Testing Services provides a platform for QA professionals to discuss and gain insights in to the business value delivered by testing, the best practices and processes that drive it and the emergence of new technologies that will shape the future of this profession.

« September 2016 | Main | November 2016 »

October 6, 2016

Testing the Internet of Things Solutions

Author: Tadimeti Srinivasan, Delivery Manager

The Internet of Things (IoT) is a network of physical objects (devices, vehicles, buildings, and other items) that are embedded with electronics, software, sensors, and network connectivity to collect and exchange data.

According to a recent report by McKinsey (source: http://www.mckinsey.com/industries/high-tech/our-insights/the-internet-of-things-sizing-up-the-opportunity ), around 30 billion objects may be connected through IoT by 2020. Enterprises are adopting IoT solutions for the benefits they offer; such as, optimization in operations, reduction in costs, and improvement in efficiency.

Unique characteristics and requirements of IoT systems

Compared to other applications, IoT applications are characterized by several unique factors, such as:                                      

  • Combination of hardware, sensors, connectors, gateways, and application software in a single system
  • Real-time stream analytics and complex event processing 
  • Support for data volume, velocity, variety, and veracity
  • Visualization of large-scale data 

Challenges of IoT Testing

These characteristics consequently present a unique set of challenges when it comes to testing IoT applications. The primary challenges include:

  • Dynamic environment: Unlike application testing performed in a defined environment, IoT has a very dynamic environment with millions of sensors and different devices in conjunction with intelligent software
  • Real-time complexity: IoT applications can have multiple, real-time scenarios and its use cases are extremely complex
  • Scalability of the system: Creating a test environment to assess functionality along with scalability and reliability is challenging
  • Security and data privacy: Potential security vulnerabilities across various layers of IoT system
  • Ability to automate: Due to innumerous scenarios and dependency on hardware (sensors, devices), automation is a challenging process

Infosys IoT Validation solution

Infosys has developed a comprehensive quality assurance (QA) strategy to handle the unique requirements and challenges associated with validating IoT applications. Infosys IoT Validation solution enables testing with a combination of actual devices, simulators, tools and frameworks. Infosys has developed specific tools for IoT validation to perform functional validation, API validation, load testing, and security verification. Infosys IoT test framework can easily integrate with various IoT protocols and platforms, thus providing interoperability.

Benefits to Business

Infosys IoT validation solution helps our customers to Engage, Integrate and Accelerate their IoT solution deployments.

  • Engage: Best-in-class user experience across multiple channels (device, web, mobile) by leveraging Infosys Customer Experience and Analytics tools and design thinking methodologies
  • Integrate: Provide an integrated approach to validate the functional and non-functional requirements of IoT solutions. Future-proofing of business through extensive interoperability, security and performance testing
  • Accelerate: Faster time-to-market by leveraging early automation, Infosys automation tools and re-usable test scripts. Infosys AI Platform for testing to optimize testing and bring in efficiencies

Please read more here.

Infosys view on relevance of Continuous Validation in DevOps Journey

Author: Prashant Burse, AVP-Senior Delivery Manager

Rapid digitization is forcing organizations to develop more and more consumer driven enterprise strategies. This is forcing businesses and IT organizations to respond with increased agility to ever changing consumer needs. IT organizations are hence shifting from traditional waterfall based delivery models to Agile and eventually DevOps methodology.

With decades of development and test work done by IT employees in waterfall model, these organizations are faced with multiple challenges on different fronts:                                                                                                  

  • How do I undertake this waterfall to DevOps methodology/ process transformation journey starting with organization change management
  • How do I address the challenge of reducing my debt across multiple segments
    o   People Debt ( How do I quickly up skill by people)
    o   Process Debt ( How do I quickly optimize if not reduce my process related debt and move from waterfall to Agile)
    o   Technical Debt:
      How do I reduce my coding debt
      How do I reduce my test toolsets debt
      How do I reduce my environments debt and provision them faster

QA teams and organizations in context of Continuous Validation too are coming under pressure to justify their existence and address challenges such as:

  • Cut down longer test cycles
  • Provide increased test coverage in lesser time
  • Skill upliftment : testers with development skills, knowledge of multiple toolsets in context of Continuous Validation

Infosys Continuous Validation (CV) solution addresses the following key considerations:

