Colony: Parallel Functions as a Service on the Cloud-Edge Continuum

Abstract

Although smart devices markets are increasing their sales figures; their computing capabilities are not sufficient to provide goodenough-quality services. This paper proposes a solution to organize the devices within the Cloud-Edge Continuum in such a way that each one, as an autonomous individual -Agent-, processes events/data on its embedded compute resources while offering its computing capacity to the rest of the infrastructure in a Function-as-a-Service manner. Unlike other FaaS solutions, the described approach proposes to transparently convert the logic of such functions into task-based workflows backing on task-based programming models; thus, agents hosting the execution of the method generate the corresponding workflow and offloading part of the workload onto other agents to improve the overall service performance. On our prototype, the function-to-workflow transformation is performed by COMPSs; thus, developers can efficiently code applications of any of the three envisaged computing scenarios -sense-process-actuate, streaming and batch processing -throughout the whole Cloud-Edge Continuum without struggling with different frameworks specifically designed for each of them.

Securing the Execution of ML Workflows across the Compute Continua

“…The second scenario, called sense-process-actuate, considers that computation is triggered by the infrastructure when sensing an event and the IT platform is expected to process it and provide an appropriate response even activating some of the actuators to produce some physical effect. This is usually handled by submitting a Machine Learning inference operation to a Function-as-a-Service provider [1,5,11,16].…”

“…In general, the currently available software development solutions target one of these scenarios and focus on tackling the specific challenges set out by the scenario they deal with; however, this technological heterogeneity incurs the adoption of a wide range of programming frameworks and models that hinder the efficient development of solutions targeting complete solutions of the continuum. To the best of our knowledge, Colony [16] is the only framework able to efficiently support the three scenarios with one single programming model thanks to its native support of the COMPSs programming model [17,19].…”

“…COMPSs/PyCOMPSs is a programming framework aiming at easing the development of general-purpose applications targeting the Cloud-Edge-IoT Continuum [16]. The core of the framework is its task-based programming model [17] which provides a sequential fashion of programming that allows its runtime toolkit to orchestrate its execution on top of any distributed infrastructure.…”

Self-balancing architectures based on liquid functions across computing continuums

“…Only few approaches for function execution in continuums have been proposed before, for instance with Delta, an intelligent function scheduler based on predictors and the funcX platform that can lead to 50%-80% execution time reduction [8]. Similar approaches are Pilot-Edge, characterised by support for heterogeneous workloads and decoupled resource management [10], and Colony, supporting three edge-cloud computing scenarios for work ows based on FaaS agents [9]. Osmotic computing has also been proposed to connect edges with data centres seamlessly [13].…”