TLess Experiments

This repository hosts the experiments for the TLess project, a system design for confidential serverless workflows.

We implement TLess on top of two confidential FaaS runtimes representative of two points in the design space for confidential serverless:

• Faasm + SGX: a port of the Faasm to run WASM sandboxes inside SGX.
• CC-Knative: a port of the Knative runtime to run Knative services as container functions inside confidential VMs (AMD SEV).

To execute any code snippet in this repository, we will assume that you have activated your virtual environment:

source ./bin/workon.sh

Pre-requisites

Install rust and rust-analyzer. Then rustup component add rust-analyzer.

Baselines

TLess currently supports being deployed on top of two serverless runtimes, Faasm and Knative.

For instructions to deploy each one of them, see:

• Deploying on top of Faasm
• Deploying on top of Knative

Workflows

This repository implements four different workflows:

• FINRA - Based on the AWS FINRA case study.
• ML Training - Ported from Orion and RMMap.
• ML Inference - Ported from RMMap.
• Word Count - Ported from the MapReduce example in the FunctionBench paper.

Progress Summary

Workflow\Baseline	Faasm	SGX-Faasm	TLess-Faasm	Knative	CC-Knative	TLess-Knative
FINRA	✅	✅	✅	✅	✅	✅
ML Training	✅	✅	✅	✅	✅	✅
ML Inference	✅	✅	✅	✅	✅	✅
Word Count	✅	✅	✅	✅	✅	✅

Experiments

We run the following experiments:

• End-to-end latency: measures the end-to-end execution latency for each workflow.