Private Dental Insurance Provider Launches "Quality Improvement" Program To Make Software Service Delivery "Word Class"

Date
05/08/2017
Services
DevOps, Functional Test Automation, Performance Engineering

Challenge:

Customer, a major Boston private dental insurance company had embarked upon a multi-year engagement to make their software service delivery lifecycle world class and state-of-the-art.

The engagement involved numerous areas of focus, including Application Development/Management, Service Delivery, Test Data Management, Functional and Non-Functional Automation/Testing, Release Management (including automated deployment & provisioning), Production Incident Remediation, Production Workload Characterization, and Application Performance Monitoring (APM) in Production.

[read more=”Click here to Read More” less=”Read Less”]

Solution:

In the beginning of December 2015 GQP, LLC (hereafter referred to as GQP) was engaged to aid the client in meeting its goals.  In the beginning, the technology stack was largely Microsoft focused, with development in C#, .NET, Team Foundation Server (TFS), SQL Server, and other supporting technologies.  Very soon after beginning the engagement however, other technologies were introduced, including Oracle, Salesforce, Tableau, Cognos, and Informatica, just to name a few.  GQP was asked to reassess, and make certain that our solution encompassed all technologies and was truly an Enterprise Solution.

GQP architected and implemented a solution in which:

  • Guilds, or Centers Of Excellence for both Performance Engineering & Functional Test Automation where established.
  • The Performance Center Of Excellence acted as a centralized Performance Service Center of Expertise, with a supervisor that managed project resource access in and out of the COE, at the Enterprise level, based upon requirements and needs. To aid the client in achieving their world-class goals, GQP, introduced Neotys’ Enterprise solution, Neoload with Team Server, to share resources across the organization, and to produce accurate results.  Interesting, the client already had the best of bread APM solution, Dynatrace for both Pre-Production and Production in house.  However, the client did not know how to utilize this tool to its full extent.  GQP educated the client in the proper use of all tools.  GQP also mentored and trained the four-member team in the discipline of Performance Engineering extensively over the first year of the engagement.
  • The Automation Center Of Excellence acted as an internal Software Development Group that produced, published, and managed an Enterprise Testing Framework. This framework consists of a set of libraries that encapsulated the intricacies of programming various technologies from database connectivity, integration of tests with TFS and MTM, Data Object Mapping, Web Services, Selenium, etc.  The library was architected by GQP and built using in Visual Studio, C#, Specflow (Cucumber/BDD), and Selenium (using the PageObjects design pattern), and various other open source libraries.  The goal was to build the library to support not only seasoned developers, but junior to intermediate programmers from SQA, whom GQP trained in the basics of C# and Object Oriented Programming (OOP), Behavior Driven Development (BDD), Cucumber, Gherkin, Selenium, and the PageObjects Design Pattern over the 1.5-year length of the engagement.  SQA resources where then embedded into SCRUM Teams, who GQP also trained in the intricacies of BDD, to facilitate the use of the Enterprise Testing Framework in complete Shift-Left, Sprint Success Testing.  The Automation Center Of Excellence also managed and executed the Enterprise Release Regression Suites for all Enterprise Products and executed these tests, reported, and managed them through TFS, MTM, and the Lab Management capabilities of MTM.
  • GQP played an assisting role with Release Management, to ensure automatic Continuous Integration and Delivery Builds, with Automated Deployments, Provisioning, and appropriate testing at each stage (smoke, full regression, etc.).

 

Results:

In conclusion, GQP helped the client achieve the following:

  • The release pipeline was automated and release to Production time was greatly reduced, with unparalleled consistency and reliability compared to pre-engagement. This also fostered trust and dependability into receiving new deployments when needed and contributed significantly to reducing environment dissimilarity, sprawl, and hording.
  • Performance and Automation COEs where established and trained in the appropriate practice capabilities.
  • 12 SCRUM teams where trained in BDD/Gherkin and Selenium/PageObjects Automation.
  • SQA team members were trained in C#, OOP, BDD/Gherkin, and Selenium/PageObjects.
  • The Enterprise Automation Framework is now utilized in all Regression Test Suites in the Enterprise Release Paths, as well as in all SCRUM teams for Sprint Success testing.
  • The Enterprise Regression Suite was increased in volume and breadth for more test coverage by over 200%.
  • The Performance COE now uses Real-World Workloads in Pre-Production load tests, characterized by analysis with Dynatrace.
  • Performance COE better equipped and well trained as Performance Engineers, with expertise in Neoload, Dynatrace, and proper engineering practice, including the use of more scientific methods, and the use of statistical analysis to deliver their results with high confidence, and the trust of their peers throughout the organization.
  • Service Level Agreements (SLA) and Operation Level Agreements (OLA) established from analyzing real-world use in Dynatrace, giving the Performance COE the knowledge to begin Shifting-Left with Non-Functional Requirements (such as Performance of individual components, and sub-systems, etc.).

