Benchmark Application	Language	Function Dependency Graph	Summary	Original Source
ImageProcessing	Python		[Workflow that performs image processing on the input image.]	FunctionBench
Text2SpeechCensoring	Python		[Workflow that turns short text segments into speech and censors any profanities within the text segment.]	Eismann et al.
RegressionTuning	Python		A regression problem using Keras, where we try to predict a numerical target y based on a numerical feature x.	no-ops-machine-learning(Jacopo Tagliabue)
VideoAnalytics	Python		Performs object recognition on images generated from a video stream.	vSwarm
DNAVisualization	Python		Generates an image from DNA file.	DNAvisualization.org

The benchmarks have been implemented in four primary categories:

1. All five benchmarks are supported for UnFaaSener with minor modifications to their base codes and are located in their respective subdirectories. In these benchmarks, you need to replace *** with your Google Cloud Project ID.
2. All benchmarks have also been implemented for Google Cloud Workflows to facilitate comparison with UnFaaSener's results. You can find these benchmarks under the gcp_workflows directory. For more information on GCP Workflows YAML parameters, please refer to this guide that assists you in setting up your Google Workflows environments using our provided code.
3. AWS SNS (Simple Notification Service)'s implementation supports ImageProcessing, RegressionTuning, and VideoAnalytics benchmarks. You can find these under the aws_SNS directory.
4. Finally, under aws_step_functions directory you can find the implementation of ImageProcessing, RegressionTuning, VideoAnalytics, and DNAVisualization benchmarks adapted for AWS Step Functions.

The memory configuration of the functions used in our benchmarks is included here. For more information on the contents of each benchmark, please refer to their respective directories.