Thin to Win? Network Performance Analysis of the OnLive Thin Client Game System By Mark Claypool, David Finkel, Alexander Grant, and Michael Solano Submitted.

Slides:



Advertisements
Similar presentations
A Graduate Course on Multimedia Technology 3. Multimedia Communication © Wolfgang Effelsberg Media Scaling and Media Filtering Definition of.
Advertisements

CS4513 Distributed Computing Systems Games in the Cloud.
Performance Analysis of Home Streaming Video Using Orb Rabin Karki, Thangam Seenivasan, Mark Claypool and Robert Kinicki Worcester Polytechnic Institute.
Playback-buffer Equalization For Streaming Media Using Stateless Transport Prioritization By Wai-tian Tan, Weidong Cui and John G. Apostolopoulos Presented.
1 Measurements of the Congestion Responsiveness of Windows Streaming Media James Nichols, Mark Claypool, Robert Kinicki and Mingzhe Li Computer Science.
Measurements of Congestion Responsiveness of Windows Streaming Media (WSM) Presented By:- Ashish Gupta.
Networks & Multimedia Amit Pande, Post-doctoral fellow, Department of Computer Science, University of California Davis
12/10/2006ConfidentialSlide 1 Video Streaming over UMTS: practical issues Stefan Rugel, Klaus Schäfer February 2008.
Application, Network and Link Layer Measurements of Streaming Video over a Wireless Campus Network Passive & Active Measurement Workshop 05 Boston, MA,
Motion and Scene Complexity for Streaming Video Games Mark Claypool Computer Science Department Worcester Polytechnic Institute Worcester, Massachusetts,
Mark Claypool’s MQP Projects Network Games Streaming Media.
Performance Analysis of Orb Rabin Karki and Thangam V. Seenivasan 1.
Perspectives, Frame Rates and Resolutions: It's all in the Game Mark Claypool Kajal Claypool Worcester Polytechnic Institute Worcester, MA, USA
Mark Claypool’s MQP Projects Network Games Streaming Media.
April 2004MQP Topic PresentationsSlide 1 Mark Claypool’s Projects Network Games Streaming Video.
An Empirical Study of RealVideo Performance Across the Internet Yubing Wang, Mark Claypool and Zheng Zuo
Characterization of Wireless Networks in the Home Mark Yarvis, Konstantina Papagiannaki, and W. Steven Conner Presented by Artur Janc, Eric Stein.
The Effects of Loss and Latency on User Performance in Unreal Tournament 2003 Tom Beigbeder, Rory Coughlan, Corey Lusher, John Plunkett, Emmanuel Agu,
The Effects of Loss and Latency on User Performance in Unreal Tournament 2003 Tom Beigbeder, Rory Coughlan, Corey Lusher, John Plunkett, Emmanuel Agu,
The Effects of Loss and Latency on User Performance in Unreal Tournament 2003 Tom Beigbeder, Rory Coughlan, Corey Lusher, John Plunkett, Emmanuel Agu,
Characterization by Measurement of a CDMA 1x EVDO Network Presenter: Mingzhe Li Wireless Internet Conference (WICON’06) Boston, Massachusetts,
The 4th IEEE International Conference on Broadband Communications, Networks and Systems (BROADNETS) Raleigh, NC, USA September 10-13, 2007 Measuring Queue.
Packet Loss and Latency in Unreal Tournament 2003 Tom Beigbeder Rory Coughlan Corey Lusher John Plunkett.
School of Computer Science and Software Engineering A Networked Virtual Environment Communications Model using Priority Updating Monash University Yang-Wai.
Traffic Analysis of Avatars in Second Life James Kinicki and Mark Claypool Computer Science Department Worcester Polytechnic Institute Worcester, Massachusetts,
March 2008MQP TopicsSlide 1 Mark Claypool’s MQP Projects Network Games Streaming Media.
On the Turbulence * of Nintendo DS and Sony PSP Hand-held Network Games Mark Claypool Computer Science Department Worcester Polytechnic Institute.
Analysis of compressed depth and image streaming on unreliable networks Pietro Zanuttigh, Andrea Zanella, Guido M. Cortelazzo.
Measurement of the Congestion Responsiveness of RealPlayer Streaming Video Over UDP Jae Chung, Mark Claypool, Yali Zhu Proceedings of the International.
Using Redundancy and Interleaving to Ameliorate the Effects of Packet Loss in a Video Stream Yali Zhu, Mark Claypool and Yanlin Liu Department of Computer.
The Effects of Frame Rate and Resolution on Users Playing First Person Shooter Games Mark Claypool CS Department Worcester Polytechnic Institute
