Hot Under the Hood: An Analysis of Ambient Temperature Impact on Heterogeneous Edge Platforms
Authors: 
Amirhossein Ahmadi, Hazem Abdelhafez, Shashwat Jaiswal, 
Karthik Pattabiraman, Matei RipeanuAuthors Info & Claims
Amirhossein Ahmadi, Hazem Abdelhafez, Shashwat Jaiswal, Karthik Pattabiraman, Matei RipeanuAuthors Info & ClaimsAbstract
Applications deployed at the edge are often subject to critical Quality of Service (QoS) objectives, such as meeting deadlines while optimizing for energy consumption. To design and operate middleware that satisfies these QoS objectives, it is crucial to understand the runtime and power consumption characteristics of the edge platform. However, while edge platforms are frequently deployed in environments where ambient factors cannot be controlled, most characterizations are performed without considering environmental factors. We characterize the impact of ambient temperature on the power consumption and runtime of machine learning inference applications running on a popular edge platform, the NVIDIA Jetson TX2. Our rigorous data collection and statistical methodology reveals a sizeable ambient temperature impact on power consumption (about 20% on average, and up to 40% on some workloads) and a moderate impact on runtime (up to 5%).
References
[1]
S. Bateni and C. Liu. 2020. Neuos: A latency-predictable multi-dimensional optimization framework for dnn-driven autonomous systems. In USENIX ATC. 371--385.
[2]
B. Cox and J. Galjaard et al. 2021. Masa: Responsive Multi-DNN Inference on the Edge. In IEEE PerCom.
[3]
C. Curtsinger and E. Berger. 2013. STABILIZER: Statistically Sound Performance Evaluation. SIGARCH Comp. Arch. News, 219--228.
[4]
A. Maricq et al. 2018. Taming Performance Variability. In USENIX OSDI. 409--425.
[5]
C. Imes et al. 2015. POET: a portable approach to minimizing energy under soft real-time constraints. In IEEE RTAS. 75--86.
[6]
M. Han et al. 2019. MOSAIC: Heterogeneity-, Communication-, and Constraint-Aware Model Slicing and Execution for Accurate and Efficient Inference. In IEEE PACT. 165--177.
[7]
N. Razali et al. 2011. Power comparisons of shapiro-wilk, kolmogorov-smirnov. lilliefors and anderson-darling tests. Journal of statistical modeling and analytics, 21--33.
[8]
P. Chestovich et al. 2022. Temperature Profiles Of Sunlight-Exposed Surfaces In A Desert Climate: Determining The Risk For Pavement Burns. Journal of Burn Care & Research: Official Publication of the American Burn Association, irac136--irac136.
[9]
S. Bateni et al. 2018. PredJoule: A Timing-Predictable Energy Optimization Framework for Deep Neural Networks. In IEEE RTSS. 107--118.
[10]
S. Han et al. 2016. Mcdnn: An approximation-based execution framework for deep stream processing under resource constraints. In Mobile Systems, Applications, and Services Conference. 123--136.
[11]
S. Maity et al. 2021. Thermal-aware adaptive platform management for heterogeneous embedded systems. ACM TECS, 1--28.
[12]
T. Baruah et al. 2018. Airavat: Improving energy efficiency of heterogeneous applications. In DATE Conference. 731--736.
[13]
T. Mytkowicz et al. 2009. Producing wrong data without doing anything obviously wrong! ACM Sigplan Notices, 265--276.
[14]
A. Karyakin and K. Salem. 2017. An analysis of memory power consumption in database systems. In Workshop on Data Management on New Hardware. 1--9.
[15]
V. Lakshminarayanan and N. Sriraam. 2014. The effect of temperature on the reliability of electronic components. In IEEE CONECCT. 1--6.
[16]
Y. Lee. 2021. Thermal-Aware Design and Management of Embedded Real-Time Systems. In DATE Conference. 1252--1255.
Index Terms
- Hot Under the Hood: An Analysis of Ambient Temperature Impact on Heterogeneous Edge Platforms
Recommendations
Snowflakes at the Edge: A Study of Variability among NVIDIA Jetson AGX Xavier Boards
EdgeSys '21: Proceedings of the 4th International Workshop on Edge Systems, Analytics and NetworkingWhile applications deployed at the edge often rely on performance stability (or, at a minimum, on a predictable level of performance), variability at the edge remains a real problem [4]. This study uncovers a surprising source of variability: intrinsic ...
On the Programmability and Performance of Heterogeneous Platforms
ICPADS '13: Proceedings of the 2013 International Conference on Parallel and Distributed SystemsGeneral-purpose computing on an ever-broadening array of parallel devices has led to an increasingly complex and multi-dimensional landscape with respect to programmability and performance optimization. The growing diversity of parallel architectures ...
Transparent Acceleration for Heterogeneous Platforms With Compilation to OpenCL
Multi-accelerator platforms combine CPUs and different accelerator architectures within a single compute node. Such systems are capable of processing parallel workloads very efficiently while being more energy efficient than regular systems consisting ...
Comments
Information & Contributors
Information
Published In
May 2023
66 pages
ISBN:9798400700828
DOI:10.1145/3578354
- Co-chairs:
- Aaron Ding,
- Ella Peltonen,
- Volker Hilt,
- Publications Chair:
- Atakan Aral
Copyright © 2023 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 08 May 2023
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
EdgeSys '23
Sponsor:
EdgeSys '23: 6th International Workshop on Edge Systems, Analytics and Networking
May 8, 2023
Rome, Italy
Acceptance Rates
Overall Acceptance Rate 10 of 23 submissions, 43%
Upcoming Conference
EuroSys '25
- Sponsor:
- sigops
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 170Total Downloads
- Downloads (Last 12 months)70
- Downloads (Last 6 weeks)5
Reflects downloads up to 02 Feb 2025
Other Metrics
Citations
Cited By
View allView Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in