Performance Related Changes and their User Impact

Slides:



Advertisements
Similar presentations
Beliefs & Biases in Web Search
Advertisements

Accurately Interpreting Clickthrough Data as Implicit Feedback Joachims, Granka, Pan, Hembrooke, Gay Paper Presentation: Vinay Goel 10/27/05.
A Provider-side View of Web Search Response Time
ATM Firewall Routers with Black Lists Hwajung LEE The George Washington University School of Engineering and Applied Science Electrical Engineering and.
WEB DESIGN TABLES, PAGE LAYOUT AND FORMS. Page Layout Page Layout is an important part of web design Why do you think your page layout is important?
Junchen Jiang (CMU) Vyas Sekar (Stony Brook U)
RED-PD: RED with Preferential Dropping Ratul Mahajan Sally Floyd David Wetherall.
©2013 AKAMAI | FASTER FORWARD TM It's all about Performance Measured and Perceived Performance on Desktop and Mobile Devices San Mateo Meetup, July 2013.
Chapter 12 Working with Forms Principles of Web Design, 4 th Edition.
University of Michigan Electrical Engineering and Computer Science Anatomizing Application Performance Differences on Smartphones Junxian Huang, Qiang.
1 Project Planning with MS Project Kishore October 2002 Notes available in the presentation: To view notes download the presentation and select View->
By Jory Cohen Made for CISC856, Spring 2010 Thanks to Dr. Amer, Mike Belshe(Google), Jon Leighton.
Dynamic Adaptive Streaming over HTTP2.0. What’s in store ▪ All about – MPEG DASH, pipelining, persistent connections and caching ▪ Google SPDY - Past,
Ao-Jan Su † Y. Charlie Hu ‡ Aleksandar Kuzmanovic † Cheng-Kok Koh ‡ † Northwestern University ‡ Purdue University How to Improve Your Google Ranking: Myths.
Broadband Internet Performance: A View from the Gateway Srikanth Sundaresan, Walter de Donato, Nick Feamster, Renata Teixeira, Sam Crawford, Antonio Pescapè.
Performance Analysis of Orb Rabin Karki and Thangam V. Seenivasan 1.
Multi-level Application-based Traffic Characterization in a Large-scale Wireless Network Maria Papadopouli 1,2 Joint Research with Thomas Karagianis 3.
Effective Agent Participation in Rescue Operations Brent Dutson CS7100 Fall 2009.
Exploiting Content Localities for Efficient Search in P2P Systems Lei Guo 1 Song Jiang 2 Li Xiao 3 and Xiaodong Zhang 1 1 College of William and Mary,
Dynamics of Hot-Potato Routing in IP Networks Renata Teixeira (UC San Diego) with Aman Shaikh (AT&T), Tim Griffin(Intel),
A Distributed Proxy Server for Wireless Mobile Web Service Kisup Kim, Hyukjoon Lee, and Kwangsue Chung Information Network 2001, 15 th Conference.
Microsoft Research Shujaat Hussain. Cloud Faster! Low latency web transactions …. especially important to our key online properties.
Performance Comparison of Congested HTTP/2 Links Brian Card, CS /7/
Inbound Statistics Slides Attract. 1 Blogging There are 31% more bloggers today than there were three years ago 46% of people read blogs more than once.
Ch. 28 Q and A IS 333 Spring Q1 Q: What is network latency? 1.Changes in delay and duration of the changes 2.time required to transfer data across.
Niranjan Balasubramanian Aruna Balasubramanian Arun Venkataramani University of Massachusetts Amherst Energy Consumption in Mobile Phones: A Measurement.
KaZaA: Behind the Scenes Shreeram Sahasrabudhe Lehigh University
HAPORI: CONTEXT-BASED LOCAL SEARCH FOR MOBILE PHONES USING COMMUNITY BEHAVIORAL MODELING AND SIMILARITY Presented By: Brandon Ochs Nicholas D. Lane, Dimitrios.
DNN Performance & Scalability Planning, Evaluating & Improving : Part 2.
Intro to Computer Networks Internet and Networking Terms Bob Bradley The University of Tennessee at Martin.
Niels Provos and Panayiotis Mavrommatis Google Google Inc. Moheeb Abu Rajab and Fabian Monrose Johns Hopkins University 17 th USENIX Security Symposium.
1 All Your iFRAMEs Point to Us Mike Burry. 2 Drive-by downloads Malicious code (typically Javascript) Downloaded without user interaction (automatic),
Authors: Maryam Kamvar and Shumeet Baluja Date of Publication: August 2007 Name of Speaker: Venkatasomeswara Pawan Addanki.
