Web Performance BootCamp 2013

104
Web Performance Boot Camp Daniel Austin PayPal, Inc. HTML5 DevConf Oct. 24, 2013 V 2.2

description

These are the slides I used for my 1-day Performance Workshop at HTML5 DevConf this year.

Transcript of Web Performance BootCamp 2013

Page 1: Web Performance BootCamp 2013

Web Performance Boot Camp

Daniel Austin

PayPal, Inc.

HTML5 DevConf

Oct. 24, 2013

V 2.2

Page 2: Web Performance BootCamp 2013

Overture: Goals of the Class

• Provide a basic understanding of Web performance for Architects, Developers, Designers, and Engineers

• Empower YOU to identify and resolve performance problems and make your pages and applications faster!!!

• Demonstrate and explain how to use common tools and techniques used in our industry to solve performance problems

Page 3: Web Performance BootCamp 2013

Scope of Web Performance

Anything that uses HTTP

Always From the End User’s Point of View

Web Request/Response Only!

Page 4: Web Performance BootCamp 2013

Current State of the Art

• Web performance is both an Art and a Science (but it’s not yet Engineering)

• Multiple tools and methodologies, large ad hoc, contend in the marketplace (but little of it is well-thought out or based on scientific reasoning).

• Things are getting better – W3C involvement and competitive pressures, as well as better infrastructure and the influx of new users in Asia is driving more attention to performance. There’s hope.

Page 6: Web Performance BootCamp 2013

Tools Used in This Class

• Excel (or similar spreadsheet program)

• Online Testing Tools – webspeedtest.org

• Desktop Testing Tools – your browser, Firebug, Visual Round Trip Analyzer (VRTA), netmon, dig, ping others

• Optional: R (and Rstudio), Mathematica, SPSS

Page 7: Web Performance BootCamp 2013

Class Structure

Schedule

• Start: 9:00 AM• Break: 10:30-1:45• Lunch: 12:30-1:15• Break: 2:30-2:45• End: 4:00 PM

Agenda

• Section I – What Is Performance?• Section I – Performance Basics• Section III – The MPPC Model• Section IV – Tools & Testing• Section V – HTML5 & Peformance• Section VI – Mobile Devices

Page 8: Web Performance BootCamp 2013

Section I What is Performance?

Page 9: Web Performance BootCamp 2013

What Problem Are We Trying To Solve?

• World-class response times compared to our competitors

• Reliable, predictable performance for users worldwide

• Efficient use of resources: cost scales linearly with traffic

• Delighted users!

Page 10: Web Performance BootCamp 2013

Impact of Scalability on Business

• Google– 500 ms reduces traffic to sites by 20%

• Yahoo!– 400 ms reduces traffic y 5-9%

• Amazon– 100 ms reduces revenue by 1%

• Compuware – 1 sec delay reduces conversion by 7%

10

Page 11: Web Performance BootCamp 2013

Who needs ‘Performance’ measurements

anyway??

Who Needs Performance Anyway?

Page 12: Web Performance BootCamp 2013

A More Rational Approach?

Page 13: Web Performance BootCamp 2013

Systemic Qualities In a Nutshell

“Anything you can say about a black box – from the outside”

• Systemic qualities are the “ilities” – physical features of the system such as capacity, performance, and scalability

• The SQs correspond to different groups of stakeholders: users, developers, operators, organizations

• SQs are the best measure of the quality of the user’s experience of the system, regardless of the feature set

Page 14: Web Performance BootCamp 2013

The Four Classes of Systemic Qualities

Manifest Qualities - What the users see

• Usability, Performance, Reliability, Availability, Accessibility

Operational Qualities - What the system operators see

• Throughput, Manageability, Security, Serviceability

Developmental Qualities - What developers see

• Buildability, Budgetability, Planability

Evolutionary Qualities - How the system changes over time

• Scalability, Maintainability, Extensibility, Reusability, Portability

Page 15: Web Performance BootCamp 2013

The Manifest Systemic Qualities

• Usability reflects the ease with which users can accomplish their goals

• Performance reflects how much little time users must wait for actions to complete

• Reliability measures how often the system fails

• Availability measures uptime vs. downtime

• Accessibility measures the systems ability to serve users regardless of location or physical condition (including I18N and L10N)

Page 16: Web Performance BootCamp 2013

Performance is a Balancing Act

Performance isn’t everything; sometimes we’re called on to make choices about which systemic qualities have priority over others. Security v. performance is a common tradeoff – what would you choose?

Page 17: Web Performance BootCamp 2013

What Is Performance?

