SlideShare a Scribd company logo
Best Practices
Protect your Online Revenue through Web
Performance Testing


  Melinda Carol-Ballou - Application Life-Cycle Management
                            pp             y        g
  & IT Executive Strategies, IDC


  Imad Mouline - CTO Gomez
                 CTO,
Optimizing Application Quality and
 Performance to Drive the Business


Melinda Ballou
Application Life-Cycle Management
IT Executive Strategies
   E     ti St t i
September 2009


Copyright 2009 IDC. Reproduction is forbidden unless authorized. All rights reserved.
Agenda


          Understanding the drivers and challenges for taking a
      life-cycle approach to application quality and performance

        Coordinating quality and performance from
      requirements building through development and
      deployment

         Establishing process maturity and organizational
      strategies for successful application deployments




© 2009 IDC                                                         3
Disruptive Trends Driving Automated
Software Quality Adoption (ASQ)
• Evolving, flexible development paradigm with services creation will
increasingly demand technology and business collaboration and drive for
service quality within and outside the firewall – strong agile emergence
• Changing approaches to application testing as application packages
evolve to new architectures
• End-user experience and business impact challenges of emerging web
technologies (rich Internet), virtualization, cloud computing and Web 2.0
• Global economic competition plus local compliance initiatives across
geographies demand quality, adaptability and rigor
• Complex sourcing/off-shoring necessitate strong teaming and effective
testing and quality management; limited infrastructure creates barriers
• License and delivery mechanisms with SaaS for new models, choices,
customer support, speed adoption, limit complexity (
            pp , p          p   ,          p     y (human and systems)
                                                               y     )


© 2009 IDC                                                                  4
“Quality Gap”: High Cost of Failure

       Poor Quality = Increased Business Risk
                  y

        ($$$$$)
     Lost Revenue       Lost Customers       Increased Costs




     Damaged Brand        Lost P d ti it
                          L t Productivity    Lower P fit
                                              L     Profits


© 2009 IDC                                                     5
Two Goals of Effective IT/Business
Alignment

             Innovation:
       Maximize Upside
     Through Technology-
      Enabled Business
          Processes
                            New Business
                                Value

                              Reduced
                              Exposure
         Compliance:
        Minimize Downside
          Through Risk
                g
           Management


© 2009 IDC                                 6
Key IT Challenges: Business-
Criticality

       Today’s applications are high-visibility, and carry a
     high cost of failure:
          cost-of-failure:
             Customer “self-serve”
             Supplier/channel integration
             Internal applications that automate critical business
             processes
       “Network effect”: failure in one applicationcan mean
     loss of service in others.
       “Outside-in” Testing is Increasingly Important Due
     to Web Application Delivery Chain Complexity



© 2009 IDC                                                           7
Performance Problems: a Leading
Cause of Failure for New Applications
     Q. When your software fails to satisfy users, the top 3 reasons are:

                                    Flawed specification
                                            p

             Under-estimated resources /time required

             Poor performance /capacity/response time

         Failure of app. to evolve with business needs

                Inability to integrate with other software

                      Dissatisfaction with user interface

                                     Data quality issues

              Flawed implementation of a correct spec

                         Quality of code /lack of testing

                    Inefficient or incorrect configuration

                                                             0   5    10   15   20   25   30   35
                                        Source: IDC n=467
                                                                     (% of respondents)
© 2009 IDC                                                                                          8
IT Business Challenges: Silos, Gaps
  The need for a quality life-cycle is key since G2000
 organizations are split across interdependent groups:
    Business/users stakeholders
    Developers
    QA professionals
     Operational staff
   Increased complexity for both technology and source
 (offshore, etc.) drives need for coordination
   Must extend the Quality life cycle across phases and
                            life-cycle
 groups




© 2009 IDC                                                9
Rise of Global Teams and Businesses


                                    European
      UI Team    Dev/QA/
                                   Business Unit
                 Support                                  APAC
                                             Dev/QA/   Business Unit
                                             Support
                  US Business
                                Content
                   Sponsors
Content
             Business
             Analysts                                       Dev/QA/
                                                            Support




© 2009 IDC                                                             10
Coordinating across the Life-Cycle

• Coordinating requirements, testing, and operational
performance are key across core emerging technologies
• Slow response times for key business areas are problematic
• Organizations should target quality life-cycle approaches
through requirements, unit t t system i t
th    h      i     t     it test,   t     integration and pre-
                                                 ti     d
deployment and performance testing and change management
• Leveraging test automation for synthetic transaction
monitoring enables users to manage user experience with
applications
• As business requirements change, a cogent life-cycle
approach enables adaptive testing and monitoring for flexible
responses


© 2009 IDC                                                      11
Establishing Process and Org Strategies
 Engage Upfront Business input with requirements and QA
and change activities via organizational and process strategy
       “Customer advocates” – attuned to business and IT
             – Design a test plan based on how real users interact with
               your applications
                 • Where they do it
                    – LA, Edmonton, London, Bangkok, NY, etc
                 • What they do
                    – Key tasks, transactions, and business processes
                 • How they do it
                    – Browsers galore, mobile devices
                 • When they do it (peak times, spikes and how frequently)
       Better estimate effort  better coverage
       Early requirements/test/design helps reveal design p
           y q                     g     p             g problems –
       managed change key

© 2009 IDC                                                                   12
Establishing Process and Org Strategies

     Involve LoB & Ops in projects earlier
             Refine test plans
               – Identify key geographies and third-party components to test
               – Help define third-party SLAs, and overall “business SLAs”

             Coordinate testing of external components and
             services delivered from outside the datacenter
             Anticipate deployment and configuration issues
     Leverage tools, tests and skill sets across the life-cycle
   to ensure functionality, availability, and performance
                         y             y      p
   goals are met




