Digital Repository

Code Driven Vendor Neural Packaging For Serverless

Show simple item record

dc.contributor.author Subashana, Anjana Supun
dc.date.accessioned 2021-07-27T02:38:14Z
dc.date.available 2021-07-27T02:38:14Z
dc.date.issued 2020
dc.identifier.citation Subhasana, Anjana Supun (2020) Code Driven Vendor Neural Packaging For Serverless, BEng. Dissertation Informatics Institute of Technology en_US
dc.identifier.other 2016406
dc.identifier.uri http://dlib.iit.ac.lk/xmlui/handle/123456789/603
dc.description.abstract Cloud computing enables developers and organizations to offload infrastructure related workloads such as maintaining, scaling, provisioning the servers to cloud providers. This model helped organizations to focus more on their business logic and it significantly reduces the cost, increases reliability and accessibility of the servers. Cloud computing has further evolved into a model that completely abstracts servers from the developers. As a result, cloud providers introduced serverless architecture allowing developers to focus only on writing code without worrying about any infrastructure. Serverless architecture offers significant advantages for the developers. This includes increased development velocity, seamless scaling including zero to scale, pay per resource usage and managed servers. Serverless architecture has revolutionized the way applications are built but being vendor locked is a critical issue for the developers and serverless adoption. Most of the solutions for addressing vendor lock-in issues are open source serveries runtimes. Majority of the serverless developers prefers vendor based serveries solutions over open source runtimes due to reasons such as increased cost, lower performance despite being locked into a vendor. This research proposes a developer centric approach to solve the vendor lock issue by introducing a vendor neutral definition language. It allows developers to programmatically define serverless functions and orchestrations using serverless concepts. The defined code will be transformed into cloud specific code in the build time by the compiler. This approach decouples the serverless code from cloud vendors and enables portability between cloud providers. en_US
dc.subject Serverless Computing en_US
dc.subject Cloud computing en_US
dc.title Code Driven Vendor Neural Packaging For Serverless en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search


Advanced Search

Browse

My Account