high performance web sites 14 rules for faster-loading pages

High Performance Web Sites

14 rules for faster-loading pagesSteve [email protected] [email protected]

Introduction

Exceptional Performancestarted in 2004quantify and improve the performance of all Yahoo! products worldwidecenter of expertisebuild tools, analyze datagather, research, and evangelize best practices

Scopeperformance breaks into two categoriesresponse timeefficiencycurrent focus is response timeof web products

Rough Cuts: nowHardcopy: Summer 2007http://www.oreilly.com/catalog/9780596514211/

The Importance of Front-End Performance

Back-end vs. Front-end

percentage of time spent on the front-end

The Performance Golden Rule80-90% of the end-user response time is spent on the front-end. Start there.

Greater potential for improvementSimplerProven to work

SchedulePerformance Researchbreak14 RulesbreakCase StudiesLive Analysis

Performance Research

slow crawl boring snail stagnant unexceptional yawn unresponsive impatient delay moderate blah subdue drag apathetic prolong slack load sluggish sleepy late unexciting reduced lag complex heavy unmemorable obscure why wait perceived response timewhat is the end users experience?performance speed enjoyable urgent instant accelerate perception snap achievement better improve action pleasant pace quick promote swift cool maximum drive prompt advance fast hurry rush satisfying feel exceptional brisk rapid exciting

User PerceptionUsability and perception are important for performance.

The users perception is more relevant than actual unload-to-onload response time.

Definition of "user onload" is undefined or varies from one web page to the next.

http://yuiblog.com/blog/2006/11/28/performance-research-part-1/

80/20 Performance RuleVilfredo Pareto: 80% of consequences come from 20% of causes

Focus on the 20% that affects 80% of the end-user response time.

Start at the front-end.

Empty vs. Full Cache

Empty vs. Full Cachewith an empty cache

Chart5

2.3857080.073697

1.9012470.03514

1.9003720.250341

1.7553970.115827

1.7364090.017868

1.735610.023736

1.7343070.022563

1.5444550.075842

1.5013610.22307

1.4913270.235455

1.3787060.027494

1.4323860.071666

1.3459750.146844

1.3452710.14719

1.1955820.062929

1.1936250.034181

1.1481810.030886

1.1473480.03236

1.0941010.048729

1.0691020.071543

1.0648260.098428

0.9710670.162706

0.8875690.232298

0.886340.1925

0.8033120.169048

0.7952380.145024

0.6135730.172498

0.6061860.044826

0.6023090.030719

0.4876780.053908

0.4441490.018247

0.4068670.103763

0.402910.026868

0.0644530.288027

00.018238

Sheet1

"wd:dItemDuration">PT0.018238SPT0.288027SPT0.026868SPT0.103763SPT0.018247SPT0.053908SPT0.030719SPT0.044826SPT0.172498SPT0.145024SPT0.169048SPT0.192500SPT0.232298SPT0.162706SPT0.098428SPT0.071543SPT0.048729SPT0.032360SPT0.030886SPT0.034181SPT0.062929SPT0.147190SPT0.146844SPT0.071666SPT0.027494SPT0.235455SPT0.223070SPT0.075842SPT0.022563SPT0.023736SPT0.017868SPT0.115827SPT0.250341SPT0.035140SPT0.073697S

Empty vs. Full Cache

Empty vs. Full CacheExpires header3user re-requests www.yahoo.comwith a full cache

Chart8

0.80.091007

0.701980.071007

00.4

Sheet1

"wd:dItemDuration">PT0.018238SPT0.288027SPT0.026868SPT0.103763SPT0.018247SPT0.053908SPT0.030719SPT0.044826SPT0.172498SPT0.145024SPT0.169048SPT0.192500SPT0.232298SPT0.162706SPT0.098428SPT0.071543SPT0.048729SPT0.032360SPT0.030886SPT0.034181SPT0.062929SPT0.147190SPT0.146844SPT0.071666SPT0.027494SPT0.235455SPT0.223070SPT0.075842SPT0.022563SPT0.023736SPT0.017868SPT0.115827SPT0.250341SPT0.035140SPT0.073697S

Empty vs. Full Cache empty cache2.4 seconds

full cache0.9 seconds83% fewer bytes90% fewer HTTP requests

How much does this benefit our users?It depends on how many users have components in cache.

What percentage of users view a page with an empty cache*?* Empty cache means the browser has to request the components instead of pulling them from the browser disk cache.

What percentage of page views are done with an empty cache*?

Browser Cache ExperimentAdd a new image to your page

with the following response headers:

Expires: Thu, 15 Apr 2004 20:00:00 GMTLast-Modified: Wed, 28 Sep 2006 23:49:57 GMT

Browser Cache ExperimentRequests from the browser will have one of these response status codes:200 The browser does not have the image in its cache.304 The browser has the image in its cache, but needs to verify the last modified date.

Browser Cache Experiment

Surprising Results40-60% ~20%

Chart2

11

0.9756970610.3511414856

0.64892420720.2480818574

0.5362557680.2172068146

0.49705641450.205786832

0.47695751880.2014351966

0.49093842630.2013420285

0.47437852410.1930321497

0.46505295010.1954734765

0.45392525710.1938962941

0.44627662250.1903622437

0.44449787410.1888348329

0.4392663860.1877929461

0.45289059970.1891742737

0.4466963780.186161091

0.44525710320.1868084979

0.43449977570.1843509261

17

18

19

20

unique users with empty cache

page views with empty cache

day of experiment

percentage

Empty Cache %

DateDay%age 200 B cookies%age 200 page viewsC + filter old BE PVsE + filter new BG PVsF + extreme 200ersI PVs