Global NetWatch Copyright © 2003 Global NetWatch, Inc. Factors Affecting Web Performance Getting Maximum Performance Out Of Your Web Server.
+ Network Level Footprints of Facebook Applications Komal Pal Gautam Bhawsar.
Web Pages with Features. Features on Web Pages Interactive Pages –Shows current date, get server’s IP, interactive quizzes Processing Forms –Serach a.
Energy Consumption in Mobile Phones: A Measurement Study and Implications for Network Applications REF:Balasubramanian, Niranjan, Aruna Balasubramanian,
Web Performance and key business metrics Part II: More Findings from the Front Line of Web Acceleration.
Streaming Media A technique for transferring data on the Internet so it can be processed as a steady and continuous stream.
McLean HIGHER COMPUTER NETWORKING Lesson 7 Search engines Description of search engine methods.
TCP behavior of a Busy Internet Server: Analysis and Improvements Y2K Oct.10 Joo Young Hwang Computer Engineering Research Laboratory KAIST. EECS.
Zibin Zheng DR 2 : Dynamic Request Routing for Tolerating Latency Variability in Cloud Applications CLOUD 2013 Jieming Zhu, Zibin.
Qi Guo Emory University Ryen White, Susan Dumais, Jue Wang, Blake Anderson Microsoft Presented by Tetsuya Sakai, Microsoft Research.
Efficient P2P Search by Exploiting Localities in Peer Community and Individual Peers A DISC’04 paper Lei Guo 1 Song Jiang 2 Li Xiao 3 and Xiaodong Zhang.
PERFORMANCE ENHANCEMENT IN ASP.NET By Hassan Tariq Session #1.
Measuring and Mitigating Web Performance Bottlenecks in Broadband Access Networks Srikanth Sundaresan, Nick Feamster (Georgia Tech) Renata Teixeira (Inria)
TCP: Transmission Control Protocol Part II : Protocol Mechanisms Computer Network System Sirak Kaewjamnong Semester 1st, 2004.
OPTIMISING USER INTERFACES FOR MOBILE COMPUTER ASSISTED INTERVIEWING DEVICES Prepared By Otyek Ronald James Uganda Bureau of Statistics Presented at the.
Case Study: Live Search – Moving to and then away from AJAX Eric Schurman Performance Development Lead Live Search Microsoft.
Performance Testing Test Complete. Performance testing and its sub categories Performance testing is performed, to determine how fast some aspect of a.
Experiences of Cloud Storage Service Monitoring Performance Assessment and Comparison  Enrico Bocchi  Idilio Drago  Marco Mellia Cloud Services for.
How Web Database Architectures Work CPS181s April 8, 2003.
Identifying “Best Bet” Web Search Results by Mining Past User Behavior Author: Eugene Agichtein, Zijian Zheng (Microsoft Research) Source: KDD2006 Reporter:
Computer Network Architecture Lecture 6: OSI Model Layers Examples 1 20/12/2012.
Web Proxy Caching: The Devil is in the Details Ramon Caceres, Fred Douglis, Anja Feldmann Young-Ho Suh Network Computing Lab. KAIST Proceedings of the.
WHAT'S THE DIFFERENCE BETWEEN A WEB APPLICATION STREAMING NETWORK AND A CDN? INSTART LOGIC.
Web Design Vocabulary #3. HTML Hypertext Markup Language - The coding scheme used to format text for use on the World Wide Web.
Week-7 (Lecture-1) Internet Protocols: TCP/IP, Client Server: Protocol: Specification in networking and Communications It defines the procedures to be.
© 2006 Cisco Systems, Inc. All rights reserved.Cisco Public 1 OSI transport layer CCNA Exploration Semester 1 – Chapter 4.
25/09/2016 INASP: Effective Network Management Workshops Unit 6: Solving Network Problems.
Reach People when it matters with Location Extensions
IsoveraDL Performance Enhancements
IST 497 Vladimir Belyavskiy 11/21/02
Do it now – PAGE 11 You will find your do it now task in your workbook – look for the start button! Wednesday, 21 November 2018.
08/03/14 Energy Consumption in Mobile Phones: A Measurement Study and Implications for Network Applications REF:Balasubramanian, Niranjan, Aruna Balasubramanian,
The Internet An Overview.
CSE 461 HTTP and the Web.
Network Controllable MP3 Player
Identifying Slow HTTP DoS/DDoS Attacks against Web Servers DEPARTMENT ANDDepartment of Computer Science & Information SPECIALIZATIONTechnology, University.
Yale Digital Conference 2019
Presentation transcript:

Performance Related Changes and their User Impact Eric Schurman Principal Development Lead Bing Jake Brutlag Decision Support Engineering Analyst Google

Experiments Server Delays (Microsoft and Google) Page Weight Increases Progressive Rendering

Server-side Delays Experiment Goal Determine impact of server delays Methodology Delay before sending results Different experiments with different delays Small number of users Monitor negative impact

Server Delays Experiment: Results - Means no statistically significant change Strong negative impacts Roughly linear changes with increasing delay Time to Click changed by roughly double the delay

Google Web Search Delay Experiments A series of experiments on a small % search traffic to measure the impact of latency on user behavior Randomly assign users to the experiment and control groups (A/B testing) Server-side delay: Emulates additional server processing time May be partially masked by network connection Varied type of delay, magnitude (in ms), and duration (number of weeks)

Search Traffic Impact Type of Delay Delay (ms) Experiment Duration (weeks) Impact on Average Daily Searches Per User Pre-header 50 4 Not measurable 100 -0.20% Post-header 200 6 -0.29% 400 -0.59% Post-ads -0.30% Increase in abandonment heuristic = less satisfaction Abandonment heuristic measures if a user stops interacting with search engine before they find what they are looking for Active users (users that search more often a priori) are more sensitive

-0.22% -0.36% -0.44% -0.74%

-0.08% -0.21%

Page Weight Experiment <html> <head> <CSS /> <!--Payload--> </head> <body> <answer>... </answer> <results> <result>... </result> </results> ... <script /> </body> </html> Goal Determine impact of a heavier page. Isolate bytes over the wire cost, not layout costs, etc. Methodology Use incompressible HTML comments Vary size (from 1.05x to 5x page size) and location of payload Experiment with payload in individual and multiple locations US-only test – mostly good broadband

Page Weight Experiment: Results Minimal impact for small payloads Payload at top of page had stronger effect Performance suffered slightly – would have been worse if tested in regions with poor connectivity Click metrics impacted more than Query metrics Largest experiment (approx 5X control page size) Any Clicks: -0.55% No changes to query metrics Results only apply to one GET – not multiple

Progressive Rendering Experiment Visual Header - Fast to compute Results - Slower to compute Goal Determine impact sending visual header before results. Methodology Build page in phases Send using HTTP 1.1 Chunked Transfer Encoding Application design impacts

Progressive Rendering Experiment: Results Metric Change Performance Faster across all latency percentiles 4-18% faster to download all HTML Roughly halved time to see visible page change Time to Click ~9% faster Query refinement +2.2% Clicks overall +0.7% Pagination +2.3% Satisfaction

Conclusion "Speed matters" is not just lip service Delays under half a second impact business metrics The cost of delay increases over time and persists Number of bytes in response is less important than what they are and when they are sent Use progressive rendering