The technology, Function-as-a-Service, is a new model that is spreading more and more nowadays, and promises a new architecture, with which any developer can spare money and time. Using this technology means that the developers can focus just to implement apps and leave the work with the infrastructure to the FaaS provider.
But other new innovations can appear in both side of DevOps. If this technology is successfully optimized, then in the future all hardware resources can be used smarter. This can only happen because if these new functions are not used by anyone, then it needs only a minimal level of resources. There are some implementations, which support "cold start". Cold start means, that a computing unit (typically a container) running the code of the function is not available when the function is called, it is initiated only after a function is invoked, therefore no resources are allocated before the initiation.
From the financial perspective there are two separated groups in FaaS. The first group is "commercial serverless services". These solutions are working well, but the user must pay for everything that he uses. Besides is the other group: open source serverless services. Using this is free for the users, but these implementations are often not well documented or not working well.
Hereinafter I want to introduce the basics of FaaS technology. I also explain the software components inside a FaaS platform, and lastly, I will measure selected open source systems.
For my project, I need to examine available measuring tools and study the above-mentioned open source systems. With the selected tools I test the FaaS systems in different measuring environments.
When the tests are done, I reshape my results to a new form to make them readable and comparable for everyone, after that conclusions can be made.