1/22/060100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%

1/23/06197.6%35.1%97.6%34.9%91.9%33.0%95.5%31.8%

1/24/06264.9%24.8%64.3%24.7%61.8%24.1%63.8%22.8%

1/25/06353.6%21.7%53.1%21.6%50.2%20.1%51.7%18.5%

1/26/06449.7%20.6%49.1%20.4%44.3%18.0%45.6%16.3%

1/27/06547.7%20.2%47.8%20.1%40.9%18.3%40.7%15.5%

1/28/06649.1%20.1%49.0%20.0%42.0%18.1%41.8%15.5%

1/29/06747.4%19.3%47.5%19.2%40.4%17.3%40.2%14.7%

1/30/06846.5%19.6%46.6%19.5%38.6%17.5%38.4%14.6%

1/31/06944.2%19.4%44.2%19.4%44.2%19.4%44.9%17.2%

2/1/061044.6%19.0%44.6%19.0%44.6%19.0%45.5%17.0%

2/2/0611

2/3/0612

2/4/0613

2/5/0614

2/6/0615

2/7/0616

2/8/0617

2/9/0618

2/10/0619

2/11/0620

2/12/0621

2/13/0622

2/14/0623

2/15/0624

2/16/0625

2/17/0626

2/18/0627

2/19/0628

2/20/0629

2/21/0630

Empty Cache %

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

00

B Cookies

Page Views

Day

Percentage

Empty Cache Percentages

Duplicate 200

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

00000000

B Cookies

Page Views

Remove Old B

Remove Old B PVs

Remove New B

Remove New B PVs

Remove Extreme 200s

Remove Extreme 200s PVs

Day

Percentage

Empty Cache Percentages

Sheet1

17938076.20%

21149811.04%

342504.08%

421552.07%

513851.33%

69480.91%

76900.66%

85350.51%

93800.36%

103190.31%

112670.26%

122390.23%

131890.18%

141620.16%

151460.14%

161200.12%

171230.12%

181100.11%

19860.08%

20970.09%

21830.08%

22500.05%

23680.07%

24580.06%

25600.06%

26560.05%

27480.05%

28450.04%

29280.03%

30400.04%

31220.02%

32280.03%

33170.02%

34170.02%

35260.02%

36230.02%

37150.01%

38230.02%

39220.02%

40180.02%

41140.01%

42250.02%

4390.01%

44140.01%

45140.01%

4690.01%

4790.01%

48110.01%

49110.01%

50110.01%

5190.01%

5290.01%

5360.01%

5470.01%

5550.00%

5640.00%

5750.00%

5850.00%

5930.00%

6050.00%

6140.00%

6250.00%

6340.00%

6470.01%

6580.01%

6640.00%

6770.01%

6840.00%

6960.01%

7080.01%

7150.00%

7230.00%

7330.00%

7430.00%

7520.00%

7650.00%

7720.00%

7810.00%

8020.00%

8140.00%

8230.00%

8330.00%

8410.00%

8530.00%

8620.00%

8830.00%

8910.00%

9010.00%

9110.00%

9210.00%

9320.00%

9420.00%

9620.00%

9710.00%

9910.00%

10130.00%

10230.00%

10310.00%

10510.00%

10610.00%

10710.00%

11210.00%

11310.00%

11830.00%

11910.00%

12010.00%

12210.00%

12310.00%

12510.00%

12610.00%

13110.00%

13310.00%

13520.00%

13810.00%

14030.00%

14130.00%

14310.00%

14910.00%

15510.00%

16610.00%

16810.00%

19110.00%

20110.00%

21110.00%

21510.00%

23510.00%

23610.00%

25510.00%

25610.00%

29410.00%

44310.00%

Sheet1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

# of 200 Responses per B Cookie

Percentage of B Cookies

Duplicate 200 Responses

Sheet3

Sheet3

11

0.9756970610.3511414856

0.64892420720.2480818574

0.5362557680.2172068146

0.49705641450.205786832

0.47695751880.2014351966

0.49093842630.2013420285

0.47437852410.1930321497

0.46505295010.1954734765

0.45392525710.1938962941

0.44627662250.1903622437

0.44449787410.1888348329

0.4392663860.1877929461

0.45289059970.1891742737

0.4466963780.186161091

0.44525710320.1868084979

0.43449977570.1843509261

17

18

19

20

unique users with empty cache

page views with empty cache

day of experiment

percentage

1

0.3511414856

0.2480818574

0.2172068146

0.205786832

0.2014351966

0.2013420285

0.1930321497

0.1954734765

0.1938962941

0.1903622437

0.1888348329

0.1877929461

0.1891742737

0.186161091

0.1868084979

0.1843509261

17

18

19

20

Baseline

Day

Percentage

Front Page - Page Views with Empty Cache

01234567891011121314151617181920

1/22/061/23/061/24/061/25/061/26/061/27/061/28/061/29/061/30/061/31/062/1/062/2/062/3/062/4/062/5/062/6/062/7/062/8/062/9/062/10/062/11/06

Baseline# B cookies with 20031,07425,96821,15119,41917,47013,08413,37718,44418,09817,72717,56316,07112,58912,62917,95917,433

total # B cookies31,84840,01739,44239,06836,62826,65128,19939,66039,87039,72239,51236,58627,79728,27240,33440,122

%age100.0%97.6%64.9%53.6%49.7%47.7%49.1%47.4%46.5%45.4%44.6%44.4%43.9%45.3%44.7%44.5%43.4%

# PVs with 20040,74439,83534,19432,06229,36221,33421,60332,10331,24630,46930,15927,48520,91020,51731,08429,889

total # PVs116,033160,572157,426155,802145,764105,959111,914164,232161,148160,058159,711146,358110,533110,211166,395162,131

%age100.0%35.1%24.8%21.7%20.6%20.1%20.1%19.3%19.5%19.4%19.0%18.9%18.8%18.9%18.6%18.7%18.4%

Filter Old B# B cookies with 20030,39525,35320,66218,94617,09012,78613,08818,05717,71317,36817,18315,75512,31912,35417,61317,051

total # B cookies31,14039,10638,54038,13335,81326,07727,57138,78338,99638,87538,62135,82627,21627,65339,51139,246

%age100.0%97.6%64.8%53.6%49.7%47.7%49.0%47.5%46.6%45.4%44.7%44.5%44.0%45.3%44.7%44.6%43.4%

# PVs with 20039,77738,86333,33631,19228,68020,82221,14931,35030,53129,80529,49126,85420,43420,01730,33929,151

total # PVs114,022157,771154,727153,013143,161104,180110,177161,360158,452157,410157,064143,980108,858108,399163,670159,435

%age100.0%34.9%24.6%21.5%20.4%20.0%20.0%19.2%19.4%19.3%18.9%18.8%18.7%18.8%18.5%18.5%18.3%

Filter Extreme# B cookies with 20030,33325,29220,60418,88317,04412,75213,04617,97717,65317,30317,12215,70112,28012,32417,54616,984

total # B cookies31,07738,96238,41537,99135,69126,00127,48438,60838,85438,72638,48635,70927,14027,57839,36339,089

%age100.0%97.6%64.9%53.6%49.7%47.8%49.0%47.5%46.6%45.4%44.7%44.5%44.0%45.2%44.7%44.6%43.4%

# PVs with 20039,58638,12932,70930,45628,04420,38120,74230,42429,63928,97828,74726,21719,88619,53229,68228,428

total # PVs107,953147,896145,375142,382133,73898,386104,437149,560148,669146,825146,713134,631102,216102,631153,147147,846

%age100.0%36.7%25.8%22.5%21.4%21.0%20.7%19.9%20.3%19.9%19.7%19.6%19.5%19.5%19.0%19.4%19.2%

Filter New B# B cookies with 20026,89319,84414,86812,85411,2638,1478,11510,82210,43910,1039,6788,8616,7376,7279,5018,967

total # B cookies27,56932,63131,08429,88427,48119,33820,05928,04527,78727,44026,73824,59817,98018,29326,17125,620

%age100.0%97.5%60.8%47.8%43.0%41.0%42.1%40.5%38.6%37.6%36.8%36.2%36.0%37.5%36.8%36.3%35.0%

# PVs with 20035,31430,95524,79221,87619,63713,67213,56419,83318,92718,25717,73316,02011,66311,20717,42616,280

total # PVs97,271126,119118,505112,486102,39172,56375,828108,251105,590103,097101,02891,43166,17366,340100,16295,255

%age100.0%36.3%24.5%20.9%19.4%19.2%18.8%17.9%18.3%17.9%17.7%17.6%17.5%17.6%16.9%17.4%17.1%

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

Baseline

Remove Old B Cookies

Remove Extreme B Cookies

Remove New B Cookies

Day

Percentage

Front Page - B Cookies with Empty Cache

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

0000

Baseline

Remove Old B Cookies

Remove Extreme B Cookies

Remove New B Cookies

Day

Percentage

Front Page - Page Views with Empty Cache

Experiment TakeawaysKeep in mind the empty cache user experience. It might be more prevalent than you think!

