General-purpose graphics processing units (GPUs) made available on embedded platforms have gained much interest in real-time cyber-physical systems. Despite the fact that GPUs generally outperform CPUs on many compute-intensive tasks in a multitasking environment, high power consumption remains a challenging problem. In this paper, we first analyze the power and energy consumption of GPU kernels scheduled with spatial multitasking, which is found to be advantageous for schedulability in recent studies, and prove that its use, however, degrades energy efficiency even in the latest commercially available embedded GPUs like NVIDIA Jetson Xavier AGX. Then, based on our observations, we propose sBEET, a real-time energy-efficient GPU scheduling framework that makes scheduling decisions at runtime to optimize the energy consumption while utilizing spatial multitasking to improve real-time performance. We evaluate the performance of the proposed sBEET framework using well-known GPU benchmarks and randomly-generated timing parameters on real hardware. The results indicate that sBEET reduces deadline misses up to 13% when the system is overloaded, and also achieves 15% to 21% lower energy consumption when the tasksets are schedulable compared to the existing works.

Balancing Energy Efficiency and Real-Time Performance in GPU Scheduling

With the rising demand for emerging DNN ap- plications in safety-critical systems, much attention has been given to the reliability and trustworthiness of DNN inference output against malicious attacks. Although prior work has been conducted to improve the privacy of DNN inference by executing the entire DNN model inside Intel SGX enclaves, ex- isting approaches pose severe performance challenges to achieve dependable and timely execution simultaneously. In this paper, we propose AegisDNN, a DNN inference framework to address this problem. AegisDNN leverages secure SGX enclaves for protecting only the critical part of real-time DNN tasks which are vulnerable to potential fault injection attacks. To choose the right set of layers for protection while ensuring the timeliness of task execution, AegisDNN includes a dynamic-programming based algorithm that finds a layer protection configuration for each task to meet the real-time and dependability requirements based on the layer-wise DNN time and SDC (Silent Data Cor- ruption) profiling mechanism. AegisDNN also utilizes a machine- learning based SDC prediction method to significantly reduce the time for estimating SDC rates for all possible layer protection configurations. We implemented AegisDNN on Caffe, PyTorch, and Tensorflow with Eigen BLAS ported into SGX enclaves to comprehensively demonstrate the effectiveness of AegisDNN against state-of-the-art DNN fault-injection attacks. Experiment results indicate that AegisDNN could satisfy both dependability and real-time requirements simultaneously, when none of the other compared approaches could do so.

AegisDNN: Dependable and Timely Execution of DNN Tasks with SGX

 **Session chair:** Kuan-Hsun Chen<br><br>

- 25 min: [**AI Meets Real-Time: Addressing Real-World Complexities in Graph Response-Time Analysis**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a082/280200a082.pdf), *Sergey Voronov, Stephen Tang, Tanya Amert and Jim Anderson*
- 25 min: [**TimeWall: Enabling Time Partitioning for Real-Time Multicore+Accelerator Platforms (Artifact Evaluated)**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a455/280200a455.pdf), *Tanya Amert, Zelin Tong, Sergey Voronov, Joshua Bakita, F. Donelson Smith and Jim Anderson*
- 25 min: [**Balancing Energy Efficiency and Real-Time Performance in GPU Scheduling**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a110/280200a110.pdf), *Yidi Wang, Mohsen Karimi, Yecheng Xiang and Hyoseung Kim*
- 5 min: [**(WiP) Evaluating Task Dropping Strategies for Overloaded Real-Time Systems**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a528/280200a528.pdf), *Yiqin Gao, Guillaume Pallez, Yves Robert and Frédéric Vivien*
- 5 min: [**(WiP) Cooling by Core-idling: Thermal-aware Thread Scheduling for Mobile Multicore Processor**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a520/280200a520.pdf), *Srijeeta Maity, Anirban Ghose, Soumyajit Dey, Sangyoung Park and Samarjit Chakraborty*
- 5 min: [**(WiP) Cloud Computing for Time-Triggered Safety-Critical Systems**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a516/280200a516.pdf), *Gautam Gala, Javier Castillo Rivera and Gerhard Fohler*
- 5 min: [**(WaP) Integrated Scheduling of Real-time and Interactive Tasks for Configurable Industrial Systems**](https://tu-dortmund.sciebo.de/s/NrQjhhzs3ynTlmz), *Suhyeon Yoo, Yewon Jo and Hyokyung Bahn*




4.21 - Heterogeneous Computing + 4 BPs

technical paper

 **RTSS 2021** has successfully concluded. Hope you have enjoyed it, and look forward to seeing you next year in person.
    Announcement for the Awards:
 * Best Paper Award goes to: Monte Carlo Response-Time Analysis, Sergey Bozhko, Georg von der Brüggen, and Björn Brandenburg
 * Best Student Paper Award goes to: Automatic Energy-Hotspot Detection and Elimination in Real-Time Deeply Embedded Systems, Mohsen Shekarisaz, Lothar Thiele, and Mehdi Kargahi.
 * Best Solution Award for Industry Session goes to: Toward Real-Time Guaranteed Scheduling for Autonomous Driving Systems. Jinghao Sun, Tianyi Wang, Kailu Duan, Bin Lu, Jiankang Ren, Zhishan Guo, and Guozhen Tan.



IEEE RTSS 2021

The IEEE Real-Time Systems Symposium (RTSS) is the premier conference in the field of real-time systems, and is a venue for researchers and practitioners to showcase innovations covering all aspects of real-time systems, including theory, design, analysis, implementation, evaluation, and experience. 

**Session chair:** Luis Almeida

- 25 min: [**Vulnerability of Controller Area Network to Schedule-Based Attacks**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a495/280200a495.pdf), *Sena Hounsinou, Mark Stidd, Uchenna Ezeobi, Habeeb Olufowobi, Mitra Nasri and Gedare Bloom*
- 25 min: [**AegisDNN: Dependable and Timely Execution of DNN Tasks with SGX**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a068/280200a068.pdf), *Yecheng Xiang, Yidi Wang, Hyunjong Choi, Mohsen Karimi and Hyoseung Kim*
- 25 min: [**Event-Triggered and Time-Triggered Duration Calculus for Model-Free Reinforcement Learning (Artifact Evaluated)**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a240/280200a240.pdf), *Kalyani Dole, John Komp, Ashutosh Gupta, Krishna S. and Ashutosh Trivedi*
- 5 min: [**(WiP) Enabling Secure Boot for Real-Time Restart-Based Cyber-Physical Systems**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a524/280200a524.pdf), *Sena Hounsinou, Vijay Kumar Banerjee, Chunhao Peng, Monowar Hasan and Gedare Bloom*
- 5 min: [**(WiP) Improving Resilience of Distributed Real-Time Applications via Security and Fault Tolerance Co-Design**](https://conferences.computer.org/rtsspub/pdfs/RTSS2021-23beZleJJDdiozCiQ9rzXD/280200a536/280200a536.pdf), *Wei Jiang, Xinke Liao, Jinyu Zhan and Ke Jiang*



Yidi Wang

2

SHORT BIO

Presentations

Balancing Energy Efficiency and Real-Time Performance in GPU Scheduling

AegisDNN: Dependable and Timely Execution of DNN Tasks with SGX

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES