SlideShare a Scribd company logo
• Philip Tellis

•                        .com
• philip@lognormal.com
• @bluesmoon
• geek paranoid speedfreak
• http://bluesmoon.info/




     NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   1
I <3 JavaScript




NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   2
So much that I wore Mustache socks to my wedding




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   3
I’m a Speedfreak (the network kind)




NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   4
We measure real user website performance




NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   5
This talk is (mostly) about how we abuse JavaScript to do it




    NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   6
Abusing JavaScript to Measure Web
            Performance

       Philip Tellis / philip@lognormal.com


           NY #WebPerf Meetup / 2012-09-13




NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   7
First, a note about the code




                      Note that in the code that follows,

                                     + new Date

                                    is equivalent to

                           new Date().getTime()




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   8
1
                           Latency



NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   9
Arrange the following in order of increasing signal latency over
a fixed distance:
  • Electricity through copper
  • Smoke Signals
  • Light through Fibre
  • Pulsar




       NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   10
1 Blinking Lights


        It takes about 16ms for light to get from SF to NYC
                       (24ms through fibre)

                                          ...




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   11
1 Blinking Lights


        It takes about 16ms for light to get from SF to NYC
                       (24ms through fibre)

           Though it takes about 100ms to ping... why?




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   11
1 HTTP




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   12
So to measure latency, we need to send 1 packet each way, and
                           time it




     NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   13
1   Network latency in JavaScript




      var ts, rtt, img = new Image;
      img.onload=function() { rtt=(+new Date - ts) };
      ts = +new Date;
      img.src="/1x1.gif";




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   14
1 Notes


    • 1x1 gif is 35 bytes
    • including HTTP headers, is smaller than a TCP packet
    • Fires onload on all browsers
    • 0 byte image fires onerror
    • which is indistinguishable from network error




          NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   15
2
                TCP handshake



NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   16
2 ACK-ACK-ACK




      NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   17
2 Connection: keep-alive




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   18
2   Network latency & TCP handshake in JavaScript
    var t=[], tcp, rtt;
    var ld = function() {
        t.push(+new Date);
        if(t.length > 2) // run 2 times
          done();
        else {
          var img = new Image;
          img.onload = ld;
          img.src="/1x1.gif?" + Math.random()
                               + ’=’ + new Date;
        }
    };
    var done = function() {
       rtt=t[2]-t[1];
       tcp=t[1]-t[0]-rtt;
    };
    ld();
         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   19
Notice that we’ve ignored DNS lookup time here... how would
                      you measure it?




     NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   20
Network Throughput
                            3
NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   21
3 Measuring Network Throughput



                                   data_length
                                  download_time




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   22
Should you fly a 747 or a 737?



       • A 747 seats 400+ passengers
       • A 737 seats about 150
       • Both take about the same time to fly from SFO to BOS
       • A 747 takes longer to load and unload than 3 737s in
           parallel




   The best selling aircraft to date is the 737




                NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   23
3   Network Throughput in JavaScript




    // Assume global object
    // image={ url: ..., size: ... }
    var ts, rtt, bw, img = new Image;
    img.onload=function() {
       rtt=(+new Date - ts);
       bw = image.size*1000/rtt;     // rtt is in ms
    };
    ts = +new Date;
    img.src=image.url;




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   24
3 Measuring Network Throughput



       If it were that simple, I wouldn’t be doing this talk.




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   25
3 TCP Slow Start




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   26
3 Measuring Network Throughput



   So to make the best use of bandwidth, we need resources that fit
                         in a TCP window




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   27
3 There is no single size that will tax all available networks




   http://www.yuiblog.com/blog/2010/04/08/analyzing-bandwidth-and-latency/




               NY #WebPerf Meetup / 2012-09-13        Abusing JavaScript to Measure Web Performance   28
3   Network Throughput in JavaScript – Take 2

    // image object is now an array of multiple images
    var i=0;
    var ld = function() {
       if(i>0)
          image[i-1].end = +new Date;
       if(i >= image.length)
          done();
       else {
          var img = new Image;
          img.onload = ld;
          image[i].start = +new Date;
          img.src=image[i].url;
       }
       i++;
    };

         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   29
3 Measuring Network Throughput



        Slow network connection, meet really huge image




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   30
3   Network Throughput in JavaScript – Take 3




       var img = new Image;
       img.onload = ld;
       image[i].start = +new Date;
       image[i].timer =
             setTimeout(function() {
                            image[i].expired=true
                        },
                        image[i].timeout);
       img.src=image[i].url;




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   31
3   Network Throughput in JavaScript – Take 3




    if(i>0) {
       image[i-1].end = +new Date;
       clearTimeout(image[i-1].timer);
    }




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   32
3   Network Throughput in JavaScript – Take 3




    if(i >= image.length
          || (i > 0 && image[i-1].expired)) {
       done();
    }




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   33
3 Measuring Network Throughput



                                  Are we done yet?




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   34
3 Measuring Network Throughput



                                  Are we done yet?
                                      sure...




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   34
3 Measuring Network Throughput



    Except network throughput is different every time you test it




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   35
Statistics to the Rescue




flickr/sophistechate/4264466015/
                    NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   36
3 Measuring Network Throughput



            The code for that is NOT gonna fit on a slide




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   37
But this is sort of what we see world-wide




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   38
And it’s different for different countries




   This is India




          NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   39
4     DNS



NY #WebPerf Meetup / 2012-09-13    Abusing JavaScript to Measure Web Performance   40
4 Measuring DNS



                time_with_dns − time_without_dns




       NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   41
4   Measuring DNS in JavaScript
    var t=[], dns, ip, hosts=[’http://hostname.com/’,
                               ’http://ip.ad.dr.ess/’];
    var ld = function() {
        t.push(+new Date);
        if(t.length > hosts.length)
          done();
        else {
          var img = new Image;
          img.onload = ld;
          img.src=hosts[t.length-1] + "/1x1.gif";
        }
    };
    var done = function() {
       ip=t[2]-t[1];
       dns=t[1]-t[0]-ip;
    };
    ld();
         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   42
4 Measuring DNS


    • What if the IP changes?
    • What if DNS is cached?
    • What if you map DNS based on geo location?




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   43
4   Wildcard DNS Entries




                             *.foo.com → IP address




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   44
4   Measuring DNS in JavaScript – take 2

    var base_url="http://*.foo.com/",
        timers = {}, gen_url="";

    function start() {
      var random = Math.random().toString(36),
          cache_bust = Math.random(),
          img = new Image();

        gen_url = base_url.replace(/*/, random);

        img.onload = A_loaded;

        timers.start = +new Date;
        img.src = gen_url + "image-l.gif?t=" + cache_bust;
    }

           NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   45
4   Measuring DNS in JavaScript – take 2



    function A_loaded() {
      var cache_bust = Math.random(),
          img = new Image();

        img.onload = B_loaded;

        timers.a_loaded = +new Date;
        img.src = gen_url + "image-l.gif?t=" + cache_bust;
    }


    I’ll let you figure out B_loaded



           NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   46
4 Measuring DNS



                Full code in boomerang’s DNS plugin




       NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   47
5     IPv6



NY #WebPerf Meetup / 2012-09-13    Abusing JavaScript to Measure Web Performance   48
5 Measuring IPv6 support and latency


    1   Try to load image from IPv6 host
          • If timeout or error, then no IPv6 support
          • If successful, then calculate latency and proceed
    2   Try to load image from hostname that resolves only to IPv6
        host
          • If timeout or error, then DNS server doesn’t support IPv6
          • If successful, calculate latency




          NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   49
5 Measuring IPv6 support and latency



                  Full code in boomerang’s IPv6 plugin
                       Note, only run this if you know IPv6 is supported by the client




        NY #WebPerf Meetup / 2012-09-13           Abusing JavaScript to Measure Web Performance   50
6
                       Other Stuff



NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   51
6   Other Stuff We Measure




     • NavTiming – navtiming.js
     • navigation.connection.type – mobile.js
     • window.performance.memory – memory.js –
       Chrome 22 reporting now.
     • Number of DOM nodes and byte size of HTML –
       memory.js




         NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   52
And we try to do it fast




NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   53
–
                            .done()



NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   54
Code/References




    • http://lognormal.github.com/boomerang/doc/
      (BSD Licensed)
    • www.lognormal.com




        NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   55
• Philip Tellis

•                        .com
• philip@lognormal.com
• @bluesmoon
• geek paranoid speedfreak
• http://bluesmoon.info/




     NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   56
Thank you
                   Ask me about a discount code




NY #WebPerf Meetup / 2012-09-13   Abusing JavaScript to Measure Web Performance   57

More Related Content

Similar to Abusing JavaScript to measure Web Performance, or, "how does boomerang work?" (20)

PDF
Abusing JavaScript to Measure Web Performance
Philip Tellis
 
PDF
Analysing network characteristics with JavaScript
Philip Tellis
 
PDF
Messing with JavaScript and the DOM to measure network characteristics
Philip Tellis
 
PPTX
Performance on a budget
Dimitry Ushakov
 
PPTX
Using Modern Browser APIs to Improve the Performance of Your Web Applications
Nicholas Jansma
 
PPTX
Measuring performance - Velocity 2016 Training
Patrick Meenan
 
PPTX
Measuring web performance
Patrick Meenan
 
PDF
The Statistics of Web Performance Analysis
Philip Tellis
 
PDF
Velocity_Conference
Anne Cypcar
 
PDF
Client-Side Performance Testing
Anand Bagmar
 
PDF
Measuring the End User
APNIC
 
PPTX
Browser Based Performance Testing and Tuning
Bala Murali Krishna Kanchukambala
 
PDF
Measuring CDN performance and why you're doing it wrong
Fastly
 
PDF
Ez performance measurement
Gaetano Giunta
 
PPTX
Why is this ASP.NET web app running slowly?
Mark Friedman
 
PDF
Ad109 - XPages Performance and Scalability
ddrschiw
 
PPTX
Monitoring web application response times^lj a hybrid approach for windows
Mark Friedman
 
PDF
GDD Japan 2009 - Designing OpenSocial Apps For Speed and Scale
Patrick Chanezon
 
PPTX
Web Performance BootCamp 2013
Daniel Austin
 
KEY
improving the performance of Rails web Applications
John McCaffrey
 
Abusing JavaScript to Measure Web Performance
Philip Tellis
 
Analysing network characteristics with JavaScript
Philip Tellis
 
Messing with JavaScript and the DOM to measure network characteristics
Philip Tellis
 
Performance on a budget
Dimitry Ushakov
 
Using Modern Browser APIs to Improve the Performance of Your Web Applications
Nicholas Jansma
 
Measuring performance - Velocity 2016 Training
Patrick Meenan
 
Measuring web performance
Patrick Meenan
 
The Statistics of Web Performance Analysis
Philip Tellis
 
Velocity_Conference
Anne Cypcar
 
Client-Side Performance Testing
Anand Bagmar
 
Measuring the End User
APNIC
 
Browser Based Performance Testing and Tuning
Bala Murali Krishna Kanchukambala
 
Measuring CDN performance and why you're doing it wrong
Fastly
 
Ez performance measurement
Gaetano Giunta
 
Why is this ASP.NET web app running slowly?
Mark Friedman
 
Ad109 - XPages Performance and Scalability
ddrschiw
 
Monitoring web application response times^lj a hybrid approach for windows
Mark Friedman
 
GDD Japan 2009 - Designing OpenSocial Apps For Speed and Scale
Patrick Chanezon
 
Web Performance BootCamp 2013
Daniel Austin
 
improving the performance of Rails web Applications
John McCaffrey
 

More from Philip Tellis (20)

PDF
Improving D3 Performance with CANVAS and other Hacks
Philip Tellis
 
PDF
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
PDF
Frontend Performance: De débutant à Expert à Fou Furieux
Philip Tellis
 
PDF
Frontend Performance: Expert to Crazy Person
Philip Tellis
 
PDF
Beyond Page Level Metrics
Philip Tellis
 
PDF
Frontend Performance: Beginner to Expert to Crazy Person (San Diego Web Perf ...
Philip Tellis
 
PDF
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
PDF
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
PDF
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
PDF
mmm... beacons
Philip Tellis
 
PDF
RUM Distillation 101 -- Part I
Philip Tellis
 
PDF
Improving 3rd Party Script Performance With IFrames
Philip Tellis
 
PDF
Extending Boomerang
Philip Tellis
 
PDF
Rum for Breakfast
Philip Tellis
 
PDF
A Node.JS bag of goodies for analyzing Web Traffic
Philip Tellis
 
PDF
Input sanitization
Philip Tellis
 
PDF
Boomerang: How fast do users think your site is?
Philip Tellis
 
PDF
Boomerang at FOSS.IN/2010
Philip Tellis
 
PDF
Measuring the web with Boomerang (YUIConf 2010)
Philip Tellis
 
PDF
Boomerang at the Boston Web Performance meetup
Philip Tellis
 
Improving D3 Performance with CANVAS and other Hacks
Philip Tellis
 
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
Frontend Performance: De débutant à Expert à Fou Furieux
Philip Tellis
 
Frontend Performance: Expert to Crazy Person
Philip Tellis
 
Beyond Page Level Metrics
Philip Tellis
 
Frontend Performance: Beginner to Expert to Crazy Person (San Diego Web Perf ...
Philip Tellis
 
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
Frontend Performance: Beginner to Expert to Crazy Person
Philip Tellis
 
mmm... beacons
Philip Tellis
 
RUM Distillation 101 -- Part I
Philip Tellis
 
Improving 3rd Party Script Performance With IFrames
Philip Tellis
 
Extending Boomerang
Philip Tellis
 
Rum for Breakfast
Philip Tellis
 
A Node.JS bag of goodies for analyzing Web Traffic
Philip Tellis
 
Input sanitization
Philip Tellis
 
Boomerang: How fast do users think your site is?
Philip Tellis
 
Boomerang at FOSS.IN/2010
Philip Tellis
 
Measuring the web with Boomerang (YUIConf 2010)
Philip Tellis
 
Boomerang at the Boston Web Performance meetup
Philip Tellis
 
Ad

Recently uploaded (20)

PDF
NewMind AI - Journal 100 Insights After The 100th Issue
NewMind AI
 
PDF
Meetup Kickoff & Welcome - Rohit Yadav, CSIUG Chairman
ShapeBlue
 
PPTX
Building and Operating a Private Cloud with CloudStack and LINBIT CloudStack ...
ShapeBlue
 
PPTX
Building a Production-Ready Barts Health Secure Data Environment Tooling, Acc...
Barts Health
 
PDF
Complete JavaScript Notes: From Basics to Advanced Concepts.pdf
haydendavispro
 
PDF
Wojciech Ciemski for Top Cyber News MAGAZINE. June 2025
Dr. Ludmila Morozova-Buss
 
PPT
Interview paper part 3, It is based on Interview Prep
SoumyadeepGhosh39
 
PPTX
MSP360 Backup Scheduling and Retention Best Practices.pptx
MSP360
 
PPTX
Webinar: Introduction to LF Energy EVerest
DanBrown980551
 
PDF
Apache CloudStack 201: Let's Design & Build an IaaS Cloud
ShapeBlue
 
PDF
July Patch Tuesday
Ivanti
 
PPTX
WooCommerce Workshop: Bring Your Laptop
Laura Hartwig
 
PDF
Why Orbit Edge Tech is a Top Next JS Development Company in 2025
mahendraalaska08
 
PDF
LLMs.txt: Easily Control How AI Crawls Your Site
Keploy
 
PDF
DevBcn - Building 10x Organizations Using Modern Productivity Metrics
Justin Reock
 
PDF
Persuasive AI: risks and opportunities in the age of digital debate
Speck&Tech
 
PDF
Human-centred design in online workplace learning and relationship to engagem...
Tracy Tang
 
PPTX
Building Search Using OpenSearch: Limitations and Workarounds
Sease
 
PDF
Building Real-Time Digital Twins with IBM Maximo & ArcGIS Indoors
Safe Software
 
PDF
CIFDAQ Token Spotlight for 9th July 2025
CIFDAQ
 
NewMind AI - Journal 100 Insights After The 100th Issue
NewMind AI
 
Meetup Kickoff & Welcome - Rohit Yadav, CSIUG Chairman
ShapeBlue
 
Building and Operating a Private Cloud with CloudStack and LINBIT CloudStack ...
ShapeBlue
 
Building a Production-Ready Barts Health Secure Data Environment Tooling, Acc...
Barts Health
 
Complete JavaScript Notes: From Basics to Advanced Concepts.pdf
haydendavispro
 
Wojciech Ciemski for Top Cyber News MAGAZINE. June 2025
Dr. Ludmila Morozova-Buss
 
Interview paper part 3, It is based on Interview Prep
SoumyadeepGhosh39
 
MSP360 Backup Scheduling and Retention Best Practices.pptx
MSP360
 
Webinar: Introduction to LF Energy EVerest
DanBrown980551
 
Apache CloudStack 201: Let's Design & Build an IaaS Cloud
ShapeBlue
 
July Patch Tuesday
Ivanti
 
WooCommerce Workshop: Bring Your Laptop
Laura Hartwig
 
Why Orbit Edge Tech is a Top Next JS Development Company in 2025
mahendraalaska08
 
LLMs.txt: Easily Control How AI Crawls Your Site
Keploy
 
DevBcn - Building 10x Organizations Using Modern Productivity Metrics
Justin Reock
 
Persuasive AI: risks and opportunities in the age of digital debate
Speck&Tech
 
Human-centred design in online workplace learning and relationship to engagem...
Tracy Tang
 
Building Search Using OpenSearch: Limitations and Workarounds
Sease
 
Building Real-Time Digital Twins with IBM Maximo & ArcGIS Indoors
Safe Software
 
CIFDAQ Token Spotlight for 9th July 2025
CIFDAQ
 
Ad

Abusing JavaScript to measure Web Performance, or, "how does boomerang work?"

  • 1. • Philip Tellis • .com • [email protected] • @bluesmoon • geek paranoid speedfreak • http://bluesmoon.info/ NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 1
  • 2. I <3 JavaScript NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 2
  • 3. So much that I wore Mustache socks to my wedding NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 3
  • 4. I’m a Speedfreak (the network kind) NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 4
  • 5. We measure real user website performance NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 5
  • 6. This talk is (mostly) about how we abuse JavaScript to do it NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 6
  • 7. Abusing JavaScript to Measure Web Performance Philip Tellis / [email protected] NY #WebPerf Meetup / 2012-09-13 NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 7
  • 8. First, a note about the code Note that in the code that follows, + new Date is equivalent to new Date().getTime() NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 8
  • 9. 1 Latency NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 9
  • 10. Arrange the following in order of increasing signal latency over a fixed distance: • Electricity through copper • Smoke Signals • Light through Fibre • Pulsar NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 10
  • 11. 1 Blinking Lights It takes about 16ms for light to get from SF to NYC (24ms through fibre) ... NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 11
  • 12. 1 Blinking Lights It takes about 16ms for light to get from SF to NYC (24ms through fibre) Though it takes about 100ms to ping... why? NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 11
  • 13. 1 HTTP NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 12
  • 14. So to measure latency, we need to send 1 packet each way, and time it NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 13
  • 15. 1 Network latency in JavaScript var ts, rtt, img = new Image; img.onload=function() { rtt=(+new Date - ts) }; ts = +new Date; img.src="/1x1.gif"; NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 14
  • 16. 1 Notes • 1x1 gif is 35 bytes • including HTTP headers, is smaller than a TCP packet • Fires onload on all browsers • 0 byte image fires onerror • which is indistinguishable from network error NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 15
  • 17. 2 TCP handshake NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 16
  • 18. 2 ACK-ACK-ACK NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 17
  • 19. 2 Connection: keep-alive NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 18
  • 20. 2 Network latency & TCP handshake in JavaScript var t=[], tcp, rtt; var ld = function() { t.push(+new Date); if(t.length > 2) // run 2 times done(); else { var img = new Image; img.onload = ld; img.src="/1x1.gif?" + Math.random() + ’=’ + new Date; } }; var done = function() { rtt=t[2]-t[1]; tcp=t[1]-t[0]-rtt; }; ld(); NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 19
  • 21. Notice that we’ve ignored DNS lookup time here... how would you measure it? NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 20
  • 22. Network Throughput 3 NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 21
  • 23. 3 Measuring Network Throughput data_length download_time NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 22
  • 24. Should you fly a 747 or a 737? • A 747 seats 400+ passengers • A 737 seats about 150 • Both take about the same time to fly from SFO to BOS • A 747 takes longer to load and unload than 3 737s in parallel The best selling aircraft to date is the 737 NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 23
  • 25. 3 Network Throughput in JavaScript // Assume global object // image={ url: ..., size: ... } var ts, rtt, bw, img = new Image; img.onload=function() { rtt=(+new Date - ts); bw = image.size*1000/rtt; // rtt is in ms }; ts = +new Date; img.src=image.url; NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 24
  • 26. 3 Measuring Network Throughput If it were that simple, I wouldn’t be doing this talk. NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 25
  • 27. 3 TCP Slow Start NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 26
  • 28. 3 Measuring Network Throughput So to make the best use of bandwidth, we need resources that fit in a TCP window NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 27
  • 29. 3 There is no single size that will tax all available networks http://www.yuiblog.com/blog/2010/04/08/analyzing-bandwidth-and-latency/ NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 28
  • 30. 3 Network Throughput in JavaScript – Take 2 // image object is now an array of multiple images var i=0; var ld = function() { if(i>0) image[i-1].end = +new Date; if(i >= image.length) done(); else { var img = new Image; img.onload = ld; image[i].start = +new Date; img.src=image[i].url; } i++; }; NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 29
  • 31. 3 Measuring Network Throughput Slow network connection, meet really huge image NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 30
  • 32. 3 Network Throughput in JavaScript – Take 3 var img = new Image; img.onload = ld; image[i].start = +new Date; image[i].timer = setTimeout(function() { image[i].expired=true }, image[i].timeout); img.src=image[i].url; NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 31
  • 33. 3 Network Throughput in JavaScript – Take 3 if(i>0) { image[i-1].end = +new Date; clearTimeout(image[i-1].timer); } NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 32
  • 34. 3 Network Throughput in JavaScript – Take 3 if(i >= image.length || (i > 0 && image[i-1].expired)) { done(); } NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 33
  • 35. 3 Measuring Network Throughput Are we done yet? NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 34
  • 36. 3 Measuring Network Throughput Are we done yet? sure... NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 34
  • 37. 3 Measuring Network Throughput Except network throughput is different every time you test it NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 35
  • 38. Statistics to the Rescue flickr/sophistechate/4264466015/ NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 36
  • 39. 3 Measuring Network Throughput The code for that is NOT gonna fit on a slide NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 37
  • 40. But this is sort of what we see world-wide NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 38
  • 41. And it’s different for different countries This is India NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 39
  • 42. 4 DNS NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 40
  • 43. 4 Measuring DNS time_with_dns − time_without_dns NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 41
  • 44. 4 Measuring DNS in JavaScript var t=[], dns, ip, hosts=[’http://hostname.com/’, ’http://ip.ad.dr.ess/’]; var ld = function() { t.push(+new Date); if(t.length > hosts.length) done(); else { var img = new Image; img.onload = ld; img.src=hosts[t.length-1] + "/1x1.gif"; } }; var done = function() { ip=t[2]-t[1]; dns=t[1]-t[0]-ip; }; ld(); NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 42
  • 45. 4 Measuring DNS • What if the IP changes? • What if DNS is cached? • What if you map DNS based on geo location? NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 43
  • 46. 4 Wildcard DNS Entries *.foo.com → IP address NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 44
  • 47. 4 Measuring DNS in JavaScript – take 2 var base_url="http://*.foo.com/", timers = {}, gen_url=""; function start() { var random = Math.random().toString(36), cache_bust = Math.random(), img = new Image(); gen_url = base_url.replace(/*/, random); img.onload = A_loaded; timers.start = +new Date; img.src = gen_url + "image-l.gif?t=" + cache_bust; } NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 45
  • 48. 4 Measuring DNS in JavaScript – take 2 function A_loaded() { var cache_bust = Math.random(), img = new Image(); img.onload = B_loaded; timers.a_loaded = +new Date; img.src = gen_url + "image-l.gif?t=" + cache_bust; } I’ll let you figure out B_loaded NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 46
  • 49. 4 Measuring DNS Full code in boomerang’s DNS plugin NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 47
  • 50. 5 IPv6 NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 48
  • 51. 5 Measuring IPv6 support and latency 1 Try to load image from IPv6 host • If timeout or error, then no IPv6 support • If successful, then calculate latency and proceed 2 Try to load image from hostname that resolves only to IPv6 host • If timeout or error, then DNS server doesn’t support IPv6 • If successful, calculate latency NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 49
  • 52. 5 Measuring IPv6 support and latency Full code in boomerang’s IPv6 plugin Note, only run this if you know IPv6 is supported by the client NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 50
  • 53. 6 Other Stuff NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 51
  • 54. 6 Other Stuff We Measure • NavTiming – navtiming.js • navigation.connection.type – mobile.js • window.performance.memory – memory.js – Chrome 22 reporting now. • Number of DOM nodes and byte size of HTML – memory.js NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 52
  • 55. And we try to do it fast NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 53
  • 56. .done() NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 54
  • 57. Code/References • http://lognormal.github.com/boomerang/doc/ (BSD Licensed) • www.lognormal.com NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 55
  • 58. • Philip Tellis • .com • [email protected] • @bluesmoon • geek paranoid speedfreak • http://bluesmoon.info/ NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 56
  • 59. Thank you Ask me about a discount code NY #WebPerf Meetup / 2012-09-13 Abusing JavaScript to Measure Web Performance 57