Use different techniques to optimize full versus empty cache experience.

http://yuiblog.com/blog/2007/03/01/performance-research-part-3

HTTP Quick ReviewHTTP response header sent by the web server:HTTP/1.1 200 OKContent-Type: text/html; charset=utf-8Set-Cookie: C=abcdefghijklmnopqrstuvwxyz; domain=.yahoo.com

HTTP Quick ReviewHTTP request header sent by the browser:GET / HTTP/1.1Host: finance.yahoo.comUser-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Cookie: C=abcdefghijklmnopqrstuvwxyz;

HTTP Quick ReviewHTTP request header sent by the browser:GET / HTTP/1.1Host: autos.yahoo.comUser-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Cookie: C=abcdefghijklmnopqrstuvwxyz;

HTTP Quick ReviewHTTP request header sent by the browser:GET / HTTP/1.1Host: mail.yahoo.comUser-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Cookie: C=abcdefghijklmnopqrstuvwxyz;

HTTP Quick ReviewHTTP request header sent by the browser:GET / HTTP/1.1Host: tech.yahoo.comUser-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Cookie: C=abcdefghijklmnopqrstuvwxyz;

Impact of Cookies on Response Time

.yahoo.com cookie sizes

Chart8

0.28860.5180.17790.0155

1-500 bytes

501-1000 bytes

1001-1500 bytes

over 1501 bytes

percentage of page views

bcookie

Cookie Histograms:

Num CookiesAll CookiesAll IEAll FFno RTBAno RTBA IEno RTBA FFNum CookiesAll CookiesAll IEAll FFno RTBAno RTBA IEno RTBA FFNum CookiesAll CookiesAll IEAll FFno RTBAno RTBA IEno RTBA FF

000051300.00%0.00%0.00%4.95%1.54%100.00%00.00%0.00%0.00%4.95%1.54%100.00%

151300015.05%1.59%100.00%0.00%0.00%0.00%15.05%1.59%100.00%4.95%1.54%100.00%

200000020.00%0.00%0.00%0.00%0.00%0.00%25.05%1.59%100.00%4.95%1.54%100.00%

300000030.00%0.00%0.00%0.00%0.00%0.00%35.05%1.59%100.00%4.95%1.54%100.00%

400000040.00%0.00%0.00%0.00%0.00%0.00%45.05%1.59%100.00%4.95%1.54%100.00%

500011050.00%0.00%0.00%0.99%1.54%0.00%55.05%1.59%100.00%5.94%3.08%100.00%

611000061.01%1.59%0.00%0.00%0.00%0.00%66.06%3.17%100.00%5.94%3.08%100.00%

700000070.00%0.00%0.00%0.00%0.00%0.00%76.06%3.17%100.00%5.94%3.08%100.00%

800000080.00%0.00%0.00%0.00%0.00%0.00%86.06%3.17%100.00%5.94%3.08%100.00%

900000090.00%0.00%0.00%0.00%0.00%0.00%96.06%3.17%100.00%5.94%3.08%100.00%

10000000100.00%0.00%0.00%0.00%0.00%0.00%106.06%3.17%100.00%5.94%3.08%100.00%

11000000110.00%0.00%0.00%0.00%0.00%0.00%116.06%3.17%100.00%5.94%3.08%100.00%

12000110120.00%0.00%0.00%0.99%1.54%0.00%126.06%3.17%100.00%6.93%4.62%100.00%

13110330131.01%1.59%0.00%2.97%4.62%0.00%137.07%4.76%100.00%9.90%9.23%100.00%

14330110143.03%4.76%0.00%0.99%1.54%0.00%1410.10%9.52%100.00%10.89%10.77%100.00%

15110220151.01%1.59%0.00%1.98%3.08%0.00%1511.11%11.11%100.00%12.87%13.85%100.00%

16220720162.02%3.17%0.00%6.93%3.08%0.00%1613.13%14.29%100.00%19.80%16.92%100.00%

1772025120177.07%3.17%0.00%24.75%18.46%0.00%1720.20%17.46%100.00%44.55%35.38%100.00%

1825120564201825.25%19.05%0.00%55.45%64.62%0.00%1845.45%36.51%100.00%100.00%100.00%100.00%

19544000001954.55%63.49%0.00%0.00%0.00%0.00%19100.00%100.00%100.00%100.00%100.00%100.00%

99633101653

Cookie Size Histograms:

Size LTEAllIEFFSize LTEAllIEFFSize LTEAllIEFF

5006235005.94%3.08%100.00%5005.94%3.08%100.00%

100017150100016.83%23.08%0.00%100022.77%26.15%100.00%

150061360150060.40%55.38%0.00%150083.17%81.54%100.00%

200016120200015.84%18.46%0.00%200099.01%100.00%100.00%

250010025000.99%0.00%0.00%2500100.00%100.00%100.00%

300000030000.00%0.00%0.00%3000100.00%100.00%100.00%

350000035000.00%0.00%0.00%3500100.00%100.00%100.00%

400000040000.00%0.00%0.00%4000100.00%100.00%100.00%

450000045000.00%0.00%0.00%4500100.00%100.00%100.00%

500000050000.00%0.00%0.00%5000100.00%100.00%100.00%

550000055000.00%0.00%0.00%5500100.00%100.00%100.00%

600000060000.00%0.00%0.00%6000100.00%100.00%100.00%

650000065000.00%0.00%0.00%6500100.00%100.00%100.00%

700000070000.00%0.00%0.00%7000100.00%100.00%100.00%

750000075000.00%0.00%0.00%7500100.00%100.00%100.00%

800000080000.00%0.00%0.00%8000100.00%100.00%100.00%

850000085000.00%0.00%0.00%8500100.00%100.00%100.00%

101653100.00%100.00%100.00%

All .yahoo.com cookies:

101 BA (sample: 'BA=ba=743&ip=71.94.10.227&t=1169430685')

95 Y (sample: 'Y=v=1&n=6f7bk96vejto2&l=9ij4l4dij864hi/o&p=m222p6eb13000100&jb=26|22|&r=eq&lg=us&intl=us&np=1')

95 F (sample: 'F=a=eEJdC.IsvXQyP.tLgPkB9WrbHgTl.myc59dvWujCFbUtFFryxx3JmZ0tz57I&b=ErWJ')

93 PH (sample: 'PH=fn=oQdSi4oEdBKDgw--')

92 C (sample: 'C=mg=1')

91 Q (sample: 'Q=q1=AACAAAAAAAAAAA--&q2=RaNEug--')

91 T (sample: 'T=z=z5RtFBz/mtFBFNDpbWwgYKHNDM0BjY0TjJOMTM3TjY-&a=YAE&sk=DAAkqtG71L4NIY&d=c2wBTXpRekFURXpPVFU1TmpRd09URS0BYQFZQUUBb2sBWlcwLQF0aXABSDVsQ2pEAXp6AXo1UnRGQmdXQQ--')

87 LYC (sample: 'LYC=l_v=0&l_lv=10&l_s=wxr2v14sv4zuuzr2y0ttqtqqs1x32wyr&l_um=0_0_1_0_0')

84 U (sample: 'U=mt=oJEOxJ2MhYoA9RYUqrdXOKQ535ndGY6O5GItOg--&ux=gJEeFB&un=3vrbop721di81')

81 I (sample: 'I=ir=h8&in=486ecde9&i1=AAACAAAEBhABqIDDACqZti')

70 PL (sample: 'PL=V=1.1&d=TzW9McffGIN5LeNvHFKh2.wmYJwxCm2VPu_t9lBj21WaASiTYWMoImBe73dzoZVVr7oiyKw0KGKAK9BF8.7v.oz7dUVqc1BzC0YyoIqPXaxKj2whaUJ8y2BsFxWCRekP4iyEAHzlpgevCvCKxvpDr_9xkOXrhjpkdkVxqLuaUA5ePpQmrityKeZsRydf5d_lWS0qFSzUz0OJFtPRhi.Zqy.cFJ3qct3YUE

