Skip to content

Commit

Permalink
Update Sensys.md
Browse files Browse the repository at this point in the history
  • Loading branch information
zahidurtalukder authored Jun 24, 2024
1 parent 7babed4 commit cafb1b2
Showing 1 changed file with 6 additions and 7 deletions.
13 changes: 6 additions & 7 deletions _publications/Sensys.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,10 @@
title: "Enabling Low-Cost Server-Level Power Monitoring in Data Centers Using Conducted EMI"
collection: publications
permalink: /publication/SENSYS
excerpt: 'Proposed an auto weighted aggregation techniques for FL'
date: November 2023
venue: 'SenSys '23: Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems'
paperurl: 'https://dl.acm.org/doi/abs/10.1145/3625687.3625801'
citation: 'Gupta, Pranjol Sen and Talukder, Zahidur and Abir, Tasnim Azad and Nguyen, Phuc and Islam, Mohammad A., "Enabling Low-Cost Server-Level Power Monitoring in Data Centers Using Conducted EMI," Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems, 2024, pp. 237–250, doi: 10.1145/3625687.3625801.'
excerpt: "Proposed an auto weighted aggregation technique for FL"
date: 2023-11-01
venue: "SenSys '23: Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems"
paperurl: "https://dl.acm.org/doi/abs/10.1145/3625687.3625801"
citation: 'Gupta, Pranjol Sen and Talukder, Zahidur and Abir, Tasnim Azad and Nguyen, Phuc and Islam, Mohammad A., "Enabling Low-Cost Server-Level Power Monitoring in Data Centers Using Conducted EMI," Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems, 2023, pp. 237–250, doi: 10.1145/3625687.3625801.'
---
Abstract:
Server-level power monitoring in data centers can significantly contribute to its efficient management. Nevertheless, due to the cost of a dedicated power meter for each server, most data center power management only focuses on UPS or cluster-level power monitoring. In this paper, we propose a low-cost novel power monitoring approach that uses only one sensor to extract power consumption information of all servers. We utilize the conducted electromagnetic interference (EMI) of server power supplies to measure their power consumption from non-intrusive single-point voltage measurements. We present a theoretical characterization of conducted EMI generation in server power supply and its propagation through the data center power network. Using a set of ten commercial-grade servers (six Dell PowerEdge and four Lenovo ThinkSystems), we demonstrate that our approach can estimate each server's power consumption with less than ~7\% mean absolute error.
Abstract: Server-level power monitoring in data centers can significantly contribute to its efficient management. Nevertheless, due to the cost of a dedicated power meter for each server, most data center power management only focuses on UPS or cluster-level power monitoring. In this paper, we propose a low-cost novel power monitoring approach that uses only one sensor to extract power consumption information of all servers. We utilize the conducted electromagnetic interference (EMI) of server power supplies to measure their power consumption from non-intrusive single-point voltage measurements. We present a theoretical characterization of conducted EMI generation in server power supply and its propagation through the data center power network. Using a set of ten commercial-grade servers (six Dell PowerEdge and four Lenovo ThinkSystems), we demonstrate that our approach can estimate each server's power consumption with less than ~7% mean absolute error.

0 comments on commit cafb1b2

Please sign in to comment.