PERFORMANCE IS RESPONSE TIME

PERFORMANCE IS RESPONSE TIME

PERFORMANCE IS RESPONSE TIME

PERFORMANCE IS RESPONSE TIME

PERFORMANCE IS RESPONSE TIME

Page 18: Web Performance BootCamp 2013

Section II Performance Basics

Page 19: Web Performance BootCamp 2013

Scales of Measure

Page 20: Web Performance BootCamp 2013

Statistics 101

Page 21: Web Performance BootCamp 2013

Comparison of Mean, Median, and Mode

Comparison of mean, median and mode of two log-normal distributions with different skewness.

Page 22: Web Performance BootCamp 2013

Outliers, or Why We Use the Median

• A: skewed to the left• B: skewed to the right• C: symmetrical

Page 23: Web Performance BootCamp 2013

Statistical Distributions

• Discrete or continuous?• Mean, median, sigma,

95%?• Is it reasonable?

|mean – median| <= sigma• Does it correlate?

Page 24: Web Performance BootCamp 2013

Understanding the Margin of Error

• Margin of error at 99% confidence = 1.29/sqrt(n)

• Margin of error at 95% confidence = 0.98/sqrt(n)

• Margin of error at 90% confidence = 0.82/sqrt(n)

(where n is the number of sample data points)

The margin of error is a measure of how close the results are likely to be.

Page 25: Web Performance BootCamp 2013

5 Number Reports

A simple way of summarizing a sample

• Shape of the distribution• Extreme values• Variance• Skewness

This is how you get a sense of the data…

Median

1st quartile 3rd quartile

Minimum Maximum

Page 26: Web Performance BootCamp 2013

5 Number Reports in Excel

Excel functions:

Min = MIN(Data Range)

Q1 = QUARTILE(Data Range, 1)

Q2 = QUARTILE(Data Range, 2)

Q3 = QUARTILE(Data Range, 3)

Max = MAX(Data Range)

Page 27: Web Performance BootCamp 2013

5 Number Reports in R

Let’s make a 5 number report in R:

http://www.r-project.org/

Page 28: Web Performance BootCamp 2013

Operational Research

• Developed during WWII for managing armies and supply chains

• A set of rules or ‘laws’ that describe the operational aspects of a system.

• Useful for understanding the performance of any system

• Utilization Law

• Forced Flow law

• Little’s Law

• Response Time Law

Page 29: Web Performance BootCamp 2013

Resources and Queues

Si: Service time

Ri: Queue Residence time

i: Queue length

In general, systems consist of many combined queues and resources

ResourceQueue

SiRi

Page 30: Web Performance BootCamp 2013

The Utilization Law

• The utilization (Ui) of resource i is the fraction of time that the resource is busy.

• Xi: average throughput of queue i, i.e. average number of requests that complete from queue i per unit of time

• Si: average service time of a request at queue i per visit to the resource

Ui = Xi * Si

Page 31: Web Performance BootCamp 2013

Interactive Response time law

IMAGE COURTESY PROF. RAJ JAIN

Page 32: Web Performance BootCamp 2013

Hands-on: Using the Response Time Law in the Real World

Let’s say Facebook’s Web servers can process 10K ‘like’ requests/second, and the number of concurrent users is 600K. If each user waits 5s between requests, how long will each request take?

R = (N/X) –Z

Page 33: Web Performance BootCamp 2013

Hands-on: Using the Response Time Law in the Real World

Let’s say Facebook’s Web servers can process 10K ‘like’ requests/second, and the number of concurrent users is 600K. If each user waits 5s between requests, how long will each request take?

R = (N/X) –Z

= 6 * 10e5 / (10 requests/ms) -5 * 10e4ms

= 6 * 10e4 – 5* 10e4 ms

= 1000ms

Facebook needs more servers!

Page 34: Web Performance BootCamp 2013

Antipattern: Keyhole optimization

Problem: Optimizing your project

Antipattern: Optimizing *your* project, at the expense of everyone else!

Pattern: Your project is part of the system – optimize for the overall system performance, not just what you can see, even if that means your part is less-than-perfectly optimized.

