PQS is an online retail shop in Sydney, Australia. Its website, where the users can explore different products and promotions and place orders, has more than 100,000 visitors per month. Any downtime experienced by the website can result in significant lost opportunities. Thus, any change made to website, which runs on co-located data centres, must be very carefully coordinated. After lots of paper works and lengthy process, only operation engineers are allowed to deploy changes into the production infrastructure. Hence, new releases may happen once in every month. At PQS, 40 developers work in different teams. Each developer has her/his own laptop which runs Windows 10. The development environment, which includes various libraries and toolchains, in each laptop is setup individually. The developers make changes in hundreds of branches of multiple repositories. Clearly, before the release, all those branches need to be merged. Any changes to the website need to pass through the QA before they are considered ready for deployment in the production system. However, QA build and test environments are much smaller in size than that of production. There are four servers for QA build and test. Since developers also have direct SSH access into these servers, many of them have been modified during releases to address hotfixes and other urgent needs. The QA team is responsible for creating various tests and ensuring that website passes all those tests before being deployed into the production system. Developers can choose to run tests, but they often skip testing since testing takes a long time and tend to pass locally but fail in QA environment. PQS realizes that this is inefficient. They want to ship new features faster and more reliably. How can DevOps fix this? Make a recommendation with justification on whether to keep the production infrastructure on-premise or to move into the Cloud, such as, AWS or Azure
PQS is an online retail shop in Sydney, Australia. Its website, where the users can explore different products and promotions and place orders, has more than 100,000 visitors per month. Any downtime experienced by the website can result in significant lost opportunities. Thus, any change made to website, which runs on co-located data centres, must be very carefully coordinated. After lots of paper works and lengthy process, only operation engineers are allowed to deploy changes into the production infrastructure. Hence, new releases may happen once in every month. At PQS, 40 developers work in different teams. Each developer has her/his own laptop which runs Windows 10. The development environment, which includes various libraries and toolchains, in each laptop is setup individually. The developers make changes in hundreds of branches of multiple repositories. Clearly, before the release, all those branches need to be merged. Any changes to the website need to pass through the QA before they are considered ready for deployment in the production system. However, QA build and test environments are much smaller in size than that of production. There are four servers for QA build and test. Since developers also have direct SSH access into these servers, many of them have been modified during releases to address hotfixes and other urgent needs. The QA team is responsible for creating various tests and ensuring that website passes all those tests before being deployed into the production system. Developers can choose to run tests, but they often skip testing since testing takes a long time and tend to pass locally but fail in QA environment. PQS realizes that this is inefficient. They want to ship new features faster and more reliably. How can DevOps fix this?
Make a recommendation with justification on whether to keep the production infrastructure on-premise or to move into the Cloud, such as, AWS or Azure
Step by step
Solved in 3 steps