© 2009 IDC                                                                     13
Application Quality – Design and Delivery

                   Quality Evaluation Matrix

                  •Easy to Use            •Easy to Use
                 •Unpredictable
                  U    di t bl            •Behaves as
                                           B h
                    Behavior               Designed          Quality of
      Design




                                                             Customer
                                                             Experience
                 •Difficult to Use       •Difficult to Use
                 •Unpredictable           •Behaves as
                   Behavior
                   Beha ior                Designed
                                           D i    d



                              Delivery

© 2009 IDC                                                                14
App Deployment & Support: Old View


 DEV

                          Define
                          SLAs,           Test/
                         Provision                  Monitor      Support
                                          Tune

                                                                  Issues
  Result:
               Little input into specifications or development
               Little leverage between development/ops of
               testing/monitoring investments
               Testing/tuning LATE in the cycle!

© 2009 IDC                                                                 15
Closing the Loop: Leverage Skills &
             Tools for Agile, Iterative Approach

 •Shared goals,
          g     ,
 •Shared metrics,
 •Share tests,                        Design
  tools,and
  tools and skills        Define
 •Shared systems
                                          Develop
                        Support


                           Deploy &     Test &
                           Monitor       Tune


© 2009 IDC                                          16
Establish a Common Language

    “Sales says       “Call Center’s              “Network statistics look great.”
  customers can t
             can’t    swamped with                  “All servers are available.”
                                                     All             available
find product in the    Order Status
     database.”         inquiries.”                “All databases are available.”
                                               “99% of our monitored apps are OK”
 “Order                      “We’re not          “Service levels are within limits.”
                                                  Service                   limits.
volume is                  meeting our first
 down.”                     call resolution
                            metric today.”




              Business                                           IT
                Bridge the communication gap!
                Establish a common language!
© 2009 IDC                                                                             17
Bridging the Customer Experience Gap

                                                           “Is the end-to-end
   Business
   Management
   M
                                                           business process
                                                           working?”

  Customer              End-to-End Usage Scenarios
                                      g
  Experience                                                 “Can customers
  Management                                                 do what they
                     Web, Voice
                     Web Voice, Web Services, email FAX
                                    Services email, FAX…     want to do?”
                                                                     do?


   Application                                              “Can I proactively
   Management                                               fix things before
                                                            they impact the
   Server, Network &                                        business?”
   Element Management


© 2009 IDC                                                                       18
Closing a ‘Customer Experience’ Gap

    1. Establish a common language between IT and the
       business, designate a liaison
    2. Agree on business process measurement priorities
    3. Implement testing/management for end-to-end
         p             g      g
       business processes – from the customer/end-user
       perspective
    4. Eliminate silos and close the loop between
       development and operations
    5. S
       Share goals, metrics, f
                             feedback with the entire team:
       make Customer Experience Management part of the
       culture


© 2009 IDC                                                    19
Summary

•Coordinate a Quality Life-Cycle approach from requirements
through to operations to enable adaptive businesses
•Schisms between business, development, testers and
operations must be addressed
•IT groups and the business must build a common language,
common metrics, and common tools and practices
•Focusing on the customer experience will help IT achieve this
along with core incentives to collaboration
•Making “customer experience” the culture is key to help bridge
the gap and to enable successful collaboration across groups



© 2009 IDC                                                    20
Best Practices: Protecting your Online Revenue

                       BP #1: Establish a Common Language
                       between IT and the business




                        BP #2: Make Customer Experience 
                        Management part of the culture




                       BP #3: Implement testing & 
                       management for end‐to‐end business 
                       processes  from the customer/end
                       processes – from the customer/end‐
                       user perspective
BP #1: Establish a Common Language Between IT & the Business:
Customer Satisfaction Index 



  Apdex (Application                    Quickly isolate and prioritize 
  Performance Index)
  P f           I d )                   opportunities for performance 
                                                  ii f       f
      Single measurement of             improvement based on real 
     user satisfaction with             customer satisfaction
      Web performance  
      W b     f                            Why is customer satisfaction 
                                             h                    f
      Based on response time               decreasing? 
      thresholds                           What do users have in 
                                           common?   ?
                                           How do I compare against 
  Satisfied   Tolerating   Frustrated      my competitors?
      http://www.apdex.org 2007
Customer Satisfaction Index: An Example

  A l di                        if i d
  As load increases customer satisfaction decreases




                                                      Customer
                                                      satisfaction
                                                      decreases




                                                       Load
                                                       increases
BP #2: Make Customer Experience Management Part of your 
Culture ‐ Design your Web site for a Multi‐browser World
 Will it work across all browsers?
  ill i     k         ll b       ?
                                              Functions correctly




                                     Does not function correctly = brand damage
Make Customer Experience Management Part of your 
Culture ‐ Test & Monitor your Key Transactions


  Further analysis shows sequence 0 of    However sequence 1, the
  the transaction - the login page - is   authentication step, is
  unaffected                              impacted b a major
                                          i     t d by      j
                                          performance issue
Make Customer Experience Management Part of your Culture 
– Ensure that your Applications Will Scale Under Load

                                   Obama family exclusively
                                   wears leading fashion brand
                                   for inauguration ceremony




                                   Leading fashion brand’s
                                   web application crashes
                                   leading to short and long
                                   term loss of revenue and
                                   brand equity
BP #3: Implement Testing/Management  for End‐to‐end 
  Business Process from the Customer/End‐user Perspective 
    Behind‐the‐                                                            …user is 
firewall testing                                                           NOT happy
     tools: “OK”