Page 35: Web Performance BootCamp 2013

Section IIIThe MPPC Model

Page 36: Web Performance BootCamp 2013

Dimensions of Performance

• Geography• Network location• Bandwidth• Transport Type• Browser/Device Type

• RT Varies by as much as 50%• Page Composition

• Client-side rendering and execution effects (JS, CSS)• Network Transport Effects

• # of Connections, CDN Use

Page 37: Web Performance BootCamp 2013

Hardware and Routing

Page 38: Web Performance BootCamp 2013

The OSI Stack Model

Page 39: Web Performance BootCamp 2013

OSI Functionality Summary

Physical

Data Link

Network

Transport

Session

Presentation

Application

Transmit bits

Organize bits into frames

Transmit packets from source to destination

Reliable message delivery end to end

Establish, manage, terminate sessions

Translate, compress, encrypt

Access network resources

Page 40: Web Performance BootCamp 2013

All People Seem To Need Data Processing

Physical

Data Link

Network

Transport

Session

Presentation

Application

Processing

Data

Need

To

Seem

People

All

Page 41: Web Performance BootCamp 2013

HTTP Connection Flow

Estimated server processing time

Handshake time

Client’s perceived response time Request

Response

Connection setup

Client Server

Request transmission time

Response transmission

time

The more HTTP requests & network roundtrips you require, the slower your performance will be: Images, CSS, JS, DNS lookups, Redirects, #of packets

Page 42: Web Performance BootCamp 2013

The MPPC Model Of Web Performance

End User

HTTP RequestDNS/Network

Resolution

Page Composition

Payload Delivery Time

HTTP Request

HTTP Response HTTP ResponseBrowser Rendering Time

Request Initiation by

User

This entire cycle, steps 1-4, is repeated once for each external reference on the page, so for a given page the total time is:

Where n is the number of external page requisites.

T = S Dt1 + Dt2 + Dt3 + Dt4

n+1

Sn+1

t1

t2

t3t4

T1

Connection Time

T2

Server Duration

T3

Transport Time

T4

Render Time

“Multiple Parallel Persistent Connections”

Page 43: Web Performance BootCamp 2013

T1 – Making the Connection

t1= tDNS + tTCP+ tSSL

• Typically a larger part of the E2E than expected

• Highly variable

• SSL is slow!

Page 44: Web Performance BootCamp 2013

Why DNS Matters

• Nothing happens before DNS!

• User does not see anything on their page waiting time

• Homework Assignment: create a host file for yourself. Try your favorite sites without DNS!

• DNS has a great impact on user’s perceptions in HTTP applications

Page 45: Web Performance BootCamp 2013

Interacting with DNS: dig

time

Page 46: Web Performance BootCamp 2013

T2 – The Server Duration

• Let ( /l m) = dr

• U = (dr)[1-(dr)W]

• X = U * m

• Navg = (dr)[W(dr)W+1 -(W+1)(dr)W+1]

• … so t2 = Navg/X

(The response time law)

Page 47: Web Performance BootCamp 2013

T3 – TCP Transport Time

• Single Object:

t3 = Sz/R+2RTT+tidle

For persistent parallel connections:

t3 = (M+1)Si/Ri+[M/kNh]*3SRTTi+tidle

… for 1 base HTML page with M objects, with Si bits, at bandwidth Ri, k connections per host, and Nh unique hostnames

Page 48: Web Performance BootCamp 2013

T4 - What the Browser Does

t4 = S Dtoff(i)

Dtoff = time offset

to parse the HTML, JS, CSS, and establish the individual connections

(to different hostnames)

t4 is especially significant for mobile devices!

n

i = 1

Page 49: Web Performance BootCamp 2013

Where are the delays?

Page 50: Web Performance BootCamp 2013

Bandwidth Efficiency

Bmax = 1.22*(L)^1/2 * MSS/RTT

Page 51: Web Performance BootCamp 2013

Bandwidth, Latency, and All That

Page 52: Web Performance BootCamp 2013

More Bandwidth? Less Latency!

Page 53: Web Performance BootCamp 2013

Hands-on: Testing DNS Response times

We’ll use nslookup for this exercise

1. Run 10 nslookup commands for a site (e.g. www.facebook.com)

2. Observe the response time for the DNS lookup

3. Calculate statistics for the results • 5 number report (summary)• Sketch the distribution

• What can you say about the response times for DNS?

Page 54: Web Performance BootCamp 2013

Antipattern: That’s Outside My Control

It’s never the case that there is ‘nothing you can do’ about a performance problem.

Antipattern: avoiding solving a performance issue because you think it’s outside your control. This path leads to despair.