 

Dynatrace utilized to measure Real-Time User Experience and Business Transactions.  Dynatrace is also utilized in Production for expedited Root Cause Analysis, usually led by the Performance COE.  In addition, Dynatrace is utilized extensively in Pre-Production by the Performance COE to deduce problems before they go into Production.
[/read]


Muli-National Fortune 1000 Semiconductor Companies "World Conference", Delivery by a 3rd-Party Video-Streaming Conference Company

Date
02/09/2017
Services
Performance Engineering

Challenge:

A Fortune 1000, multi-national semiconductor company (hereafter known as the “Client”) has moved to a new implementation of  a 3rd-party video-streaming solution to deliver their quarterly “Worldcasts”.  It had a requirement to ensure that the video solution would be able to successfully deliver content under an expected level of user concurrency to all participating locations throughout North America and Asia Pacific regions.

Each location has an “Media Server” which fronts the 3rd-party cloud-based solution and delivers video to the users at that location. Performance of each “Media Server” needed to be tested under the realistic workload. Traffic needed to be directed from the end-point to the media-server(s) and back, just as if a real human was accessing them.

[read more=”Click here to Read More” less=”Read Less”]

Solution:

The Client engaged GQP through the Neotys Professional Services team, to execute this testing project. GQP/Neotys Professional Services, working with the Client and the 3rd-party:

  • Developed a realistic workload scenario
  • Scripted a custom solution reflecting the Client’s particular 3rd-party solution implementation
  • Lead a joint GQP/Neotys/Client/3rd-party virtual performance testing team to create test infrastructure in all participating locations
  • Ran a series of performance, capacity, and endurance test to verify Client/3rd-party readiness for the first quarterly “Worldcast”
  • Provided performance reports, analysis, and recommendations to the Client/3rd-party

Results:

As a result of this testing, the Client was able to verify readiness of its network to sustain the traffic equal to the expected result (as well as the available additional capacity of 40%). During testing it was determined that the hardware capacity of the Media servers servicing larger population was not sufficient. The Client had proper and timely information that enabled it to increase the needed capacity.  Thus due to performance testing done in time, the first Client video “Worldcast” went smoothly without any problems.  The Client was ready.
[/read]


 

A Major Boston Private Research University Moves It's ERP System To An Open Source Solution

Date
05/08/2017
Services
Performance Engineering

Challenge:

Customer, a major Boston private university has embarked on a multi-year project of transitioning university ERP systems to a modern system based on open source platform and service architecture. Such an ERP system would be used as backbone of university systems for financial processing, course management, event and housing management, and enrollment.

The project had a requirement to introduce performance engineering practice in order to assure that key user and business process transactions would be within service level and operational level agreements.

[read more=”Click here to Read More” less=”Read Less”]

Solution:

In 2015 GQP, LLC was engaged to build for the client application performance management and performance engineering practice for the project. We have been providing such a service till present. Following our best of breed approach we put Dynatrace tools at the technological core of our offering.

 

Using Dynatrace we have:

  • introduced Application Performance Monitoring across the full stack
  • Conducted Architectural Validation of the platform under development
  • Assisted executive team (CTO, Project Leadership) in data driven vendor management using diagnostics from Dynatrace
  • Created Incident Diagnostics and Remediation process and practice
  • Trained Development and Operations team in using Dynatrace for monitoring and incident remediation
  • Provided ongoing performance analysis and reporting

Results:

As a result of GQP building Dynatrace powered application performance management practice the client acquired much needed visibility into the system performance, gained skills in and improved incident troubleshooting and remediation process, and is able to better interact with and control software vendors.
[/read]