                                                                             Users
 Load 
 Balancing
 Web 
 Servers
 App 
 Servers                The traditional view of Web Application Delivery
                        The traditional view of Web Application Delivery
 DB 
 Servers
 Storage
 Mobile 
 Components




    Traditional zone 
       of control
Critical to Test Across Entire Web Application Delivery Chain

    Behind‐the‐                                                                                             …user is 
firewall testing                                                                                            NOT happy
     tools: “OK”
                                     The Web Application Delivery Chain

                                                   3rd Party/                                  Browsers 
                                                 Cloud Services             Local ISP         and Devices     Users
 Load                       •   Inconsistent geo performance         • Network peering           • Poorly 
 Balancing                  •   Bad performance under load
                                Bad performance under load             p
                                                                       problems                    p
                                                                                                   performing g
                            •   Blocking content delivery            • Bandwidth throttling        JavaScript
 Web                                                                 • Inconsistent              • Browser/device 
 Servers                    •   Incorrect geo‐targeted content
                                                                       connectivity                incompatibility
         • Configuration                                                                         • Page size 
 App                                    • Network peering                                          too big
 Servers errors                           problems
         • Application                                                                           •TToo many 
 DB        design issues                • Outages Internet                                         objects
                                                                     • Network resource          • Low cache 
 Servers • Code defects     Major                                      shortage                    hit rate
         • Insufficient      ISP • Configuration issues              • Faulty content 
 Storage infrastructure                                                transcoding 
                                     • Oversubscribed POP
 Mobile                              • Poor routing optimization       SMS routing / 
                                                                     • SMS       i /
 Components                          • Low cache hit rate              latency issues 
                                               Content Delivery             Mobile 
                                                  Networks                  Carrier

    Traditional zone                         Zone of customer expectation
       of control
Load Testing 2.0
 Performance Testing For A Web 2.0 World

                                Self‐service testing gives you 
                                control to test as needed
                                Test internal & external web 
                                T ti t      l& t       l b
                                application components to ensure 
                                transactions perform under load
                                Test cloud‐based applications & 
                                Test cloud‐based applications &
                                services
                                SaaS based Load Testing 2.0 
                                solutions enable on‐demand,  ,
                                scalable testing without associated 
                                hardware investment & 
                                maintenance costs
Combine hi h volume
C   bi    high   l              Test the full web application 
                                delivery chain all the way to your 
application load testing        end users’ desktops
with broad geographic
experience testing from
the outside-in
Gomez Reality Load XF: 
On‐Demand Realistic Load Testing from Browser to Data Center
                                  Backbone                           Last Mile
                                                                     Last Mile
                High volume load (HTTP, Browser)                Real‐world load
                     Find infrastructure breaking points     Find user experience breaking points
                     Define capacity headroom                Accurately measure response time

                                  3rd Party/                                Browsers 
                                Cloud Services             Local ISP       and Devices   Users
Load                                                                   100,000+ 
Balancing                                                              consumer‐
                                                                       grade 
                                                                          d
Web                                                                    desktops
Servers
                                                                       168+ 
App                          100+                                      countries
Servers                      commercial‐
                             grade 
DB                           nodes & Internet                          2,500+ 
                             data                                      ISPs
Servers      Major           centers
              ISP
Storage                                                                Major 
                                                                       mobile 
Mobile                                                                 carriers 
                                                                       around 
Components                                                             the globe
                              Content Delivery             Mobile 
                                 Networks                  Carrier
Ensuring Performance of All 3rd Party Components

     Company: Online Retailer
      • Several 3rd Parties now involved in serving up key content
      • G l was to validate performance of entire application
        Goal      t   lid t    f           f ti        li ti


                                  3rd Party/
                                          y/                   Browsers 
                                Cloud Services    Local ISP   and devices   Users
Load                       *
Balancing
Web 
Servers
App 
Servers
DB                                 Internet
Servers            Major
                    ISP
Storage
Mobile 
Components
                               Content Delivery    Mobile 
                                  Networks         Carrier
Response Times Rise Due To 3rd Party Object Error

The load increases                              The transaction rate
throughout the test                             increases and then
                                                falls off as response
                                                times climb




                                                   Errors are seen, all
                                                   on a 3rd party object




  • 3 d party h d
    3rd    t hardware was i
                          insufficient f overall d
                              ffi i t for     ll demands on application
                                                      d        li ti
  • Based on SLAs 3rd party had to improve performance to get paid
Ensuring Performance in Key Markets

     Company: Regional Online News Source
      • Began testing for the 2008 election season
      • G l was to validate overall performance f
        Goal     t     lid t      ll   f         focusing i 2 k regions
                                                      i in key     i


                                 3rd Party/
                                         y/                       Browsers 
                               Cloud Services        Local ISP   and devices   Users
Load 
Balancing
                                                 *
Web 
Servers
App 
Servers
DB                                Internet
Servers            Major
                    ISP
Storage
Mobile 
Components
                              Content Delivery       Mobile 
                                 Networks            Carrier
No Performance Issues Detected From Data-Center

Increase and hold load and not exceed
response times of 4 seconds and Success     There was only
Rate of 99%                                 1 page error and
                                            11 errors total
                                            out of 60000+
Page response                               transactions
times stayed under
4 seconds, outside
of one brief blip




    By traditional test standards the test passed
Performance Issues Detected From Real User
Desktops


Key geographies for this customer are
     New York and Pennsylvania




Last Mile data showing substantial number of
   measurements greater than 4 seconds
The Gomez Platform: Web Application Experience Management