The Effects of Latency on User Performance in Warcraft III Nathan Sheldon, Eric Gerard, Seth Borg, Mark Claypool, Emmanuel Agu Computer Science Department.
Network Analysis of Counter-strike and Starcraft Mark Claypool, David LaPoint, Josh Winslow Worcester Polytechnic Institute Worcester, MA, USA
The Effects of Latency on Player Performance in Cloud-based Games Mark Claypool and David Finkel Computer Science and Interactive.
Mark Claypool’s MQP Projects Network Games Streaming Media.
Ryan Shea and Jiangchuan Liu, Simon Fraser University Edith C.-H. Ngai, Uppsala University, Yong Cui, Tsinghua University Published in August 2013 Presented.
CS 529 Multimedia Networking Review. Digital Audio What is the tradeoff between sample size and fidelity? Between sample rate and fidelity? What uses.
1 Assessing The Real Impact of WLANs: A Large-Scale Comparison of Wired and Wireless Traffic Maria Papadopouli * Assistant Professor Department.
Killing Zombies with Rate Controlled Adaptive Intra Refresh over Wireless HDMI by Nicholas Jamba.
Jani Pousi Supervisor: Jukka Manner Espoo,
Profiling Grid Data Transfer Protocols and Servers George Kola, Tevfik Kosar and Miron Livny University of Wisconsin-Madison USA.
Exploiting Proxy-Based Transcoding to Increase the User Quality of Experience in Networked Applications Maarten Wijnants Patrick Monsieurs Peter Quax Wim.
Roger Allen, Principal Architect, GRID Architecture NVIDIA GRID & GAMESTREAM.
Warcraft III, by Blizzard CSE 381 – Advanced Game Programming Networked Gaming Issues.
Mark Claypool’s MQP Projects Network Games Streaming Media.
The Way Forward Factors Driving Video Conferencing Dr. Jan Linden, VP of Engineering Global IP Solutions.
ONLINE GAME NETWORK TRAFFIC OPTIMIZATION Jaewoo kim Youngho yi Minsik cho.
TCP and SCTP RTO Restart draft-hurtig-tcpm-rtorestart-02 Michael Welzl 1.
Assignment of Games to Servers in the OnLive Cloud Game System David Finkel, Mark Claypool, Sam Jaffe, Thinh Nguyen, Brendan Stephen Computer Science and.
Thin to Win? Network Performance Analysis of the OnLive Thin Client Game System Mark Claypool, David Finkel, Alexander Grant and Michael Solano Computer.
1 Measuring Congestion Responsiveness of Windows Streaming Media James Nichols Advisors: Prof. Mark Claypool Prof. Bob Kinicki Reader: Prof. David Finkel.
Internet Measurment Multimedia 1. Properties Challenges Tools State of the Art 2.
Paper Review: Latency Evaluation of Networking Mechanisms for Game Traffic Jin, Da-Jhong.
Improving application layer latency for reliable thin-stream By: Joel Fichter & Andrew Sitosky Src:
Latency and Player Actions in Online Games Mark Claypool & Kajal Claypool Worcester Polytechnic Institute Communications of the ACM, Nov Presented.
Measurement-based Analysis of the Video Characteristics of Twitch.tv Mark Claypool, Daniel Farrington, and Nicholas Muesch Computer.
Adaptive Content-Aware Scaling for Improved Video Streaming. Avanish Tripathi Advisor: Mark Claypool Reader: Bob Kinicki.
Quality of Service Schemes for IEEE Wireless LANs-An Evaluation 主講人 : 黃政偉.
COPYRIGHT © 2012 ALCATEL-LUCENT. ALL RIGHTS RESERVED. lightRadio TM Network Demonstration October 22, 2013 The LTE End User Experience.
Doc.: IEEE /1162r1 Submission Sept 2013 Guoqing Li (Intel)Slide 1 Video Application Categories and Characteristics Date: Authors: NameAffiliationsAddressPhone .
The Effects of Latency on Player Performance in Cloud-based Games Mark Claypool and David Finkel Worcester Polytechnic Institute 1 In Proceedings of the.
Accelerating Peer-to-Peer Networks for Video Streaming
Types of Operating System
Custom Configurations
A First Look at Performance of TV Streaming Sticks
Roaming Interval Measurements
Custom Configurations
Assignment of Games to Servers in the OnLive Cloud Game System
Hardware Accelerated Video Decoding in
DL MU MIMO Error Handling and Simulation Results
Presentation transcript:

Thin to Win? Network Performance Analysis of the OnLive Thin Client Game System By Mark Claypool, David Finkel, Alexander Grant, and Michael Solano Submitted March 2013 Presented by Anthony Botelho