Pattern: Compensate in some other part of the E2E. Think outside the box

Page 55: Web Performance BootCamp 2013
Page 56: Web Performance BootCamp 2013

Section IV Tools and Testing

Page 57: Web Performance BootCamp 2013

Let’s Talk Tools

Site Performance Services– Gomez– Keynote– AlertSite– ThousandEyes

• ‘Wholesale’ Testing– Statistical data for many

page views under different conditions– Operational testing – Best for understanding

global and network effects

Page Analysis Tools– YSlow– MS Virtual RoundTrip Analyzer,

HTTPWatch, Many Others– F12 in your browser

• ‘Retail’ Testing– One Page or App– Diagnostic– Best for functional testing

Page 58: Web Performance BootCamp 2013

Commercial Testing Services

• Gomez, AlertSite, and Keynote toolsets are similar in many ways• Synthetic Test Setup• Test nodes in large datacenters and/or end user’s machines• Statistical data about response times

Page 59: Web Performance BootCamp 2013

Performance Testing Locations

Your Data Centers

QA & Test

Page 60: Web Performance BootCamp 2013

HTTP Object Model

Web Page(s)

Page Objects

(or Components)A Test is a sequence of one or more URLs for which HTTP requests will be made.

A Monitor is a set of predefined Tests to be run at specific times and places

Each Page Object has 4

associated time segments, t1, t2, t3, t4

Page 61: Web Performance BootCamp 2013

Desktop Tools

• Methodology– DOM Crawler and

Packet Sniffer– More accurate– Analyzes components– Stats view

• Implements the 14 18 22 105 YSlow Rules– All browsers except IE– Mobile bookmarklet– Best tool for page

analysis

Page 62: Web Performance BootCamp 2013

Unix Performance Testing Tools

• Ping• nslookup• dig• traceroute• netstat

Page 63: Web Performance BootCamp 2013

Task-Based Performance Thinking

Welcome(A)

Inbox(B)

Bulk(C)

Compose(E)

ReadMessage(D) Verify

(F)

SendConfirm(G)

P(A,B) = 0.5168

9.85%

35.02%

66.25%10.04%

61.77%

34.52%

58.07%64.53%

28.04%

34.38%22.30%

Welcome 1400msInbox 2200msBulk 3200ms…

Exercise: What % of users follow the path:

A->E->F->G?

Page 64: Web Performance BootCamp 2013

Testing Your Competitors for Fun & Profit

Page 65: Web Performance BootCamp 2013

Stormcat: Global Performance Testing

• Cannot compare performance data out-of-region• There are many global factors involved in performance:

• Bandwidth• ISP• Infrastructure• Secular cycles (weeks, holidays, usage patterns)

• The best approach: use the ‘StormCat’ system!• Best case (Northern California high broadband @3 AM)• Worst Case (rural Indonesia on VSNL @ 2PM local)• Divide the range into 5 categories equally spaced between the

best & worst: some locales will be in Cat I, some in Cat II, some in Cat III, etc.

Page 66: Web Performance BootCamp 2013

Hands-on: Analyzing Waterfall Diagrams

http://www.webpagetest.org• Choose a location and a

browser and test:

http://www.yahoo.com

Page 67: Web Performance BootCamp 2013

Cached v. Uncached

Page 68: Web Performance BootCamp 2013

Antipattern: Design-time Failure

Performance is a design time activity!

Anti-Pattern: Releasing a new or modified product without testing its

performance

“Bake it in up front!”

Page 69: Web Performance BootCamp 2013

Section VHTML5 & Performance

Page 70: Web Performance BootCamp 2013

A Federated Model for HTML

Core HTML5

HTML Markup

HTML Media

IndexDB

Web Storage

Web Sockets

Web Workers

Canvas 2D

Source: Sergey Mavrody c. 2013

This is XHTML 1.1

Page 71: Web Performance BootCamp 2013

The Co-Evolution of HTML, JS, CSS, and XML

Source: Sergey Mavrody c. 2013

Document Object Model

JavaScript…

JSON…

XML CoreXSLTXSD

Xpath/XQuery

Page 72: Web Performance BootCamp 2013

Tower of Babel: A Problem We Have Yet to Solve

Page 73: Web Performance BootCamp 2013

The Current Browser Landscape

Page 74: Web Performance BootCamp 2013

HTML5 Performance

Page 75: Web Performance BootCamp 2013

W3C Resource Timing

Page 76: Web Performance BootCamp 2013

W3C Navigation Timing

Page 77: Web Performance BootCamp 2013

Hands-on Exercise: Testing Performance the W3C Way

• Use the Navlet:

http://code.google.com/p/navlet/

Make a bookmark or favorite using the code

Page 78: Web Performance BootCamp 2013

Antipattern: We’ll Be Done With This Soon

Performance is an ongoing activity, not fire and forget!

Antipattern: Not treating performance as a property of the system, or only testing at release time.