                                   Self‐
                                   Self‐Service SaaS Portal


         Web                  Web Load                     Web                      Web 
    Cross‐Browser          and Performance             Performance              Performance
       Testing                  Testing                Management             Business Analysis
     • Compatibility           High volume 
                             • High‐volume                 20
                                                     Real user monitoring
                                                   • Real‐user monitoring           Web performance 
                                                                                  • Web performance
     • Functionality           backbone load       • Last Mile monitoring           analytics
     • Performance           • Real‐world          • Backbone monitoring          • Benchmarks
                               Last Mile load                                     • Business dashboard

                                Web 2.0, RIA, Streaming, Mobile 
                                Web 2 0 RIA Streaming Mobile


   Dashboards,             Recording &          Alerting &            APIs &                Education &
 Metrics & Analytics       Provisioning         Diagnostics         Data Feeds             Best Practices


                       World’s Most Comprehensive Testing Network
                                   168+ countries    2,500+ ISPs

              500+               100+                 5,000+               100,000+ 
            combos of      commercial nodes         supported           consumer‐grade 
          browsers & O/S    and data centers       mobile devices          desktops
Gomez Customers Enjoy Measurable Benefits

Increased conversions 10% 

Reduced homepage load time from 
11.3 seconds to 3 4 seconds
11 3      d t 3.4        d

Improved page load times 23%

Saved 50%+ in staff and fees

Reduced seven‐step 
transaction time by 50%
                  y

Reduced downtime 45% 

Achieved under 3 second 
Achieved under 3 second
response time and 99%+ 
availability
Validated decision to consolidate 
Validated decision to consolidate
three data centers
Appendix
Ensuring Performance Of Login Process

     Company: Online presence for a popular TV show
          • Following episodes of the TV show the web site sees high traffic spikes
          • Goal was to achieve 1500 logins per minute
          • Load tested DB to improve performance in anticipation of another traffic spike
                                          3rd Party/
                                                  y/                         Browsers 
                                        Cloud Services        Local ISP     and devices   Users
Load 
Balancing    1
Web 
Servers
App 
Servers
DB                                         Internet
Servers                  Major
                          ISP
Storage
Mobile 
Components
                                       Content Delivery        Mobile 
                                          Networks             Carrier
Application Bottleneck Causes Immediate Response
Time Issue



                            • As users were added the
                                              added,
                              response time of step 3 (the
                              login) climbed immediately

                            • The test bottlenecked at 160
                              logins per minute (Goal 1500)

                            • But quickly dropped off as
                              users received server errors

                            • New login query was not
                              optimized and was
                              bottlenecking the database
                              servers’ CPU
                                      ’ CPUs

More Related Content

PPTX
Application Management and Support - Shared Services Featuring the Pay Per Ti...
PDF
What is Application Management?
PDF
Ibm pure flex client presentation
PPT
IBM Social Business Agenda template
PDF
Rational
 
PDF
Dextrys IT Services Organization Overview
PDF
Application Lifecycle Management & VSTS
PPT
Selling Quality to Management: How to Get Management Buy-In to New Quality In...
Application Management and Support - Shared Services Featuring the Pay Per Ti...
What is Application Management?
Ibm pure flex client presentation
IBM Social Business Agenda template
Rational
 
Dextrys IT Services Organization Overview
Application Lifecycle Management & VSTS
Selling Quality to Management: How to Get Management Buy-In to New Quality In...

What's hot (20)

PDF
6 rpt oracle_plugin-anitha_krishnamurthy
 
PDF
Optimising and prioritising your SDLC using business intelligence
PPTX
Enterprise quality data for the supply chain
PDF
How to make_it_real-hayden_lindsey
 
PPTX
Modern Apps and App Lifecycle
PDF
Monetizing Open Source Software
PPT
Computer Enterprises Inc
PDF
Software Quality Df
PDF
Opportunities in challenging_times-steve_robinson
 
PDF
Enterprise Applications
PDF
Dnv Improving Your Process Performances With Agile
PDF
Application Management by Siemens
PPT
Les outils de Devops IBM
PDF
THE OPEN SOURCE OPPORTUNITY: Monetizing Open Source Though Partnerships
PPTX
CMI Overview
PDF
Organizing Design-Driven Development Using Rational Requirements Composer
PPTX
Dvsl enterprise solutions.v1
PDF
Opposites Attract SOA, Agile, MDA
PDF
Asq Voc Article 0210
6 rpt oracle_plugin-anitha_krishnamurthy
 
Optimising and prioritising your SDLC using business intelligence
Enterprise quality data for the supply chain
How to make_it_real-hayden_lindsey
 
Modern Apps and App Lifecycle
Monetizing Open Source Software
Computer Enterprises Inc
Software Quality Df
Opportunities in challenging_times-steve_robinson
 
Enterprise Applications
Dnv Improving Your Process Performances With Agile
Application Management by Siemens
Les outils de Devops IBM
THE OPEN SOURCE OPPORTUNITY: Monetizing Open Source Though Partnerships
CMI Overview
Organizing Design-Driven Development Using Rational Requirements Composer
Dvsl enterprise solutions.v1
Opposites Attract SOA, Agile, MDA
Asq Voc Article 0210
Ad

Similar to IDC & Gomez Webinar --Best Practices: Protect Your Online Revenue Through Web Performance Testing (20)