69 SO (sample: 'SO=v=0.4&t=1168368499')

66 CP (sample: 'CP=v=60306&br=i&sp=')

60 YLS (sample: 'YLS=v=1&p=0&n=9')

53 HP (sample: 'HP=1')

51 sS (sample: 'sS=shf=2&ftrs=ourwebact%7Cprs&rdsng=1&l=9ij4l4dij864hi/o')

39 PA (sample: 'PA=p0=5kr3St1K6kvSSw--&e=cVotFB&p3=ANw5IHkACpU86Dgged05AAAACpU85zkged05AACACpU8zTkged05AACACpU80Dgged05AAA-')

38 LYS (sample: 'LYS=l_fh=0&l_vo=myla')

35 D (sample: 'D=_ylc=X3IDMwRmcANyLzIx&_ylg=X3oDMTRocWtxMTgzBGZwdWxtAzEwMDAEeXB1bHNlA2NvbnNlcnIEcGEDLTEEcGNpZANHYW1lc19Nb3N0X1BvcHVsYXJfV2FyY3JhZnRfZWxmd29tYW5fM0UtNTI5OTM2BHBwaWQDMTE2OTI2NTY3MgR0bXBsA2luZGV4LWwEX1MDMjcxNjE0OQRwaWQDMTE2OTUwMjE0MAR0ZXN0A

33 YGCV (sample: 'YGCV=d=')

30 YMPS (sample: 'YMPS=auth=e%2FYXFbeRDhRzHJBNvqxa3Q%3D%3D&t=1169503229&yid=6ZOcCjBst%2FIm3B%2F1I5C9WA%3D%3D&bf=0')

28 l%5FPD3 (sample: 'l%5FPD3=840')

27 YMUSIC (sample: 'YMUSIC=t=1')

25 MVUserInfo3 (sample: 'MVUserInfo3=123_9')

24 YMRAD (sample: 'YMRAD=187841048*0_0_1204_1')

23 yme (sample: 'yme=1')

15 playerFullVersion (sample: 'playerFullVersion=10')

14 PVL (sample: 'PVL=38856941_37868940_35881672_37188836_36110701_36465596_34613518_31363511_36886779_37987962_37161962_35466915_2154453_39155872_36511587_37159165_37943612_37146469_37159933_36394684_38510337_39155866_36962718_35112301|26_26_26_26_26_26

13 op122geohomeliid (sample: 'op122geohomeliid=a00602d0mc231ci0hk2fsbd34')

13 op122geohomegum (sample: 'op122geohomegum=a00602d0mc231ci0hk2fsbd34')

13 LCGuLi (sample: 'LCGuLi=1049601=1')

9 UC (sample: 'UC=E90efOXSQJVcrIc7kJ28hFmklH8cDxAknk9DVBgWTV5VQl.sCjp1jE46F8ScvZnwTLF.8RMJ0hjgPOH_jtC4lV9fmf12qkAw4HvEoYIGk7xbEjouSRLdFRdbY.7fd4nreXhaqskdFUkhdJDW10v7HpFmI8hkJct3ZfUCSzVYBDU1_eqPzSmMZ8sVamn3w9FFkondBc6H0kb2iJzy7QD4emIyq6Np90foalixVmL7TNV

9 sB (sample: 'sB=vm=p&subscrs=')

8 adx (sample: 'adx=c9809@1169484121@1')

6 LCSoftStick (sample: 'LCSoftStick=ym=20070')

5 op122WHhomeliid (sample: 'op122WHhomeliid=a002022034231mh07e34g56e5')

5 vCodec (sample: 'vCodec=9')

5 LCTrSk (sample: 'LCTrSk=1169499580_0_0_171226794')

5 op122WHhomegum (sample: 'op122WHhomegum=a002022034231mh07e34g56e5')

3 ykH (sample: 'ykH=w=&f=&c=1&pc=1&p=&csm=NNGjVeqjhLG0EdKJOkRagQ--&ts=1145483966')

3 ykPCD (sample: 'ykPCD=d=5Wj8mVM-&csm=NNGjVeqjhLG0EdKJOkRagQ--')

2 LSundefined_LYS (sample: 'LSundefined_LYS=l_fh=1&l_vo=myla')

2 LSundefined_LYC (sample: 'LSundefined_LYC=l_v=0&l_lv=10&l_s=xwyxt053s35ru3y31s52xy3wsy0yrt1s&l_lc=0_1_0_0_-1&l_um=0_0_1_0_0&l_mv=56_50')

2 op122homepagegum (sample: 'op122homepagegum=a0070230kf2316j0612dn1870')

2 op122homepageliid (sample: 'op122homepageliid=a0070230kf2316j0612dn1870')

2 RT (sample: 'RT=s=1169503282553&u=&r=http%3A//www.yahoo.com/')

2 SS (sample: 'SS=u1=todsteno&[email protected]&v=1.0&sig=CTnepscty8NoYdGVIw4jyg--')

2 MVFSWarning (sample: 'MVFSWarning=1')

2 YM.BM (sample: 'YM.BM=domain=zender-engr.net&sid=2291779466&silo=ms333&farm=b3&account=sgilbreath&fwdaddr=&fwdtemp=&quota=2147483648&BD=b')

2 FPS (sample: 'FPS=dl')

2 INT (sample: 'INT=globeph=1:')

2 R (sample: 'R=o=1/i=search/iy/c3231_f1970/jackson+guitar/shopping_narrower_main/sig=12h4q2fe7/t=1169435451')

2 ymttrlsvst (sample: 'ymttrlsvst=000000010000000000000')

1 YCorp (sample: 'YCorp=id%3D111312%26userid%3Dlehmbecj%26sign%3D69b72a54a0ae09cf7f54e4f1b15f0006%26time%3D1169482827%26expires%3D480%26ip%3D10.72.55.109%26roles%3D%7C1.IE%7C121.U%7C13.V%7C20.U%7C212.E%7C50.U%7C6878.U%7C6951.I%7C6982.I%7C7033.U%7C7181.I

1 ASLTRG1 (sample: 'ASLTRG1=28#42#.net#3133#20993#.sd.sd.cox.net#72.197.249.102#en-us%2Cen%3Bq%3D0.5#ISO-8859-1%2Cutf-8%3Bq%3D0.7%2C*%3Bq%3D0.7#32706454#20332#14659#825#5#0')

1 FRQSTR (sample: 'FRQSTR=19487461|19487461|19487461|19487461|19487461')

1 FRhp_BouyguesCUST (sample: 'FRhp_BouyguesCUST=t=1')

1 K (sample: 'K=kl=ggs09e9e/a&kn=cXEyYWpvam8-&km=cXEyYUBzaW5hLmNvbS50dw--&kp=aa1vvvjcN&ko=ggs09e9e/a&kk=ggs09e9e/o&kc=0')

1 KIDYMD (sample: 'KIDYMD=#342018:GATA#')

1 KR_DIC (sample: 'KR_DIC=qh=cmluZyB0aGUgcmVnaXN0ZXJ8fA--')

1 s_cc (sample: 's_cc=true')

1 s_sq (sample: 's_sq=undefinedtoJSONString%2CtoJSONString%3Dfunction%2520%2528%2529%2520%257B%250A%2520%2520%2520%2520return%2520s.object%2528this%2529%253B%250A%257D%26function%20%28%29%20%7B%0A%20%20%20%20return%20s.object%28this%29%3B%0A%7D%3DtoJSON

1 stageLYC (sample: 'stageLYC=l_v=0&l_lv=9&l_s=yt3vx3s0qyrzux1r0tv54uqw1qtrv1su&l_um=0_0_1_0_0')

1 stageLYS (sample: 'stageLYS=l_fh=0&l_vo=myla')

