Request
- Defines required throughput based on requested
video quality
- Detects uplink request by YouTube with DPI
Algorithm 2: Dynamic Offloading Based on
Buffer Estimation
- Constant monitoring of the buffer level
- Adaption of the required throughput based on the
buffer level
Algorithm 3: Burst-wise Offloading Based on
Buffer Estimation
- Make use of the complete bandwidth until the
buffer is filled
- Disables 3G link until the buffer gets low
Algorithm 1 Time
- Req. throughput
Algorithm 2
- Req. throughput
Time Time
- req. throughput = max.
- req. throughput = 0 WiFi-only
Algorithm 3
SLIDE 10
%
Florian Wamser 10 10
Time Series of Algorithm 1
Time series of one video with 1080p resolution
Wi-Fi and 3G available
10 20 30 40 50 Buffered time [s] 50 100 150 200 250 300 Time [s] Request detected 5 Throughput [Mbps] Request detected Adaption to required throughput 6 4 3 2 1 50 100 150 200 250 300 Time [s] No WiFi or 3G activity
Buffer filled
Static req. throughput = 6 Mbps
SLIDE 11
%
Florian Wamser 11 11
Time Series of Algorithm 2
10 20 30 40 50 Buffered time [s] 50 100 150 200 250 300 Time [s] 5 Throughput [Mbps] 6 4 3 2 1 50 100 150 200 250 300 Time [s]
Algorithm 2 dynamically adjusts required throughput according to
video playback buffer
Request detected Request detected
Buffer filled Low buffer Throughput decreases
SLIDE 12
%
Florian Wamser 12 12
Comparison of Algorithm 1 and 2
Consumption of 3G bandwidth
per 1000 seconds playtime
Total 3G bandwidth consumed in MB 100 200 300 400 2.5 3 3.5 4 4.5 5 Average bit-rate in Mbps
Resolution based Buffer based 720p 98 MB 35 MB 1080p 321 MB 223 MB
Algorithm 1 (resolution based) Algorithm 2 (buffer based)
1.6 1.8 2 2.2 2.4 2.6 100 200 300 400 Average bit-rate in Mbps Total 3G bandwidth consumed in MB
- 64,3%
- 30,5%
720p 1080p
SLIDE 13
%
Florian Wamser 13 13
Comparison of Algorithm 1 and 2
Average amount of consumed energy
per 1000 seconds playtime
Resolution based Buffer based
Resolution based Buffer based 720p 1180 J 591 J 1080p 1404 J 1299 J
2.5 3 3.5 4 4.5 5 200 400 600 800 1000 1200 1400 1600 Average bit-rate in Mbps Energy consumption per video in Joule 1.6 1.8 2 2.2 2.4 2.6 200 400 600 800 1000 1200 1400 Average bit-rate in Mbps Energy consumption per video in Joule
- 49,9%
- 7,5%
720p 1080p
- J. Huang et al. „A close examination of performance
and power characteristics of 4g lte networks”
SLIDE 14
%
Florian Wamser 14 14
Time Series of Algorithm 3
10 20 30 40 50 Buffered time [s] 50 100 150 200 250 300 Time [s] 5 Throughput [Mbps] 6 4 3 2 1 50 100 150 200 250 300 Time [s]
Full Buffer Full Buffer Low Buffer Request detected
Algorithm 3 activates 3G link in bursts
Request detected 3G link disabled Low Buffer
SLIDE 15
%
Florian Wamser 15 15
Comparison of Algorithm 2 and 3
Average amount of consumed energy
per 1000 seconds playtime Dynamic buffer Burst-wise buffer 720p 591 J 519 J 1080p 1299 J 1183 J
Dynamic buffer Burst-wise buffer
2.5 3 3.5 4 4.5 5 200 400 600 800 1000 1200 1400 1600 Average bit-rate in Mbps Energy consumption per video in Joule 1.6 1.8 2 2.2 2.4 2.6 200 400 600 800 1000 1200 1400 Average bit-rate in Mbps Energy consumption per video in Joule
- 12,2%
- 8,9%
720p 1080p
SLIDE 16
%
Florian Wamser 16 16
Conclusion and Outlook
Contribution of the work
Assessment and quantification of the benefits of cross-layer resource management on the example of YouTube Analysis of three application-aware algorithms which differ in complexity and impact on user and network
Results of the evaluation
The application-aware algorithms can
– enhance the QoE level for end users (if both networks provide enough resources) – save costs in terms of energy & Cellular resources
Future work
Investigations on scalability of our approach and field trials with many users Providing a holistic resource allocation for popular applications with respect to their instaneneous needs
SLIDE 17
%
Florian Wamser 17 17
http://dl.acm.org/authorize?N71341 Florian Wamser, Thomas Zinner, Phuoc Tran-Gia and Jing Zhu Dynamic Bandwidth Allocation for Multiple Network Connections: Improving User QoE and Network Usage of YouTube in Mobile Broadband
SLIDE 18
%
Florian Wamser 18 18
http://dl.acm.org/authorize?N71209 Florian Wamser, Thomas Zinner, Lukas Iffländer, Phuoc Tran-Gia Demonstrating the Prospects of Dynamic Application-Aware Networking in a Home Environment
SLIDE 19
%
Florian Wamser 19 19
http://dl.acm.org/authorize?N71235 Steffen Gebert, David Hock, Thomas Zinner, Phuoc Tran-Gia (University
- f Würzburg); Marco Hoffmann, Michael Jarschel, Ernst-Dieter Schmidt