PPT
Rational collaborative-lifecycle-management-2012
PPTX
Amazing Winter Keynote - IT as a Team Sport
PPTX
My view on Lean IT
PPTX
SoftServe Corporate Presentation 2013
PDF
Big Data, Big Problems: Avoid System Failure with Quality Analysis - Webinar ...
DOCX
Bs case study
PPT
DevOps vs. ShadowOps (Pulse 2013)
PDF
Curated Computing
PDF
Req Pro - Andreas gschwind
PDF
IBM Rational Software Conference 2009: Requirements Definition & Management T...
PDF
Ibm smarter quality_management
PDF
Everware cbdi opposites attract 04-12-11
PPTX
Infopulse presentation
PDF
2011 National BDPA Tech Conf Presenation Final
PDF
Creating Business Value Through User Experience
PPTX
[QUATIC 2012] A Multi-Model Case Study: High Maturity in Development + Servic...
PDF
Career Choice for Graduates
PDF
The new role of CIO, Borut Kolmanič, S&T Slovenija
PDF
[CMMI Portugal 2012] CMMI ML5 Multimodel in a Portuguese company
PPTX
5 pedrohenriques-2confcmmiportugal-v2-3-split
Rational collaborative-lifecycle-management-2012
Amazing Winter Keynote - IT as a Team Sport
My view on Lean IT
SoftServe Corporate Presentation 2013
Big Data, Big Problems: Avoid System Failure with Quality Analysis - Webinar ...
Bs case study
DevOps vs. ShadowOps (Pulse 2013)
Curated Computing
Req Pro - Andreas gschwind
IBM Rational Software Conference 2009: Requirements Definition & Management T...
Ibm smarter quality_management
Everware cbdi opposites attract 04-12-11
Infopulse presentation
2011 National BDPA Tech Conf Presenation Final
Creating Business Value Through User Experience
[QUATIC 2012] A Multi-Model Case Study: High Maturity in Development + Servic...
Career Choice for Graduates
The new role of CIO, Borut Kolmanič, S&T Slovenija
[CMMI Portugal 2012] CMMI ML5 Multimodel in a Portuguese company
5 pedrohenriques-2confcmmiportugal-v2-3-split
Ad

More from Compuware APM (20)

PPTX
Complexity At The Edge How To Maximize The Mobile Opportunity
PPTX
Complexity At The Edge How To Maximize The Mobile Opportunity In China
PDF
Managing Cost in Public Cloud Environments
PPTX
Don't Lose Your Viewers: Keep Their Attention With High Performance Streaming
PPTX
Why Server Virtualization Demands an End-user’s Perspective
PPTX
Webcast: CIO Insights: How to Optimize User Experience Across 60 Hospitals
PDF
3 Keys to Great Customer Experience When Launching Web and Mobile Applications
PDF
The State Of Mobile Commerce – Are You Meeting Your Mobile Customers’ Expecta...
PDF
Speed Your Apps Launch to Market: Learn Six Best Practices for DevOps
PDF
Recent Survey Reveals that Optimized APM Approaches Increase Business Efficie...
PDF
Web Load Testing for Dummies
PDF
Measure Twice, Cut Once: 5 Best Practices For Selecting Your Cloud Service Pr...
PDF
How to Mitigate the Performance Risk of Third-party Web Components
PDF
Optimizing web and mobile site performance using page speed
PDF
Delta Air Lines Shares Best Practices for Becoming a Top Performing Website
PDF
"What Users Want from Mobile - A study of consumers’ mobile web and applicati...
PDF
Are Your Applications Delivering What Your End-Users Expect?
PDF
Managing Complexity Across Today’s Application Delivery Chain:Six key indicat...
PDF
Top Tips To Deliver Quality Mobile Web And App Experiences To Smartphone Users
PDF
5 Best Practices for Successful Cloud Deployments – and the Pitfalls to Avoid
Complexity At The Edge How To Maximize The Mobile Opportunity
Complexity At The Edge How To Maximize The Mobile Opportunity In China
Managing Cost in Public Cloud Environments
Don't Lose Your Viewers: Keep Their Attention With High Performance Streaming
Why Server Virtualization Demands an End-user’s Perspective
Webcast: CIO Insights: How to Optimize User Experience Across 60 Hospitals
3 Keys to Great Customer Experience When Launching Web and Mobile Applications
The State Of Mobile Commerce – Are You Meeting Your Mobile Customers’ Expecta...
Speed Your Apps Launch to Market: Learn Six Best Practices for DevOps
Recent Survey Reveals that Optimized APM Approaches Increase Business Efficie...
Web Load Testing for Dummies
Measure Twice, Cut Once: 5 Best Practices For Selecting Your Cloud Service Pr...
How to Mitigate the Performance Risk of Third-party Web Components
Optimizing web and mobile site performance using page speed
Delta Air Lines Shares Best Practices for Becoming a Top Performing Website
"What Users Want from Mobile - A study of consumers’ mobile web and applicati...
Are Your Applications Delivering What Your End-Users Expect?
Managing Complexity Across Today’s Application Delivery Chain:Six key indicat...
Top Tips To Deliver Quality Mobile Web And App Experiences To Smartphone Users
5 Best Practices for Successful Cloud Deployments – and the Pitfalls to Avoid

Recently uploaded (20)