  • Customized assessment framework to benchmark and systematically plan the maturity of Continuous Validation roadmaps across 5 different levels
  • Infosys Point of View across 20+ different parameters on Multiple toolsets in context of CV across SOA/Performance/functional automation
  • Tried and tested pipelines across different technologies / domains both open sourced and licensed toolsets
  • Infosys DevOps Platform can be leveraged to accelerate building continuous validation pipelines
  • Accelerators to speed up DevOps - CV journeys by test teams 
  • Skill Upliftment: Infosys has developed customized training programs on Continuous Validation for 3 categories - Executives, Managers and practitioners

Execution benefits:

Infosys is actively working to implement DevOps on 200+ programs/ Projects. In one such payments transformation programs of 300 applications with leading US based financial services company, Infosys implemented Agile/ CI-CD pipeline. Automated testing and service virtualization, Automated acceptance and Behavior Driven development using Jbehave was implemented.
Tools used: SVN, GIT, Nexus, Jenkins, Selenium and CA Lisa.
Benefits realized: Time to market was reduced from 3-6 months to 2-4 weeks and 90% defects are captured phases due to increased test coverage.


More info on the offerings here.

October 3, 2016

Evolution of mobile devices and Impact on Performance

Author: Yakub Reddy Gurijala, Senior Technology Architect

In last decade, mobile devices evolved from point to point communication like phone calls and SMS to smart features with advanced OS capable of executing native applications. These change created lot of opportunities and challenges for online businesses and application developers.

With increased connectivity options like 2G/3G/4G/LTE technologies, these mobile devices are rapidly expanding into different areas and share mobile device users is ever expanding when compared to traditional Desktop/Laptop users.  In case of some of the e-commerce applications, number of transactions from Mobile users are exceeding as compared to the other users types. Also, number of mobile users have already exceeded the number of traditional users, globally.

As mobile devices are increasingly adopted by users, it also poses serious challenges to application experience and performance. These devices will come with different configurations related to screen size, memory and cpu.  This is a big challenge in application performance and following factors will have huge impact on performance / user experience, of applications.

Layout/Display:  Mobile devices have different display sizes, with varying screen resolutions. This will make normal web applications not fit the screen as expected in mobile devices.  To display the application properly, the web applications needs to re-adjust for mobiles possibly with new layout for mobiles.  Easy way to do that is to build the web application for the mobiles and later extend it to desktop clients. This will make huge difference in user experience. If you have not tuned the applications for mobiles, most of the mobile users not able to use the application and can lead into huge financial loses. For more better user experience it is suggested adopt the native applications instead of mobile web application.

Payload/Page Sizes: For mobile devices all pages/payload sizes should be optimal and minimal. High payload will have longer download time, consumption of bandwidth to users and higher cpu/memory utilization at client (Device) side.

Client Resources:  Mobile devices comes in different CPU, Memory options and they are very minimal when compared traditional desktop/laptops.  Applications which are designed for mobile devices are to be ensured for optimal usage of memory and CPU.  Heavy client side processing, rich GUI and huge payloads will have huge impact on the performance of mobile users.

  • Typical Desktop Configuration: Intel I Series CPU, 8 GB RAM, 500GB to 1TB HDD, 15/21 Inch Display
  • Typical Mobile Devise Configuration: Single Core Mobile Processors, 1 GB RAM, 16 GB HDD, 4.5 to 5 Inch display

Server Resources: Mobile devices will connect to the server using different networks, limited CPU, limited Memory and different connection speeds (Often varying network conditions).  These connections are generally slow in nature and spend more time to download data from servers, hence consumes more server resources, as each mobile connection is active more time on servers when compared desktop client because of network conditions/CPU/Memory. This necessitates more server resource requirement to serve mobile user when compared similar number of traditional desktop users

Network Conditions: Mobile devices uses different network connection to connect the servers like 2G/3G/4G/LTE/Wi-Fi connections. In cases, connection speed varies (Device is switching between different network conditions), will have an impact on performance and user experience of mobile application.

In today's dynamic world each user is important for the business and application teams needs to make sure both desktop and mobile users experience should be on par with each other. All types of users should be able to use the application with ease and performance should be good for both desktop and mobile users. As such, Performance engineers needs to look mobile user base and ask right questions to create proper Load Model for the performance assessment. Also, Performance engineers should simulate both desktop and mobile users (including geographical and network simulations) together, to measure and baseline the performance of the applications.


