FaaScinating Resilience for Serverless Function Choreographies in Federated Clouds

Cloud applications often benefit from deployment on serverless technology Function-as-a-Service (FaaS), which may instantly spawn numerous functions and charges users for the period when are running. Maximum is achieved orchestrated in a workflow or function choreographies (FCs). However, many provider limitations specific FaaS, such as maximum concurrency duration increase failure rate, can severely hamper execution of entire FCs. Current support resilience limited to retries try-catch, applicable within same cloud region only. To overcome these limitations, we introduce rAFCL, middleware platform that maintains reliability complex FCs federated clouds. In order resilient FC under our model creates an alternative strategy each based required availability specified by user. Alternative strategies not restricted region, but contain across five providers, invoked concurrently single plan executed subsequently multiple plans. With this approach, rAFCL offers flexibility terms cost-performance trade-off. We evaluated running three real-life providers. Experimental results demonstrated outperforms AWS Step Functions, increasing success rate 53.45%, while invoking only 3.94% more with zero wasted invocations. significantly improves almost 1 survives even after massive failures functions.