PPT
“AI and Expert System Decision Support & Business Intelligence Systems”
PDF
NewMind AI Monthly Chronicles - July 2025
PDF
Diabetes mellitus diagnosis method based random forest with bat algorithm
PDF
Shreyas Phanse Resume: Experienced Backend Engineer | Java • Spring Boot • Ka...
PPTX
Big Data Technologies - Introduction.pptx
PDF
Machine learning based COVID-19 study performance prediction
PDF
Empathic Computing: Creating Shared Understanding
PDF
KodekX | Application Modernization Development
PDF
The Rise and Fall of 3GPP – Time for a Sabbatical?
PDF
Electronic commerce courselecture one. Pdf
PDF
Network Security Unit 5.pdf for BCA BBA.
PPT
Teaching material agriculture food technology
PDF
CIFDAQ's Market Insight: SEC Turns Pro Crypto
PPTX
20250228 LYD VKU AI Blended-Learning.pptx
PDF
Reach Out and Touch Someone: Haptics and Empathic Computing
PDF
NewMind AI Weekly Chronicles - August'25 Week I
PDF
Peak of Data & AI Encore- AI for Metadata and Smarter Workflows
PPTX
PA Analog/Digital System: The Backbone of Modern Surveillance and Communication
PDF
Per capita expenditure prediction using model stacking based on satellite ima...
PPTX
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
“AI and Expert System Decision Support & Business Intelligence Systems”
NewMind AI Monthly Chronicles - July 2025
Diabetes mellitus diagnosis method based random forest with bat algorithm
Shreyas Phanse Resume: Experienced Backend Engineer | Java • Spring Boot • Ka...
Big Data Technologies - Introduction.pptx
Machine learning based COVID-19 study performance prediction
Empathic Computing: Creating Shared Understanding
KodekX | Application Modernization Development
The Rise and Fall of 3GPP – Time for a Sabbatical?
Electronic commerce courselecture one. Pdf
Network Security Unit 5.pdf for BCA BBA.
Teaching material agriculture food technology
CIFDAQ's Market Insight: SEC Turns Pro Crypto
20250228 LYD VKU AI Blended-Learning.pptx
Reach Out and Touch Someone: Haptics and Empathic Computing
NewMind AI Weekly Chronicles - August'25 Week I
Peak of Data & AI Encore- AI for Metadata and Smarter Workflows
PA Analog/Digital System: The Backbone of Modern Surveillance and Communication
Per capita expenditure prediction using model stacking based on satellite ima...
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx

IDC & Gomez Webinar --Best Practices: Protect Your Online Revenue Through Web Performance Testing