Thin Clients All (or most) of the game computations occur on a server, and send back rendered frames The client needs only send small amounts of data, and stream back the resulting images (like a video)

Motivation Network load is important to plan infrastructures A variety of applications have been studied PC & console games Video streaming (both live and pre-recorded) No comparisons exist between thin clients and these other applications

Turbulence A way is needed to compare applications in similar terms Turbulence describes common network characteristics: Overall bitrate Packet size Frequency of packets Together can be used to build traffic models and classifiers

Overview Background/Motivation OnLive turbulence Comparison of game genres OnLive vs similar applications Reactions to network conditions Summary/Conclusions

Research Question 1 What is the network turbulence for OnLive games? Games are shown to have: High downstream bitrates: ~5Mb/s with 1000 byte packets Moderate upstream bitrates: ~100Kb/s with 150 byte packets

OnLive Thin Client MicroConsole runs games connected to HDTV over HDMI running at 1080p All OnLive traffic uses UDP for both up and down streams Want to first look at the overall turbulence of games using OnLive

First Experiment Three popular games with comparable graphics qualities chosen to represent possible differences in turbulence Unreal Tournament III (2007) Batman: Arkham Asylum (2009) Grand Ages: Rome (2009) Each has similar system requirements, differing slightly in minimum RAM and processor requirements

First Experiment Each game was played for 2.5 minutes in a chosen scenario that best represents that game’s gameplay Long enough to experience all core game interactions Unreal Tournament III Batman: Arkham Asylum Grand Ages: Rome

Overall Turbulence OnLive appears very asymmetric Approx. 50:1 bitrates, 6:1 inter-packet times, 10:1 packet sizes Onlive Up:Down Turbulance (mean values)

Overview Background/Motivation OnLive turbulence Comparison of game genres OnLive vs similar applications Reactions to network conditions Summary/Conclusions

Research Question 2 Does the network turbulence for different game genres differ from each other? The games observed show: Similar trends, but different rates Large variance in downstream and upstream traffic

Experiment 2 Each game’s bitrate, packet size, and frequency of packets is compared Both downstream and upstream are compared separately Any differences in the graphs would suggest differences in the genres represented by each game

OnLive Client Bitrates

OnLive Client Packet Size

OnLive Client Inter-Packet Times

Comparison There appear to be similar trends Unreal Tournament III shows higher bitrates and more frequent packets Illustrates certain genres (like a fast-paced FPS) produce a larger network load Aspects such as camera angle, and the pace of the game may effect downstream requirements

Overview Background/Motivation OnLive turbulence Comparison of game genres OnLive vs similar applications Reactions to network conditions Summary/Conclusions

Research Question 3 Does the network turbulence for OnLive games differ from traditional games? When compared: Upstream traffic appears similar to traditional games Downstream traffic more resembles live video

Experiment 3 Observe OnLive compared to similar applications as well Traditional Games (UTIII on PC) Virtual Environment (Second Life) Youtube (pre-recorded video) Skype (live video) Only Unreal Tournament III is compared

Comparison of Turbulance Medians of downstream turbulence

Comparison of Turbulance Medians of downstream turbulence OnLive behaves more like live video streaming than any other in terms of down stream Even OnLive upstream produces more turbulence than traditional games Suggests large network impact when using thin clients

Overview Background/Motivation OnLive turbulence Comparison of game genres OnLive vs similar applications Reactions to network conditions Summary/Conclusions

Research Question 4 Does the network turbulence for OnLive change with different network conditions? The OnLive client: Adapts bitrates to capacity limits, but not to loss or latency Adapts frame rate to capacity limits and loss, but not latency

Experiment 4 Different network conditions are simulated to view effects of the OnLive downstream Bitrate capacity restrictions 10 Mb/s and 5 Mb/s Packet Loss 1% and 1.5% Loss Latency 40ms and 70ms latency Only Unreal Tournament III is observed

Onlive Responses to Network Conditions Bitrate under capacity restrictions Bitrate with latency and packet loss

OnLive Frame Rates The impact on frame rate is also observed The OnLive client responds to capacity and loss, but does not respond to latency in any observed metrics

Overview Background/Motivation OnLive turbulence Comparison of game genres OnLive vs similar applications Reactions to network conditions Summary/Conclusions

Summary OnLive traffic is asymmetric, and illustrate similar trends in turbulence Genres do seem to display variance in actual values The use of a thin client reduces hardware requirements, but increases network turbulence Even when comparing the upstream to traditional games

Summary The client behaves more similarly to live video streaming in terms of downstream The OnLive client does not appear to respond to latency It does respond to capacity limitations in terms of lowering bitrate and frame rate It also responds to loss by lowering frame rate