1 KR_S1 (sample: 'KR_S1=qh=uei9vbHifHy44bfQfHzDtbjtyMZ8fLnpvcLB1nx8x8.8rsH4fHxzaWxrIHJvYWQgsNTA03x8&qc=2')

1 wgb (sample: 'wgb=b=5C3a42bfd101597645b5288a6882ed542bfd16a2f1418bd546eff7cc4978b7ac4&l=&s=Bpf9buBCCYoezbPVRdXaiqNzMbg-')

1 idm (sample: 'idm=b7CfG_fQ2whCkBWIzt52zOlagiGZh9saHtwfKL4OMe5Pk0tFo2zUsHeXyl2gB_Vr9JCBp65Htg3Pt6VrXAejhNCYbHZyvFeLzzS_DQ3O8y3gh3fCRpDCFgPkLyT0')

1 SESTEST (sample: 'SESTEST=1')

1 WIDYMD (sample: 'WIDYMD=#40669:GAT#')

1 YBY (sample: 'YBY=id%3D111312%26userid%3Dlehmbecj%26sign%3DLE8Q9fZad9YJALlii6IFYASCWP_Zukl4mkHu4.7dTGiTAs3cybltpZ2ZSjaN1viKKzMs1wWCjR1CbAZf27JTJta.RGQE0pI_F0xpCzpKPiERAF_ImptD1HK3jqQNXyr0a40q3Dlmag9EFwjNswEAoQQ5HEDyE9xAdYWKjJ7rk6w-%26time%3D1169482827

bcookie

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

000000

All Cookies

All IE

All FF

no RTBA

no RTBA IE

no RTBA FF

Number of Cookies

Cumulative Distribution of Number of Cookies in Page Views w/out B-Cookie

Sheet1

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

All

IE

FF

Size (bytes)

Size of Cookies in Page Views w/out B-Cookie

cookie-expt

dffdfsff

A.yahoo.comadvertiser.yahoo.com statsDocument Link

[email protected] ad debuggingDocument Link

[email protected] just enough information

to rebuild a user's session

on failover

AFFINITYpersonals.yahoo.comPersonals affinity searchDocument Link

[email protected] Options

("AO") cookie and

sub-cookie "o"

(opt-out)

[email protected].*.yahoo.comIncremental value to store the

number of times the user has

viewed the various AUNZ

mediapanes

[email protected] avatar cookie

B.yahoo.comunique browseridDocument Link

[email protected] cookie - containsDocument Link

user's bandwidth capacity in

kbps

[email protected] JS / NagBar Info

C.yahoo.comclient software capabilitiesDocument Link

chanAreatv.yahoo.comtv listings areaDocument Link

CHT.yahoo.comchat.yahoo.com useridDocument Link

CNMSGRcn.messenger.yahoo.comCN messenger cookieDocument Link

CN_AUTOS_DIAcn.autos.yahoo.comY! CN autos cookieDocument Link

CN_ENT_DIAcn.ent.yahoo.comY! CN ent cookieDocument Link

[email protected] .yahoo.com.cnCN_FID cookie for external flow

.3721.comstatistic

[email protected]

[email protected]

[email protected]!CN web messenger for mail

CN_NEWS_DIAcn.news.yahoo.comY! CN news cookieDocument Link

CN_REcn.realestate.yahoo.comY! CN realestate cookieDocument Link

CN_RE_DIAcn.realestate.yahoo.comY! CN realestate cookieDocument Link

[email protected] cn sds team

CN_SPORTS_DIAcn.sports.yahoo.comY! CN sports cookieDocument Link

CN_TBCTcn.sports.yahoo.comY! CN TextBroadcast cookieDocument Link

[email protected] by merchants during theDocument Link

checkout v3 beta period for

redirect purpose

CP.yahoo.comCompanion toolbar detectionDocument Link

CUcn.yahoo.comCN popunder adDocument Link

cz_INTcn.yahoo.comCN crazyadDocument Link

[email protected] ULT Data CookieDocument Link

E.groups.yahoo.comYahoo! Groups session cookieDocument Link

(timezone, preferences, etc)

ezvezventure.yahoo.comEZVentureDocument Link

ezwezwheels.yahoo.comEZWheelsDocument Link

F.yahoo.comfriend/fingerprintDocument Link

FIFAfifaworldcup.yahoo.comdefault language forDocument Link

fifaworldcup

[email protected] Sports -

timezone Fantasy Sports -

timezone information

Gclubs.yahoo.comclub-specific information (v1Document Link

and v2)

gbversiongeocities.yahoo.comGeoBuilder versionDocument Link

GeoIdgeocities.yahoo.com?Document Link

[email protected]"Gladiator" cookie to

track homepage or seach page tab

buttons - user selection

preference

[email protected] cookie is a session cookie

holding state settings for the

YHS vertical in GSP

[email protected]*.hothire.hotjobs.comThis is the cookie to be used byDocument Link

the HotJobs HotHire and is used

for identifying users.

[email protected] cookie is used for Hotjobs

customer when they login to

Hotjobs(e.g for posting jobs).

[email protected] for anonymous user

session in Hotjobs property.

This will contain user selected

items in shopping cart.

[email protected] is used for storing yahooDocument Link

internal users who access

Hotjobs system through our

Hotjobs application. Yahoo

internal users are capable of

managing users for customers and

also post jobs for customers.

HMSSpersonals.yahoo.comPersonals homepage my savedDocument Link

search cookie

[email protected] indicates what Yahoo! page

is the users home page

HWpersonals.yahoo.comPersonals homepage wildcardDocument Link

cookie

I.yahoo.comdata-mining user interestsDocument Link

[email protected] token for

o-inc.combusiness user in YSM Panama

INTcn.yahoo.comCN floating and couplet adDocument Link

JC2.yahoo.comOBSOLETE - jfriedl page checkerDocument Link

JFriedlPageChecker.yahoo.comjeffrey's page checkerDocument Link

jrunsessionidgeocities.yahoo.compagebuilder infoDocument Link

K.yahoo.coma child (

Analysis of Cookie Sizes across the Web

Total Cookie SizeAmazon60 bytesGoogle72 bytesYahoo122 bytesCNN184 bytesYouTube218 bytesMSN268 byteseBay331 bytesMySpace500 bytes

Experiment Takeawayseliminate unnecessary cookieskeep cookie sizes lowset cookies at appropriate domain levelset Expires date appropriatelyearlier date or none removes cookie sooner

Parallel DownloadsTwo componentsin parallelper hostnameHTTP/1.1

Parallel DownloadsTwo in parallelFour in parallelEight in parallel

Chart17

0.50.3

0.50.3

0.20.3

0.20.3

0.20.3

0.20.3

0.20.3

0.20.3

0.20.3

0.20.3

00.2

Sheet1

"wd:dItemDuration">PT0.018238SPT0.288027SPT0.026868SPT0.103763SPT0.018247SPT0.053908SPT0.030719SPT0.044826SPT0.172498SPT0.145024SPT0.169048SPT0.192500SPT0.232298SPT0.162706SPT0.098428SPT0.071543SPT0.048729SPT0.032360SPT0.030886SPT0.034181SPT0.062929SPT0.147190SPT0.146844SPT0.071666SPT0.027494SPT0.235455SPT0.223070SPT0.075842SPT0.022563SPT0.023736SPT0.017868SPT0.115827SPT0.250341SPT0.035140SPT0.073697SPT0.018238SPT0.288027SPT0.026868SPT0.103763SPT0.018247SPT0.053908SPT0.030719SPT0.044826SPT0.172498SPT0.145024SPT0.169048SPT0.192500SPT0.232298SPT0.162706SPT0.098428SPT0.071543SPT0.048729SPT0.032360SPT0.030886SPT0.034181SPT0.062929SPT0.147190SPT0.146844SPT0.071666SPT0.027494SPT0.235455SPT0.223070SPT0.075842SPT0.022563SPT0.023736SPT0.017868SPT0.115827SPT0.250341SPT0.035140SPT0.073697SPT0.018238SPT0.288027SPT0.026868SPT0.103763SPT0.018247SPT0.053908SPT0.030719SPT0.044826SPT0.172498SPT0.145024SPT0.169048SPT0.192500SPT0.232298SPT0.162706SPT0.098428SPT0.071543SPT0.048729SPT0.032360SPT0.030886SPT0.034181SPT0.062929SPT0.147190SPT0.146844SPT0.071666SPT0.027494SPT0.235455SPT0.223070SPT0.075842SPT0.022563SPT0.023736SPT0.017868SPT0.115827SPT0.250341SPT0.035140SPT0.073697S