October 2, 2016

Shifting extreme left during cloud migration - A case for involving QA teams from day one for cloud migration projects

Posted on behalf of Sheetal Swaroop Payyavula, Research Analyst, Infosys Validation Solutions

Due to the vast benefits being realized, many companies are moving to the cloud. But, a recent survey shows that only 27% of participants were extremely satisfied with their cloud migration experience.

The typical lifecycle of a cloud migration project can look like this.

During the initial analysis and estimation phases, many decisions and discussions are centered on Financial Assessment doing ROI analysis, Security and Compliance Assessment, and Technical Assessment. Decisions are made on type of Cloud - Public, Private, Hybrid; Cloud layer - PaaS, SaaS, IaaS; Selection of cloud vendor etc. Gaps are identified between the current traditional architecture and next-generation cloud architecture.  Every decision taken at this stage can introduce certain risks.

As examples: 1) Selecting one layer of cloud may cut off access to underlying layers, impeding certain types of testing. 2) Tests such as penetration testing may not even be allowed by the cloud vendor or permitted only in liaison with a third party service provider. 3) Companies deploy updates to their softwares/platforms/stacks frequently - sometimes several times a day. Lack of frequent and automated testing may pose risk in certain cloud deployments.

To mitigate these risks, it's imperative that test teams are involved from day one in a cloud migration project. As compared to traditional deployments, testing for cloud requires a lot of additional non-functional test dimensions like Elasticity, Security, Performance, Compatibility, API Integration, Live and Multi-tenancy testing. As there are a myriad of cloud configurations, test planning taking into account all of these dimensions or "a one plan fits all" will be wasteful if extensive and risky if not. There is a clear need to catalog all the risks involved for the project's specific cloud configuration and tailor the project's test strategy to mitigate them.

The Infosys Risk Based Cloud Testing Accelerator is one such tool that helps projects define a customized risk based test strategy by creating a thorough risk profile, and using it to determine the suitable test measures that could help mitigate their cloud related risks. It ensures Comprehensive Risk Assessment and Complete Test Dimensions coverage.

Here is a "shift extreme left" lifecycle of a cloud migration from QA perspective.

To summarize, for a successful cloud migration, shape an ideal risk based cloud test strategy by involving QA teams from day one. QA teams cannot be separate functions in cloud deployments. Test strategy should start during the first stage i.e. assessment stage. Automated tests and optimization should continue long after the successful cloud deployment thus making QA teams presence needed from extreme left to extreme right of a cloud deployment.

Do visit Infosys experts in booth #59 at STARWEST 2016 and engage in insightful conversations around cloud migration and testing. More info on our participation is here. Looking forward to meeting you.

October 1, 2016

Testing of Legacy Modernization using Infosys Big Data framework - A case study

Posted on behalf of Surya Prakash G., Delivery Manager, Infosys Validation Solutions

Many organizations would like to reduce the data being processed on legacy systems to minimize maintenance and support costs. Modernization to an open source big data solution (Ex: Hadoop solution) is an option. Such modernization enables faster time to market and helps leverage big data technology features to handle not just volume, but also velocity and variety.

Infosys was involved for one such migration testing work, where Infosys big data service offering helped client to achieve faster time to market and reduce costs as well.

The client, a leading UK based bank with global presence in Retail, Investment and Commercial banking, wanted to develop an Anti-Money Laundering (AML) solution for its correspondent banks and third party agents and integrate the same with existing NORKOM which covers direct clients of bank. Bank was looking for a flexible architecture that can ingest huge amount of data coming from around 60 million customers in 80 different countries. Testing objective of this complex migration was to validate huge amount of structured data coming from different customers and third party agents in different formats to Hadoop and ensure the integration of Oracle Mantas with Hadoop with enriched data quality.

Challenges for testing were multifold:
End-to-end testing to ensure that the Implementation of modern AML architecture integrates well with an enterprise data hub, Hadoop
• Setting up a stable testing environment
• Testing massive complex and long term descriptive data from different sources and to ensure quality data
• Consolidation and validation of data from multiple sources which are in different formats

To solve the above challenges of huge amount of data and variety of data, Infosys came out with an approach to validate 100% of the data by testing at each stage to cover all the permutations with automated tools as well as manually. Also, since structured & unstructured data was coming from different sources and resides in Hive tables, HOF-Hive output formatter was used to get the pipe delimited files. Data validation was done using automated utilities developed for different stages.

