CPU power management in video transcoding servers

Minseok Song; Yeongju Lee; Jinhan Park

doi:10.1145/2578260.2578269

CPU power management in video transcoding servers

Minseok Song, Yeongju Lee, Jinhan Park

Inha University

Research output: Contribution to conference › Paper › peer-review

3 Scopus citations

Abstract

To support the diverse requirements of mobile clients and fluctuating network bandwidth, video content for wireless streaming is usually transcoded into several versions. However, transcoding requires a lot of computation, resulting in high power consumption by the server. We examine the characteristics of transcoding workloads and characterize them in terms of deadline and computation time. We then propose a new dynamic voltage and frequency scaling (DVFS) scheme that allocates a frequency and a workload to each CPU with the aim of minimizing power consumption while meeting all transcoding deadlines. Experimental results show that our scheme can reduce system-wide energy consumption by 10% to 16% compared with the Linux Ondemand governor while ensuring all the transcoding tasks meet their deadlines.

Original language	English
Pages	91-96
Number of pages	6
DOIs	https://doi.org/10.1145/2578260.2578269
State	Published - 2014
Event	24th ACM Workshop on Network and Operating Systems Support for Digital Audio and Video, NOSSDAV 2014 - Singapore, Singapore Duration: 19 Mar 2014 → 20 Mar 2014

Conference

Conference	24th ACM Workshop on Network and Operating Systems Support for Digital Audio and Video, NOSSDAV 2014
Country/Territory	Singapore
City	Singapore
Period	19/03/14 → 20/03/14

Keywords

Dynamic voltage scaling
Low-power systems
Multimedia systems

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1145/2578260.2578269

Cite this

@conference{7406d4df02c942a4b11ae300a5d93bf4,

title = "CPU power management in video transcoding servers",

abstract = "To support the diverse requirements of mobile clients and fluctuating network bandwidth, video content for wireless streaming is usually transcoded into several versions. However, transcoding requires a lot of computation, resulting in high power consumption by the server. We examine the characteristics of transcoding workloads and characterize them in terms of deadline and computation time. We then propose a new dynamic voltage and frequency scaling (DVFS) scheme that allocates a frequency and a workload to each CPU with the aim of minimizing power consumption while meeting all transcoding deadlines. Experimental results show that our scheme can reduce system-wide energy consumption by 10% to 16% compared with the Linux Ondemand governor while ensuring all the transcoding tasks meet their deadlines.",

keywords = "Dynamic voltage scaling, Low-power systems, Multimedia systems",

author = "Minseok Song and Yeongju Lee and Jinhan Park",

year = "2014",

doi = "10.1145/2578260.2578269",

language = "English",

pages = "91--96",

note = "24th ACM Workshop on Network and Operating Systems Support for Digital Audio and Video, NOSSDAV 2014 ; Conference date: 19-03-2014 Through 20-03-2014",

}

TY - CONF

T1 - CPU power management in video transcoding servers

AU - Song, Minseok

AU - Lee, Yeongju

AU - Park, Jinhan

PY - 2014

Y1 - 2014

N2 - To support the diverse requirements of mobile clients and fluctuating network bandwidth, video content for wireless streaming is usually transcoded into several versions. However, transcoding requires a lot of computation, resulting in high power consumption by the server. We examine the characteristics of transcoding workloads and characterize them in terms of deadline and computation time. We then propose a new dynamic voltage and frequency scaling (DVFS) scheme that allocates a frequency and a workload to each CPU with the aim of minimizing power consumption while meeting all transcoding deadlines. Experimental results show that our scheme can reduce system-wide energy consumption by 10% to 16% compared with the Linux Ondemand governor while ensuring all the transcoding tasks meet their deadlines.

AB - To support the diverse requirements of mobile clients and fluctuating network bandwidth, video content for wireless streaming is usually transcoded into several versions. However, transcoding requires a lot of computation, resulting in high power consumption by the server. We examine the characteristics of transcoding workloads and characterize them in terms of deadline and computation time. We then propose a new dynamic voltage and frequency scaling (DVFS) scheme that allocates a frequency and a workload to each CPU with the aim of minimizing power consumption while meeting all transcoding deadlines. Experimental results show that our scheme can reduce system-wide energy consumption by 10% to 16% compared with the Linux Ondemand governor while ensuring all the transcoding tasks meet their deadlines.

KW - Dynamic voltage scaling

KW - Low-power systems

KW - Multimedia systems

UR - http://www.scopus.com/inward/record.url?scp=84898965746&partnerID=8YFLogxK

U2 - 10.1145/2578260.2578269

DO - 10.1145/2578260.2578269

M3 - Paper

AN - SCOPUS:84898965746

SP - 91

EP - 96

T2 - 24th ACM Workshop on Network and Operating Systems Support for Digital Audio and Video, NOSSDAV 2014

Y2 - 19 March 2014 through 20 March 2014

ER -

CPU power management in video transcoding servers

Abstract

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this