  • 1. Best Practices Protect your Online Revenue through Web Performance Testing Melinda Carol-Ballou - Application Life-Cycle Management pp y g & IT Executive Strategies, IDC Imad Mouline - CTO Gomez CTO,
  • 2. Optimizing Application Quality and Performance to Drive the Business Melinda Ballou Application Life-Cycle Management IT Executive Strategies E ti St t i September 2009 Copyright 2009 IDC. Reproduction is forbidden unless authorized. All rights reserved.
  • 3. Agenda Understanding the drivers and challenges for taking a life-cycle approach to application quality and performance Coordinating quality and performance from requirements building through development and deployment Establishing process maturity and organizational strategies for successful application deployments © 2009 IDC 3
  • 4. Disruptive Trends Driving Automated Software Quality Adoption (ASQ) • Evolving, flexible development paradigm with services creation will increasingly demand technology and business collaboration and drive for service quality within and outside the firewall – strong agile emergence • Changing approaches to application testing as application packages evolve to new architectures • End-user experience and business impact challenges of emerging web technologies (rich Internet), virtualization, cloud computing and Web 2.0 • Global economic competition plus local compliance initiatives across geographies demand quality, adaptability and rigor • Complex sourcing/off-shoring necessitate strong teaming and effective testing and quality management; limited infrastructure creates barriers • License and delivery mechanisms with SaaS for new models, choices, customer support, speed adoption, limit complexity ( pp , p p , p y (human and systems) y ) © 2009 IDC 4
  • 5. “Quality Gap”: High Cost of Failure Poor Quality = Increased Business Risk y ($$$$$) Lost Revenue Lost Customers Increased Costs Damaged Brand Lost P d ti it L t Productivity Lower P fit L Profits © 2009 IDC 5
  • 6. Two Goals of Effective IT/Business Alignment Innovation: Maximize Upside Through Technology- Enabled Business Processes New Business Value Reduced Exposure Compliance: Minimize Downside Through Risk g Management © 2009 IDC 6
  • 7. Key IT Challenges: Business- Criticality Today’s applications are high-visibility, and carry a high cost of failure: cost-of-failure: Customer “self-serve” Supplier/channel integration Internal applications that automate critical business processes “Network effect”: failure in one applicationcan mean loss of service in others. “Outside-in” Testing is Increasingly Important Due to Web Application Delivery Chain Complexity © 2009 IDC 7
  • 8. Performance Problems: a Leading Cause of Failure for New Applications Q. When your software fails to satisfy users, the top 3 reasons are: Flawed specification p Under-estimated resources /time required Poor performance /capacity/response time Failure of app. to evolve with business needs Inability to integrate with other software Dissatisfaction with user interface Data quality issues Flawed implementation of a correct spec Quality of code /lack of testing Inefficient or incorrect configuration 0 5 10 15 20 25 30 35 Source: IDC n=467 (% of respondents) © 2009 IDC 8
  • 9. IT Business Challenges: Silos, Gaps The need for a quality life-cycle is key since G2000 organizations are split across interdependent groups: Business/users stakeholders Developers QA professionals Operational staff Increased complexity for both technology and source (offshore, etc.) drives need for coordination Must extend the Quality life cycle across phases and life-cycle groups © 2009 IDC 9
  • 10. Rise of Global Teams and Businesses European UI Team Dev/QA/ Business Unit Support APAC Dev/QA/ Business Unit Support US Business Content Sponsors Content Business Analysts Dev/QA/ Support © 2009 IDC 10
  • 11. Coordinating across the Life-Cycle • Coordinating requirements, testing, and operational performance are key across core emerging technologies • Slow response times for key business areas are problematic • Organizations should target quality life-cycle approaches through requirements, unit t t system i t th h i t it test, t integration and pre- ti d deployment and performance testing and change management • Leveraging test automation for synthetic transaction monitoring enables users to manage user experience with applications • As business requirements change, a cogent life-cycle approach enables adaptive testing and monitoring for flexible responses © 2009 IDC 11
  • 12. Establishing Process and Org Strategies Engage Upfront Business input with requirements and QA and change activities via organizational and process strategy “Customer advocates” – attuned to business and IT – Design a test plan based on how real users interact with your applications • Where they do it – LA, Edmonton, London, Bangkok, NY, etc • What they do – Key tasks, transactions, and business processes • How they do it – Browsers galore, mobile devices • When they do it (peak times, spikes and how frequently) Better estimate effort better coverage Early requirements/test/design helps reveal design p y q g p g problems – managed change key © 2009 IDC 12
  • 13. Establishing Process and Org Strategies Involve LoB & Ops in projects earlier Refine test plans – Identify key geographies and third-party components to test – Help define third-party SLAs, and overall “business SLAs” Coordinate testing of external components and services delivered from outside the datacenter Anticipate deployment and configuration issues Leverage tools, tests and skill sets across the life-cycle to ensure functionality, availability, and performance y y p goals are met © 2009 IDC 13
  • 14. Application Quality – Design and Delivery Quality Evaluation Matrix •Easy to Use •Easy to Use •Unpredictable U di t bl •Behaves as B h Behavior Designed Quality of Design Customer Experience •Difficult to Use •Difficult to Use •Unpredictable •Behaves as Behavior Beha ior Designed D i d Delivery © 2009 IDC 14
  • 15. App Deployment & Support: Old View DEV Define SLAs, Test/ Provision Monitor Support Tune Issues Result: Little input into specifications or development Little leverage between development/ops of testing/monitoring investments Testing/tuning LATE in the cycle! © 2009 IDC 15
  • 16. Closing the Loop: Leverage Skills & Tools for Agile, Iterative Approach •Shared goals, g , •Shared metrics, •Share tests, Design tools,and tools and skills Define •Shared systems Develop Support Deploy & Test & Monitor Tune © 2009 IDC 16
  • 17. Establish a Common Language “Sales says “Call Center’s “Network statistics look great.” customers can t can’t swamped with “All servers are available.” All available find product in the Order Status database.” inquiries.” “All databases are available.” “99% of our monitored apps are OK” “Order “We’re not “Service levels are within limits.” Service limits. volume is meeting our first down.” call resolution metric today.” Business IT Bridge the communication gap! Establish a common language! © 2009 IDC 17
  • 18. Bridging the Customer Experience Gap “Is the end-to-end Business Management M business process working?” Customer End-to-End Usage Scenarios g Experience “Can customers Management do what they Web, Voice Web Voice, Web Services, email FAX Services email, FAX… want to do?” do? Application “Can I proactively Management fix things before they impact the Server, Network & business?” Element Management © 2009 IDC 18
  • 19. Closing a ‘Customer Experience’ Gap 1. Establish a common language between IT and the business, designate a liaison 2. Agree on business process measurement priorities 3. Implement testing/management for end-to-end p g g business processes – from the customer/end-user perspective 4. Eliminate silos and close the loop between development and operations 5. S Share goals, metrics, f feedback with the entire team: make Customer Experience Management part of the culture © 2009 IDC 19
  • 20. Summary •Coordinate a Quality Life-Cycle approach from requirements through to operations to enable adaptive businesses •Schisms between business, development, testers and operations must be addressed •IT groups and the business must build a common language, common metrics, and common tools and practices •Focusing on the customer experience will help IT achieve this along with core incentives to collaboration •Making “customer experience” the culture is key to help bridge the gap and to enable successful collaboration across groups © 2009 IDC 20