Key benefits included:
Reduction of COQ by 10-15%  through early detection of data issues at source
10% reduction in test execution effort for every execution cycle through automated execution / error monitoring of workflows by triggering corresponding sequence through UNIX scripts
100% test coverage through automated approach for validation of all scenarios
100% compliance to AML through reconciliation of activities across direct clients (NORKOM) and correspondent banks (Oracle Mantas)
I invite you to visit Infosys booth #59 at STARWEST 2016 and meet our experts who can definitely help you out in your big data journey. More info on our participation is here.

A case for mobile accessibility

Posted on behalf of Smitha Sidharthan P., Project Manager, Infosys Validation Solutions

I like to think of my commute to office as a very productive time. So, like my fellow commuters, I spend this glued to my smartphone. I catch up on news, connect with friends and family, do my grocery shopping and check up on my office emails. I also listen to music or podcasts, take the occasional MOOC and sometimes brush up on my Spanish on Duolingo or follow office conversations on Yammer. I even have an app that lets me read a summary of a new non-fiction book every day in under 15 minutes! So I breeze into office feeling quite happy with myself, having ticked off many items from my to-do list.

This is a common scenario that plays out everywhere and with most of us. While the real benefits of such "productivity" are debatable, what is amazing is the revolution that has happened over the last few years making all of this possible. The digital revolution has impacted all areas be it retail, education, entertainment, communication, and employee engagement and has made it possible to access information on the go any time and through different channels.

But has this revolution been truly universal? Does it benefit everyone, irrespective of age and physical ability? Is all this information truly accessible?

The internet and mobile technologies have tremendous potential in impacting the lives of people by providing access to information and resources from anywhere. However, this potential can only be realized if these resources are accessible to all people irrespective of their physical ability.

Consider the smartphone. Increasing adoption of smartphones and proliferation of mobile apps along with improvement in technologies like object recognition has opened up a lot of opportunities for physically impaired people to better navigate the world around them. For example, there are multiple applications available for people with vision impairments that can help in day-to-day activities like identifying objects like currency bills or identifying colors, navigation, seeing better by magnifying objects etc.

Mobile manufacturers are also increasingly providing features to make these devices accessible to people with vision, hearing, motor or cognitive disabilities. Let's take a look at the iOS and Android platforms. For vision impairment, they provide inbuilt screen readers (VoiceOver for iOS and TalkBack for Android), digital assistants and voice recognition technology that lets you perform activities through voice, ability to magnify portions of the screen, resize text and support for braille displays. For hearing, we have support for closed captions, mono audio, ability to set up visual instead of audio alerts, etc. People with motor disabilities have the option to customize gestures, support for hardware keyboards and switch access. While the above is not an exhaustive list of assistive features available today, these represent significant improvements in terms of mobile accessibility.

Do these guarantee that all iOS or Android apps are accessible? Not necessarily. They can be accessible only if developed so as to utilize the accessibility features of the operating system. For example, if the developer has not provided a meaningful label for a button, the screen reader will not be able to communicate the purpose of the button to the user. Similarly an image which does not have a text description will not be accessible to a user using a screen reader. Hence it is extremely important to perform an accessibility assessment to identify any issues from the user perspective. Accessibility testers need to understand the needs of differently-abled people as well as accessibility barriers inherent to the mobile platform. They also need to understand any country specific legal requirements related to accessibility that need to be addressed. Automated accessibility scans supplemented by manual user testing can help in ensuring that apps that we build are truly accessible to all users.

Building accessible applications has the added advantage that they tend to be more usable for everyone in general. Hence, focusing on accessibility can improve the overall customer experience and improve the adoption of these applications. 

I for one, find it very useful to turn captions on while listening to a video lecture, especially if the professor's accent is new to me or if I am passing through noisy traffic. Ending phone calls using the Power button is another accessibility feature that was useful when the Touch screen UI was not working properly. And would it not be good to get my shopping done using Voice while I stare at the sky outside! Well, I have added an item to cart but the screen reader says 0 items in cart while I hover over the shopping cart icon. Some other day then, or may be should just try a more accessible app!

Infosys team of experts would be available in booth #59 at STARWEST 2016 to discuss more on accessibility testing and other offerings. More info on our participation is here. Do drop in!

Subscribe to this blog's feed

Follow us on

Infosys on Twitter