Projects

SAAF supports profiling Function-as-a-Service (FaaS) workload performance, resource utilization, and infrastructure to enable accurate performance and cost characterizations of software deployments. SAAF provides tools including the FaaSRunner, ReportGenerator, function deployment tools, and the all-in-one FaaSET Juypter notebook to enable profiling FaaS software deployments on the AWS Lambda, Google Cloud Functions, IBM Cloud Functions, Azure Functions, OpenFaaS, and Apache OpenWhisk FaaS platforms written in Java, Python, Node.js, Go, and BASH.



The ContainerProfiler provides a Linux tool to capture resource utilization profiles of container-based software at the VM, Container, and process level. Time series sampling and graphical visualization is supported to support performance analysis and identification of performance bottlenecks.



In this project, we provide client and server-side software to automate parallel execution of CPU, disk, and network performance benchmarks across ec2 cloud VMs. These benchmarks can be deployed on isolated ec2 dedicated hosts (physical servers) to characterize performance degradation from co-located VMs. Benchmark performance data can be leveraged as features to train machine learning models to infer the number of co-located VM instances on IaaS clouds.



The BioDepot-workflow-builder (Bwb) is a software tool enabling users to create and execute reproducible bioinformatics workflows using a drag-and-drop interface where graphical widgets represent Docker containers executing a modular task. Widgets are linked graphically to build bioinformatics workflows that can be reproducibly deployed across different local and cloud platforms.