  • 21. Best Practices: Protecting your Online Revenue BP #1: Establish a Common Language between IT and the business BP #2: Make Customer Experience  Management part of the culture BP #3: Implement testing &  management for end‐to‐end business  processes  from the customer/end processes – from the customer/end‐ user perspective
  • 22. BP #1: Establish a Common Language Between IT & the Business: Customer Satisfaction Index  Apdex (Application  Quickly isolate and prioritize  Performance Index) P f I d ) opportunities for performance  ii f f Single measurement of  improvement based on real  user satisfaction with  customer satisfaction Web performance   W b f Why is customer satisfaction  h f Based on response time   decreasing?  thresholds What do users have in  common?  ? How do I compare against  Satisfied Tolerating Frustrated my competitors? http://www.apdex.org 2007
  • 23. Customer Satisfaction Index: An Example A l di if i d As load increases customer satisfaction decreases Customer satisfaction decreases Load increases
  • 25. Make Customer Experience Management Part of your  Culture ‐ Test & Monitor your Key Transactions Further analysis shows sequence 0 of However sequence 1, the the transaction - the login page - is authentication step, is unaffected impacted b a major i t d by j performance issue
  • 26. Make Customer Experience Management Part of your Culture  – Ensure that your Applications Will Scale Under Load Obama family exclusively wears leading fashion brand for inauguration ceremony Leading fashion brand’s web application crashes leading to short and long term loss of revenue and brand equity
  • 27. BP #3: Implement Testing/Management  for End‐to‐end  Business Process from the Customer/End‐user Perspective  Behind‐the‐ …user is  firewall testing  NOT happy tools: “OK” Users Load  Balancing Web  Servers App  Servers The traditional view of Web Application Delivery The traditional view of Web Application Delivery DB  Servers Storage Mobile  Components Traditional zone  of control
  • 28. Critical to Test Across Entire Web Application Delivery Chain Behind‐the‐ …user is  firewall testing  NOT happy tools: “OK” The Web Application Delivery Chain 3rd Party/ Browsers  Cloud Services Local ISP and Devices Users Load  • Inconsistent geo performance • Network peering  • Poorly  Balancing • Bad performance under load Bad performance under load p problems p performing g • Blocking content delivery • Bandwidth throttling JavaScript Web  • Inconsistent  • Browser/device  Servers • Incorrect geo‐targeted content connectivity incompatibility • Configuration  • Page size  App  • Network peering  too big Servers errors problems • Application  •TToo many  DB  design issues • Outages Internet objects • Network resource  • Low cache  Servers • Code defects Major shortage  hit rate • Insufficient  ISP • Configuration issues • Faulty content  Storage infrastructure transcoding  • Oversubscribed POP Mobile  • Poor routing optimization SMS routing /  • SMS i / Components • Low cache hit rate latency issues  Content Delivery Mobile  Networks Carrier Traditional zone  Zone of customer expectation of control
  • 29. Load Testing 2.0 Performance Testing For A Web 2.0 World Self‐service testing gives you  control to test as needed Test internal & external web  T ti t l& t l b application components to ensure  transactions perform under load Test cloud‐based applications &  Test cloud‐based applications & services SaaS based Load Testing 2.0  solutions enable on‐demand,  , scalable testing without associated  hardware investment &  maintenance costs Combine hi h volume C bi high l Test the full web application  delivery chain all the way to your  application load testing end users’ desktops with broad geographic experience testing from the outside-in
  • 30. Gomez Reality Load XF:  On‐Demand Realistic Load Testing from Browser to Data Center Backbone Last Mile Last Mile High volume load (HTTP, Browser) Real‐world load Find infrastructure breaking points Find user experience breaking points Define capacity headroom Accurately measure response time 3rd Party/ Browsers  Cloud Services Local ISP and Devices Users Load  100,000+  Balancing consumer‐ grade  d Web  desktops Servers 168+  App  100+  countries Servers commercial‐ grade  DB  nodes & Internet 2,500+  data  ISPs Servers Major centers ISP Storage Major  mobile  Mobile  carriers  around  Components the globe Content Delivery Mobile  Networks Carrier
  • 31. Ensuring Performance of All 3rd Party Components Company: Online Retailer • Several 3rd Parties now involved in serving up key content • G l was to validate performance of entire application Goal t lid t f f ti li ti 3rd Party/ y/ Browsers  Cloud Services Local ISP and devices Users Load  * Balancing Web  Servers App  Servers DB  Internet Servers Major ISP Storage Mobile  Components Content Delivery Mobile  Networks Carrier
  • 32. Response Times Rise Due To 3rd Party Object Error The load increases The transaction rate throughout the test increases and then falls off as response times climb Errors are seen, all on a 3rd party object • 3 d party h d 3rd t hardware was i insufficient f overall d ffi i t for ll demands on application d li ti • Based on SLAs 3rd party had to improve performance to get paid
  • 33. Ensuring Performance in Key Markets Company: Regional Online News Source • Began testing for the 2008 election season • G l was to validate overall performance f Goal t lid t ll f focusing i 2 k regions i in key i 3rd Party/ y/ Browsers  Cloud Services Local ISP and devices Users Load  Balancing * Web  Servers App  Servers DB  Internet Servers Major ISP Storage Mobile  Components Content Delivery Mobile  Networks Carrier
  • 34. No Performance Issues Detected From Data-Center Increase and hold load and not exceed response times of 4 seconds and Success There was only Rate of 99% 1 page error and 11 errors total out of 60000+ Page response transactions times stayed under 4 seconds, outside of one brief blip By traditional test standards the test passed
  • 35. Performance Issues Detected From Real User Desktops Key geographies for this customer are New York and Pennsylvania Last Mile data showing substantial number of measurements greater than 4 seconds
  • 36. The Gomez Platform: Web Application Experience Management Self‐ Self‐Service SaaS Portal Web  Web Load  Web  Web  Cross‐Browser  and Performance  Performance  Performance Testing Testing Management Business Analysis • Compatibility High volume  • High‐volume 20 Real user monitoring • Real‐user monitoring Web performance  • Web performance • Functionality backbone load • Last Mile monitoring analytics • Performance • Real‐world  • Backbone monitoring • Benchmarks Last Mile load • Business dashboard Web 2.0, RIA, Streaming, Mobile  Web 2 0 RIA Streaming Mobile Dashboards,  Recording &  Alerting &  APIs & Education & Metrics & Analytics Provisioning Diagnostics Data Feeds  Best Practices World’s Most Comprehensive Testing Network 168+ countries    2,500+ ISPs 500+  100+  5,000+  100,000+  combos of  commercial nodes  supported  consumer‐grade  browsers & O/S and data centers mobile devices desktops
  • 37. Gomez Customers Enjoy Measurable Benefits Increased conversions 10%  Reduced homepage load time from  11.3 seconds to 3 4 seconds 11 3 d t 3.4 d Improved page load times 23% Saved 50%+ in staff and fees Reduced seven‐step  transaction time by 50% y Reduced downtime 45%  Achieved under 3 second  Achieved under 3 second response time and 99%+  availability Validated decision to consolidate  Validated decision to consolidate three data centers
  • 39. Ensuring Performance Of Login Process Company: Online presence for a popular TV show • Following episodes of the TV show the web site sees high traffic spikes • Goal was to achieve 1500 logins per minute • Load tested DB to improve performance in anticipation of another traffic spike 3rd Party/ y/ Browsers  Cloud Services Local ISP and devices Users Load  Balancing 1 Web  Servers App  Servers DB  Internet Servers Major ISP Storage Mobile  Components Content Delivery Mobile  Networks Carrier
  • 40. Application Bottleneck Causes Immediate Response Time Issue • As users were added the added, response time of step 3 (the login) climbed immediately • The test bottlenecked at 160 logins per minute (Goal 1500) • But quickly dropped off as users received server errors • New login query was not optimized and was bottlenecking the database servers’ CPU ’ CPUs