Maximizing Parallel Downloadsresponse time (seconds)aliases


Chart18

0.5940.781

0.3440.578

0.3281.016

0.3281.141

0.3131.313

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

SRP

PRELOAD SRP COMPONENTS

Sample Source URL:

preload OFFpreload ONpreload OFFpreload ONpreload OFFpreload ON

Summary(average of medians)dialupbroadbandlan

fspl=0fspl=1fspl=0fspl=1fspl=0fspl=1

RT# pointsRT# pointsdiff%changeRT# pointsRT# pointsdiff%changeRT# pointsRT# pointsdiff%change

00000.000.00%0.39087560.96887500.58147.90%00000.000.00%

# hostsdialupbroadbandlan

fspl=0fspl=1s=ss=mfspl=0fspl=1


10.6251540.844154

20.3751530.578149

40.3281521.031151

50.3131481.094148

100.3131491.297148

0.6251540.844154

0.3751530.578149

0.3281521.031151

0.3131481.094148

0.3131491.297148

SRP

00

00

00

00

00

&A

Page &P

SYC

sidDateDialup MedianDialup 95th PctlDialup Data PointsBroadband MedianBroadband 95th PctlBroadband Data PointsLan MedianLan 95th PctlLan Data Points

958104692/20/0776302662514515786484206969279779

958104692/21/0742182154610115164296222953317260

958104692/22/0754062575012814015094223781223455

958104692/23/0771882077010711875141216987265781

958104692/24/075516222229312094000212761242666

958104692/25/0763752448412512973745180906234456

958104692/26/0773002390614812814579265969232369

958104692/27/0743971700089121947461371142300063

958104692/28/07551816234471875646978907346032

958104693/1/073031152345716726700631232447623

958104693/2/07759319750491216487539791275023

958104693/3/0754851785913214224765217610245368

958104693/4/0773352296010513315047202844248461

958104693/5/0750472164112612344640224787307861

958104693/6/07442220349163142050772471234339092

958104693/7/07482818526155134452082188913093102

958104693/8/07570818837134129750072271032243862

958104693/9/0765312214013512114125273861309470

958104693/10/0758752372411414425203222711207883

958104693/11/0776722345411711514718223750209471

958104693/12/07715722484160146948522661094262554

958104693/13/0762352046817813755308228931285979

958104693/14/0771722675012512344547209984365669

958104693/15/0763132525013515474297247591276594

958104693/16/0747032504717616225219216930256377

958104693/17/0786102759411911885157199843276675

958104693/18/0768282712514713433906205984271986

958104693/19/07598424204157137549382391046345394

958104693/20/0750632066012713624500219984253178

958104693/21/07700021040215158447972031062408671

958104693/22/07632822437151137543402361072359471

SYC

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

Dialup Median

Broadband Median

Lan Median

Sheet1

Sheet1

0.5940.7810.6875

0.3440.5780.461

0.3281.0160.672

0.3281.1410.7345

0.3131.3130.813

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average of median

Sheet3

number hostsnnumber imagesimage typeMedian Response Time# of Data PointsNumber of Hosts36 x 36 px (0.9 Kb)116 x 61 px (3.4 Kb)LargeaverageSmallMediumLargeAverageDialup

12020s=l3.1424610.5940.7813.1400.6881.0781.4686.2182.92133333334.719

21020s=l2.95324620.3440.5782.9530.4610.611.0475.7652.4744.610

4520s=l2.93724740.3281.0162.9370.6720.5471.2185.7342.49966666679.391

5420s=l2.93724650.3281.1412.9370.7350.5471.255.7342.51033333339.312

10220s=l3.5245100.3131.3133.5000.8130.5471.4228.625

12020s=m0.781247Number of Hosts36 x 36 px (0.9 Kb)

21020s=m0.57824714.719

4520s=m1.01624724.610

5420s=m1.14124749.391

10220s=m1.31324559.312

12020s=s0.594246108.625

21020s=s0.344247

4520s=s0.328247

5420s=s0.328247

10220s=s0.313245

14040s=l6.218233

22040s=l5.765236

41040s=l5.734241

5840s=l5.734244

14040s=m1.468246

22040s=m1.047246

41040s=m1.218247

5840s=m1.25245

10440s=m1.422245

14040s=s1.078247

22040s=s0.61247

41040s=s0.547246

5840s=s0.547247

10440s=s0.547245

Sheet3

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)


Chart16

0.5940.7810.6875

0.3440.5780.461

0.3281.0160.672

0.3281.1410.7345

0.3131.3130.813

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average

SRP


Sample Source URL:





00000.000.00%0.39087560.96887500.58147.90%00000.000.00%




10.6251540.844154

20.3751530.578149

40.3281521.031151

50.3131481.094148

100.3131491.297148

0.6251540.844154

0.3751530.578149

0.3281521.031151

0.3131481.094148

0.3131491.297148

SRP

00

00

00

00

00

&A

Page &P

SYC


958104692/20/0776302662514515786484206969279779

958104692/21/0742182154610115164296222953317260

958104692/22/0754062575012814015094223781223455

958104692/23/0771882077010711875141216987265781

958104692/24/075516222229312094000212761242666

958104692/25/0763752448412512973745180906234456

958104692/26/0773002390614812814579265969232369

958104692/27/0743971700089121947461371142300063

958104692/28/07551816234471875646978907346032

958104693/1/073031152345716726700631232447623

958104693/2/07759319750491216487539791275023

958104693/3/0754851785913214224765217610245368

958104693/4/0773352296010513315047202844248461

958104693/5/0750472164112612344640224787307861

958104693/6/07442220349163142050772471234339092

958104693/7/07482818526155134452082188913093102

958104693/8/07570818837134129750072271032243862

958104693/9/0765312214013512114125273861309470

958104693/10/0758752372411414425203222711207883

958104693/11/0776722345411711514718223750209471

958104693/12/07715722484160146948522661094262554

958104693/13/0762352046817813755308228931285979

958104693/14/0771722675012512344547209984365669

958104693/15/0763132525013515474297247591276594

958104693/16/0747032504717616225219216930256377

958104693/17/0786102759411911885157199843276675

958104693/18/0768282712514713433906205984271986

958104693/19/07598424204157137549382391046345394

958104693/20/0750632066012713624500219984253178

958104693/21/07700021040215158447972031062408671

958104693/22/07632822437151137543402361072359471

SYC

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

Dialup Median

Broadband Median

Lan Median

Sheet1

Sheet1

0.5940.7810.6875

0.3440.5780.461

0.3281.0160.672

0.3281.1410.7345

0.3131.3130.813

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average of median

Sheet3


12020s=l3.1424610.5940.7813.1400.6881.0781.4686.2182.92133333334.719

21020s=l2.95324620.3440.5782.9530.4610.611.0475.7652.4744.610

4520s=l2.93724740.3281.0162.9370.6720.5471.2185.7342.49966666679.391

5420s=l2.93724650.3281.1412.9370.7350.5471.255.7342.51033333339.312

10220s=l3.5245100.3131.3133.5000.8130.5471.4228.625


21020s=m0.57824714.719

4520s=m1.01624724.610

5420s=m1.14124749.391

10220s=m1.31324559.312

12020s=s0.594246108.625

21020s=s0.344247

4520s=s0.328247

5420s=s0.328247

10220s=s0.313245

14040s=l6.218233

22040s=l5.765236

41040s=l5.734241

5840s=l5.734244

14040s=m1.468246

22040s=m1.047246

41040s=m1.218247

5840s=m1.25245

10440s=m1.422245

14040s=s1.078247

22040s=s0.61247

41040s=s0.547246

5840s=s0.547247

10440s=s0.547245

Sheet3

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average

Maximizing Parallel Downloadsresponse time (seconds)rule of thumb: use at least two but no more than four aliases

Chart16

0.5940.7810.6875

0.3440.5780.461

0.3281.0160.672

0.3281.1410.7345

0.3131.3130.813

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average

SRP


Sample Source URL:





00000.000.00%0.39087560.96887500.58147.90%00000.000.00%




10.6251540.844154

20.3751530.578149

40.3281521.031151

50.3131481.094148

100.3131491.297148

0.6251540.844154

0.3751530.578149

0.3281521.031151

0.3131481.094148

0.3131491.297148

SRP

00

00

00

00

00

&A

Page &P

SYC


958104692/20/0776302662514515786484206969279779

958104692/21/0742182154610115164296222953317260

958104692/22/0754062575012814015094223781223455

958104692/23/0771882077010711875141216987265781

958104692/24/075516222229312094000212761242666

958104692/25/0763752448412512973745180906234456

958104692/26/0773002390614812814579265969232369

958104692/27/0743971700089121947461371142300063

958104692/28/07551816234471875646978907346032

958104693/1/073031152345716726700631232447623

958104693/2/07759319750491216487539791275023

958104693/3/0754851785913214224765217610245368

958104693/4/0773352296010513315047202844248461

958104693/5/0750472164112612344640224787307861

958104693/6/07442220349163142050772471234339092

958104693/7/07482818526155134452082188913093102

958104693/8/07570818837134129750072271032243862

958104693/9/0765312214013512114125273861309470

958104693/10/0758752372411414425203222711207883

958104693/11/0776722345411711514718223750209471

958104693/12/07715722484160146948522661094262554

958104693/13/0762352046817813755308228931285979

958104693/14/0771722675012512344547209984365669

958104693/15/0763132525013515474297247591276594

958104693/16/0747032504717616225219216930256377

958104693/17/0786102759411911885157199843276675

958104693/18/0768282712514713433906205984271986

958104693/19/07598424204157137549382391046345394

958104693/20/0750632066012713624500219984253178

958104693/21/07700021040215158447972031062408671

958104693/22/07632822437151137543402361072359471

SYC

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

000

Dialup Median

Broadband Median

Lan Median

Sheet1

Sheet1

0.5940.7810.6875

0.3440.5780.461

0.3281.0160.672

0.3281.1410.7345

0.3131.3130.813

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average of median

Sheet3


12020s=l3.1424610.5940.7813.1400.6881.0781.4686.2182.92133333334.719

21020s=l2.95324620.3440.5782.9530.4610.611.0475.7652.4744.610

4520s=l2.93724740.3281.0162.9370.6720.5471.2185.7342.49966666679.391

5420s=l2.93724650.3281.1412.9370.7350.5471.255.7342.51033333339.312

10220s=l3.5245100.3131.3133.5000.8130.5471.4228.625


21020s=m0.57824714.719

4520s=m1.01624724.610

5420s=m1.14124749.391

10220s=m1.31324559.312

12020s=s0.594246108.625

21020s=s0.344247

4520s=s0.328247

5420s=s0.328247

10220s=s0.313245

14040s=l6.218233

22040s=l5.765236

41040s=l5.734241

5840s=l5.734244

14040s=m1.468246

22040s=m1.047246

41040s=m1.218247

5840s=m1.25245

10440s=m1.422245

14040s=s1.078247

22040s=s0.61247

41040s=s0.547246

5840s=s0.547247

10440s=s0.547245

Sheet3

36 x 36 px (0.9 Kb)

116 x 61 px (3.4 Kb)

average

Experiment Takeawaysconsider the effects of CPU thrashingDNS lookup times vary across ISPs and geographic locationsdomain names may not be cached

SummaryWhat the 80/20 Rule Tells Us about Reducing HTTP Requestshttp://yuiblog.com/blog/2007/04/11/performance-research-part-4/Browser Cache Usage Exposed!http://yuiblog.com/blog/2007/01/04/performance-research-part-2/When the Cookie Crumbleshttp://yuiblog.com/blog/2007/01/04/performance-research-part-2/Maximizing Parallel Downloads in the Carpool Lanehttp://yuiblog.com/blog/2007/04/11/performance-research-part-4/

14 Rules

14 RulesMake fewer HTTP requestsUse a CDNAdd an Expires headerGzip componentsPut CSS at the topMove JS to the bottomAvoid CSS expressionsMake JS and CSS externalReduce DNS lookupsMinify JSAvoid redirectsRemove duplicate scriptsTurn off ETagsMake AJAX cacheable and small

Rule 1: Make fewer HTTP requestsimage mapsCSS spritesinline imagescombined scripts, combined stylesheets

Image mapsserver-side

http://.../navbar.cgi?127,13client-side preferred

drawbacks:must be contiguousdefining area coordinates tedious, errors

http://www.w3.org/TR/html401/struct/objects.html#h-13.6

CSS Sprites Preferredsize of combined image is lessnot supported in Opera 6http://alistapart.com/articles/sprites

Inline Imagesdata: URL schemedata:[][;base64],

not supported in IEavoid increasing size of HTML pages:put inline images in cached stylesheetshttp://tools.ietf.org/html/rfc2397

Combined Scripts, Combined Stylesheets

Combined Scripts,Combined Stylesheetscombining six scripts into one eliminates five HTTP requestschallenges: develop as separate modulesnumber of possible combinations vs. loading more than neededmaximize browser cacheone solution:dynamically combine and cache

Rule 2: Use a CDN

distribute your static content before distributing your dynamic content

Rule 3: Add an Expires header not just for images

ImagesStylesheetsScripts%Median Ageamazon.com0/620/10/30%114 daysaol.com23/431/16/1848%217 dayscnn.com0/1380/22/111%227 daysebay.com16/200/20/755%140 daysfroogle.google.com1/230/10/14%454 daysmsn.com32/351/13/980%34 daysmyspace.com0/180/20/20%1 daywikipedia.org6/81/12/375%1 dayyahoo.com23/231/14/4100%n/ayoutube.com0/320/30/70%26 days

Rule 4: Gzip components you can affect users' download times90%+ of browsers support compression

Gzip vs. Deflate

Gzip compresses moreGzip supported in more browsers

Gzip: not just for HTML

gzip scripts, stylesheets, XML, JSON (not images, PDF)

HTMLScriptsStylesheetsamazon.comxaol.comxsomesomecnn.comebay.comxfroogle.google.comxxxmsn.comxdeflatedeflatemyspace.comxxxwikipedia.orgxxxyahoo.comxxxyoutube.comxsomesome

