CO-DMO-CT (Co-optimization of DVS, Memory placement, and Offloading with compatible TEE)

This project implements real-time task scheduling using steady-state genetic algorithms to minimize power consumption in CPU, memory, and network subsystems while meeting deadline constraints.

CO-DMO-CT integrates three energy-saving techniques across different system layers:

• DVS (Dynamic Voltage Scaling): Adjusting voltage and frequency to reduce CPU power consumption.

• Persistent Memory Placement: Optimizing data placement to minimize memory energy usage.

• Task Offloading: Executing tasks on edge servers that support a Trusted Execution Environment (TEE) to ensure secure execution.

Two executables included in this project, which can simulate CO-DMO-CT in comparison with DVS, Offloading, and basic configurations.

• gasgen: task generation tool based on CPU and total utilization
• gastask: scheduling scheme generator based on GA

Build

To build gastask and gasgen, use CMake:

$ mkdir -p build && cd build
$ cmake ..
$ make

Run

• Create a new configuration file. Refer to gastask.conf.tmpl.
• run gasgen

# ./gasgen gastask.conf

• Tasks list will be generated into task_generated.txt network_generated.txt network_commander_generated.txtaccording to gastask.conf
• paste task_generated.txt into the task section of gastask.conf
• paste network_generated.txt into the network section of gastask.conf
• paste network_commander_generated.txt into the net_commander_ section of gastask.conf
• run gastask

# ./gastask gastask.conf

• scheduling information is generated in task.txt, which can be used as an input to simrts.

CO-DMO-CT: IIoT Application Offloading with AWS Nitro Enclaves

This project demonstrates how to securely offload Industrial IoT (IIoT) application tasks to a Trusted Execution Environment (TEE) using AWS Nitro Enclaves.

System Overview

• IIoT Device: Simulates a constrained edge device generating tasks that can be either locally executed or offloaded.
• Gateway: Receives IIoT tasks and forwards them to the enclave.
• Enclave: Executes the offloaded computation securely within the Nitro Enclave. Currently, the enclave implementation simply prints a basic message to demonstrate successful offloading.

---

1. Prepare IIoT and Enclave Code

• Modify iiot_device.c for IIoT task logic. This code can execute either locally or be offloaded to the enclave, depending on system configuration.
• Implement the offloaded logic in enclave.c. The same computation code may exist in both components, enabling flexible offloading.
• Add any required libraries to CMakeLists.txt.

---

2. Build the System Components

$ mkdir build
$ cd build
$ cmake ..
$ make

This builds the following binaries:

• iiot_device: IIoT device emulator that generates tasks.
• gateway: Forwards tasks from the IIoT device to the enclave.
• enclave: Executes offloaded tasks inside the TEE (Nitro Enclave).

---

3. Build the Enclave Image

$ ./build_enclave.sh

This script generates the Enclave Image File (EIF) using a Docker-based build process:

1. A Docker container is created based on the provided Dockerfile.ne, which defines the enclave runtime environment.
2. The Docker image includes the following required files:
   • libnsm.so: The Nitro Secure Module library for enclave-host communication.
   • enclave: The compiled binary that executes within the enclave.
   • bootstrap.sh: An entrypoint script for enclave execution.
3. The container is then converted into an EIF using nitro-cli build-enclave.

Ensure that Dockerfile.ne and all required files are located in the correct context directory when running the build script.

---

4. Launch Gateway and Enclave

In one terminal:

$ ./gateway

In another terminal:

$ ./run_enclave.sh

This script launches the Nitro Enclave using the EIF created earlier.

---

5. Run the IIoT Emulator

$ ./iiot_device -g <gateway_ip>

This simulates IIoT task generation and sends tasks to the gateway for secure offloading.

If successful, the terminal will display logs confirming secure transmission and execution.

---

Notes

• Make sure your EC2 instance supports Nitro Enclaves.
• Ensure required permissions and enclave-enabled AMI settings are configured.
• Refer to AWS Nitro Enclaves documentation for detailed enclave configuration.