Pattern: establishing a long-term performance management plan as part of your cycle.

Page 79: Web Performance BootCamp 2013

Section VIMobile Devices

Page 80: Web Performance BootCamp 2013

The Big Picture – Mobile is Growing

Page 81: Web Performance BootCamp 2013

Native Apps v. HTML5 v. Desktop

• Native Apps will run ~ 5x faster than HTML5

• Roughly 10x slower than desktop

• HTML5 on the mobile device can be 50x slower– 10x from the ARM chip– 5x from JavaScript

Page 82: Web Performance BootCamp 2013

If you’re designing for mobile, it’s safe to assume you’re going to incur 2000ms of 3G latency.

Mobile Apps Are Slow

Page 83: Web Performance BootCamp 2013

Slow Compared to What?

Since 2009 mobile browsers went from 30x to 5x slower than desktops

– Better than Moore’s Law improvement (!)

– JavaScript v. Native code ~ 5x– 4g/LTE ~ 27% faster than 3g

Page 84: Web Performance BootCamp 2013

Mobile Speeds by Example

Page 85: Web Performance BootCamp 2013

Users Expect More. Now.

Page 86: Web Performance BootCamp 2013

Mobile JavaScript Performance

• That 5x is in the code interpretation?

• Typed arrays• JSON layout• DOM Manipulation• Garbage Collection

Page 87: Web Performance BootCamp 2013

How 3G/4G Networks Work

Page 88: Web Performance BootCamp 2013

3G to 4G Migration

Page 89: Web Performance BootCamp 2013

HTML5: New Features for Mobile

http://mobilehtml5.org/

Page 90: Web Performance BootCamp 2013

Delay-tolerant Application Design

• Plan for offline/intermittent connectivity

• Caching local content– Local storage– Don’t be afraid to use sessions– Use HTTP Caching headers

wisely• Always have failure modes built-in

Page 91: Web Performance BootCamp 2013

The Right Tool For the Right Job

Source: Nick Zakas

Page 92: Web Performance BootCamp 2013

Best Practices for Mobile

• Tread lightly on the JavaScript• Don’t touch the DOM!• CDNs are less effective due to network challenges• TTFB is not a good measure of server duration• Use Web Workers for preloading• Test performance on different transport types• Test battery consumption!

Page 93: Web Performance BootCamp 2013

4 Takeaways on Mobile Performance

Mobile HTML apps are slow compared to native apps

…but it’s not all about JavaScript

Mobile networking is a big challenge

…so design for delay-tolerance

HTML5 is designed for Mobile

…so use it (wisely)!

Use the right tool for the right job

…including the right design patterns for Mobile

Page 94: Web Performance BootCamp 2013

Tools for Mobile Testing

• Speedtest/Ookla– Variability– Characteristics of different kinds of networks

• iCurl – Simple HTTP Operations on your device

• HTTPWatch Basic– Look at the Waterfall– Gather detailed data along with iCurl

Page 95: Web Performance BootCamp 2013

Hands-on Exercise: Testing Mobile Performance

Page 96: Web Performance BootCamp 2013

Antipattern: It’s The Application Stupid!

T2 (server duration ~ 35% of total E2E– More on mobile however!

Antipattern: Failing to recognize that the distribution of the Mobile E2E is very different from a desktop performance profile

Pattern: Carefully analyze the MPPC numbers for your site and identify the problems that need to be solved and in what order.

Page 97: Web Performance BootCamp 2013

Finale – Summing Up

Page 98: Web Performance BootCamp 2013

The 7 Habits of Exceptional Performance

1. Make Performance a Priority

2. Test, Measure, Test Again

3. Learn about the Tools

4. Balance Performance with Features

5. Track Results Over Time

6. Set Targets

7. Ask Questions; Check It for Yourself!

Thanks to Tenni Theurer

Page 100: Web Performance BootCamp 2013

Every Tool Has Its Place in the Universe

Page 101: Web Performance BootCamp 2013

The One Number of Truth

“42”

Page 102: Web Performance BootCamp 2013

“…a single user-interface to many large classes of stored information such as reports, notes, data-bases, computer documentation and on-line systems help”

WorldWideWeb: Proposal for a HyperText Project

Berners-Lee & Caillau, 1990

About:HTML

Page 103: Web Performance BootCamp 2013

Theme of the Work

Ultimately, Performance is about Respect.

Page 104: Web Performance BootCamp 2013

Thank You!Daniel Austin

PayPal, Inc.

HTML5 DevConf

Oct. 24, 2013

@daniel_b_austin

[email protected]