Gzip Configuration Apache 2.x: mod_deflateAddOutputFilterByType DEFLATE text/html text/css application/x-javascriptHTTP requestAccept-Encoding: gzip, deflateHTTP responseContent-Encoding: gzipVary: Accept-Encoding

needed for proxies

Gzip Edge Cases

Rule 5: Put CSS at the top stylesheets block rendering in IEhttp://stevesouders.com/examples/css-bottom.phpsolution: put stylesheets in HEAD (per spec)avoids Flash of Unstyled Contentuse LINK (not @import)

Slowest is Fastest

Rule 6: Move scripts to the bottom scripts block parallel downloads across all hostnamesscripts block rendering of everything below them in the pageIE and FFhttp://stevesouders.com/examples/js-middle.php

Rule 6: Move scripts to the bottom script defer attribute is not a solutionblocks rendering and downloads in FFslight blocking in IEsolution: move them as low in the page as possible

Rule 7: Avoid CSS expressions used to set CSS properties dynamically in IEwidth: expression( document.body.clientWidth < 600 ? 600px : auto );problem: expressions execute many timesmouse move, key press, resize, scroll, etc.

http://stevesouders.com/examples/expression-counter.php

One-Time Expressionsexpression overwrites itself

P { background-color: expression(altBgcolor(this));}

function altBgcolor(elem) { elem.style.backgroundColor = (new Date()).getHours()%2 ? "#F08A00" : "#B8D4FF";}

Event Handlerstie behavior to (fewer) specific eventswindow.onresize = setMinWidth;function setMinWidth() { var aElements = document.getElementsByTagName("p"); for ( var i = 0; i < aElements.length; i++ ) { aElements[i].runtimeStyle.width = ( document.body.clientWidth

Rule 8: Make JS and CSS external inline: HTML document is biggerexternal: more HTTP requests, but cachedvariablespage views per user (per session)empty vs. full cache statscomponent re-useexternal is typically betterhome pages may be an exception

Post-Onload Downloadinline in front pagedownload external files after onloadwindow.onload = downloadComponents;function downloadComponents() { var elem = document.createElement("script"); elem.src = "http://.../file1.js"; document.body.appendChild(elem); ...}speeds up secondary pages

Dynamic Inliningstart with post-onload downloadset cookie after components downloadedserver-side: if cookie, use externalelse, do inline with post-onload downloadcookie expiration date is keyspeeds up all pages

Rule 9: Reduce DNS lookups typically 20-120 msblock parallel downloadsOS and browser both have DNS caches

TTL (Time To Live)

TTL how long record can be cachedbrowser settings override TTL

Browser DNS CacheIEDnsCacheTimeout: 30 minutesKeepAliveTimeout: 1 minuteServerInfoTimeout: 2 minutesFirefoxnetwork.dnsCacheExpiration: 1 minutenetwork.dnsCacheEntries: 20network.http.keep-alive.timeout: 5 minutesFasterfox: 1 hour, 512 entries, 30 seconds

Reducing DNS Lookupsfewer hostnames 2-4keep-alive

Rule 10: Minify JavaScript

minify inline scripts, too

Minify vs. Obfuscate

minify it's saferhttp://crockford.com/javascript/jsminhttp://dojotoolkit.org/docs/shrinksafe

Rule 11: Avoid redirects3xx status codes mostly 301 and 302HTTP/1.1 301 Moved PermanentlyLocation: http://stevesouders.com/newuriadd Expires headers to cache redirectsworst form of blockinghttp://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html

Redirects

Avoid Redirectsmissing trailing slashhttp://astrology.yahoo.com/astrologyuse Alias or DirectorySlashmod_rewriteCNAMEslog referer track internal linksoutbound links harderbeacons beware of race conditionXHR bail at readyState 2

Rule 12: Remove duplicate scriptshurts performanceextra HTTP requests (IE only)extra executionsatypical?2 of 10 top sites contain duplicate scriptsteam size, # of scripts

Script Insertion Functions

Rule 13: Turn off ETagsunique identifier returned in responseETag: "c8897e-aee-4165acf0"Last-Modified: Thu, 07 Oct 2004 20:54:08 GMTused in conditional GET requestsIf-None-Match: "c8897e-aee-4165acf0"If-Modified-Since: Thu, 07 Oct 2004 20:54:08 GMTif ETag doesn't match, can't send 304

The Problem with ETagsETag for a single entity is always different across serversETag formatApache: inode-size-timestampIIS: Filetimestamp:ChangeNumberSites with >1 server return too few 304s(n-1)/nRemove themApache: FileETag noneIIS: http://support.microsoft.com/kb/922703/

Rule 14: Make AJAX cacheable and smallXHR, JSON, iframe, dynamic scripts can still be cached, minified, and gzippeda personalized response should still be cacheable by that person

AJAX Example: Yahoo! Mail Betaaddress book XML request GET /yab/[...]&r=0.5289571053069156 HTTP/1.1 Host: us.xxx.mail.yahoo.com HTTP/1.1 200 OK Date: Thu, 12 Apr 2007 19:39:09 GMT Cache-Control: private,max-age=0 Last-Modified: Sat, 31 Mar 2007 01:17:17 GMT Content-Type: text/xml; charset=utf-8 Content-Encoding: gzipaddress book changes infrequentlycache it; add last-modified-time in URL

Live Analysis

IBM Page Detailerpacket snifferWindows onlyIE, FF, any .exec:\windows\wd_WS2s.iniExecutable=(NETSCAPE.EXE),(NETSCP6.EXE),(firefox.exe)free trial, $300 licensehttp://alphaworks.ibm.com/tech/pagedetailer

http://alphaworks.ibm.com/tech/pagedetailer

Fasterfoxmeasures load time of pagesalters config settings for faster loadingFirefox extensionfreehttp://fasterfox.mozdev.org/

LiveHTTPHeadersview HTTP headersFirefox extensionfreehttp://livehttpheaders.mozdev.org/

Firebugweb development evolvedinspect and edit HTMLtweak and visualize CSSdebug and profile JavaScriptmonitor network activity (caveat)Firefox extensionfreehttp://getfirebug.com/

http://getfirebug.com/

YSlowperformance lint toolgrades web pages for each ruleFirefox extensionYahoo! internal tool

Conclusion

Takeawaysfocus on the front-endharvest the low-hanging fruityou do control user response timesLOFNO be an advocate for your users

Linksbook: http://www.oreilly.com/catalog/9780596514211/ examples: http://stevesouders.com/examples/ image maps: http://www.w3.org/TR/html401/struct/objects.html#h-13.6 CSS sprites: http://alistapart.com/articles/sprites inline images: http://tools.ietf.org/html/rfc2397 jsmin: http://crockford.com/javascript/jsmin dojo compressor: http://dojotoolkit.org/docs/shrinksafe HTTP status codes: http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html IBM Page Detailer: http://alphaworks.ibm.com/tech/pagedetailer Fasterfox: http://fasterfox.mozdev.org/LiveHTTPHeaders: http://livehttpheaders.mozdev.org/Firebug: http://getfirebug.com/ YUIBlog: http://yuiblog.com/blog/2006/11/28/performance-research-part-1/ http://yuiblog.com/blog/2007/01/04/performance-research-part-2/ http://yuiblog.com/blog/2007/03/01/performance-research-part-3/ http://yuiblog.com/blog/2007/04/11/performance-research-part-4/YDN: http://developer.yahoo.net/blog/archives/2007/03/high_performanc.html http://developer.yahoo.net/blog/archives/2007/04/rule_1_make_few.html

high performance web sites 14 rules for faster-loading pages

Documents