Presentation is loading. Please wait.

Presentation is loading. Please wait.

Understanding Website Complexity: Measurements, Metrics, and Implications Michael Butkiewicz Harsha Madhyastha UC Riverside 1 Vyas Sekar Intel Labs.

Similar presentations


Presentation on theme: "Understanding Website Complexity: Measurements, Metrics, and Implications Michael Butkiewicz Harsha Madhyastha UC Riverside 1 Vyas Sekar Intel Labs."— Presentation transcript:

1 Understanding Website Complexity: Measurements, Metrics, and Implications Michael Butkiewicz Harsha Madhyastha UC Riverside 1 Vyas Sekar Intel Labs

2 2 doubleclick.net cnn.com facebook.com cdn.turner.com ads.cnn.com Websites today are very complex! Diverse content from many servers and third party services

3 67% of users encounter “slow” sites once a week (gomez.com) 30% > 3 Seconds Median = 2 Seconds Users see slow loading websites! 3

4 Why does load time matter? 4 Source: gomez.com Implications for: Website owners End users Browser developers Customization

5 Our work Comprehensive study of website complexity – Analysis of sites across rank and category – Content and Service level metrics Key metrics that impact performance 5

6 Roadmap Introduction Measurement Setup Complexity Performance Implications Discussion and Summary 6

7 Measurement Setup Selecting websites – 1,700 websites from Quantcast top 20k – Primary focus on landing (home) page – Annotated with Alexa Categories Tools – Firefox + Firebug – No Local Caching Approach – 4 vantage points (3 EC2, 1 UCR) – Every 60 second one page loaded – ~30 measurements per site per vantage point over 9 weeks 7

8 Example Site Download 8 CSS Image Script HTML Time A B C D Objects

9 "log":{ "browser":{ "name":"Firefox”, }, "pages":[{ "startedDateTime":"18:12: :00", "title":"Wired News”, "pageTimings":{ "onLoad":4630 }] "entries":[{ "startedDateTime":"18:12: :00", "time”:75, "request":{... "headers":[{ "name":"Host", "value":"www.wired.com" }, } “response":{ "content":{ "mimeType":"text/html", "size":186013,... Example Firebug Log 9

10 Roadmap Introduction Measurement Setup Complexity – Content-level – Service-level Performance Implications Discussion and future work 10

11 11 Median Site = 57 Objects! Number of Objects: Across Categories Median 125 Objects! Median 125 Objects!

12 12 20% > 100! Number of Objects: Across Ranks Not as much difference across rank ranges

13 Types of Content 13 Median site: 33 Images, 10 JavaScript, 3 CSS, 0 Flash Flash

14 14 Normalized types of content Type % Mostly Homogeneous; Flash Skewed

15 Roadmap Introduction Measurement Setup Complexity – Content-level – Service-level Performance Implications Discussion and Summary 15

16 Median site requires contacting 8 servers 16 Median News 30 Servers! Median News 30 Servers! Number of Servers

17 17 Median News 20 Origins! Median News 20 Origins! 20% Sites > 13 Origins Median site requires contacting 6 origins Number of Origins

18 Popular non-origin providers Name% of sites google-analytics 58 doubleclick 45 quantserve 30 scorecardresearch 27 2mdn 24 googleadservices 18 facebook 17 yieldmanager Most common services: Analytics & Advertising Most common objects: Image (small!) and Javascript bluekai.com imrworldwide.com invitemedia.com > 5% of sites each! Not just usual suspects!

19 Contribution of non-origin services 19 Median Site Objects / Bytes 30% 35% Median Site Objects / Bytes 30% 35% 20% sites > 80% from 3 rd Party 20% sites > 80% from 3 rd Party

20 Contribution of non-origin services 20 Time only 15% Median Site Objects / Bytes 30% 35% Median Site Objects / Bytes 30% 35% 80% from 3 rd Party

21 Roadmap Introduction Measurement Setup Complexity Performance Implications Discussion and Summary 21

22 Metric Review 22 Content-Level Characteristics Total Objects Object Type: Number, Size Absolute & Normalized Service-Level Characteristics Number: Server & Origins Non-Origin Fraction: Servers / Objects / Time Total combination of 33 metrics

23 23 Load Time vs. Metric Correlation

24 Load Time Variability vs. Metrics 24

25 Roadmap Introduction Measurement Setup Complexity – Content-level – Service-level Performance Implications Discussion and Summary 25

26 Discussion: Many other variables Client-side plugins – NoScript reduces #objects by half! Mobile-specific customizations – Mobile version reduces #objects to a quarter Landing vs. non-landing pages – Non-landing seem less complex 26

27 Conclusions Comprehensive study of Website Complexity Median site: 57 objects, contacts 8 servers across 6 origins Categories show more differences than popularity ranks Non-origin content: – Analytics & advertising popular – large # objects, bytes, servers, not time Key performance indicators: – Load Time  # objects – Variability  # servers Data: 27


Download ppt "Understanding Website Complexity: Measurements, Metrics, and Implications Michael Butkiewicz Harsha Madhyastha UC Riverside 1 Vyas Sekar Intel Labs."

Similar presentations


Ads by Google