Scalable bandwidth shaping scheme via adaptively managed parallel heaps in manycore-based network processors

Taehyun Kim, Jongbum Lim, Jinku Kim, Woo Cheol Cho, Eui-Young Chung, Hyuk Jun Lee

Research output: Contribution to journalArticle

Abstract

Scalability of network processor-based routers heavily depends on limitations imposed by memory accesses and associated power consumption. Bandwidth shaping of a flow is a key function, which requires a token bucket per output queue and abuses memory bandwidth. As the number of output queues increases, managing token buckets becomes prohibitively expensive and limits scalability. In this work, we propose a scalable software-based token bucket management scheme that can reduce memory accesses and power consumption significantly. To satisfy real-Time and low-cost constraints, we propose novel parallel heap data structures running on a manycore-based network processor. By using cache locking, the performance of heap processing is enhanced significantly and is more predictable. In addition, we quantitatively analyze the performance and memory footprint of the proposed software scheme using stochastic modeling and the Lyapunov central limit theorem. Finally, the proposed scheme provides an adaptive method to limit the size of heaps in the case of oversubscribed queues, which can successfully isolate the queues showing unideal behavior. The proposed scheme reduces memory accesses by up to three orders of magnitude for one million queues sharing a 100Gbps interface of the router while maintaining stability under stressful scenarios.

Original languageEnglish
Article number59
JournalACM Transactions on Design Automation of Electronic Systems
Volume22
Issue number4
DOIs
Publication statusPublished - 2017 May 1

Fingerprint

Bandwidth
Data storage equipment
Routers
Scalability
Electric power utilization
Data structures
Processing
Costs

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Electrical and Electronic Engineering

Cite this

@article{11e74a1996f74996969efb1847746d12,
title = "Scalable bandwidth shaping scheme via adaptively managed parallel heaps in manycore-based network processors",
abstract = "Scalability of network processor-based routers heavily depends on limitations imposed by memory accesses and associated power consumption. Bandwidth shaping of a flow is a key function, which requires a token bucket per output queue and abuses memory bandwidth. As the number of output queues increases, managing token buckets becomes prohibitively expensive and limits scalability. In this work, we propose a scalable software-based token bucket management scheme that can reduce memory accesses and power consumption significantly. To satisfy real-Time and low-cost constraints, we propose novel parallel heap data structures running on a manycore-based network processor. By using cache locking, the performance of heap processing is enhanced significantly and is more predictable. In addition, we quantitatively analyze the performance and memory footprint of the proposed software scheme using stochastic modeling and the Lyapunov central limit theorem. Finally, the proposed scheme provides an adaptive method to limit the size of heaps in the case of oversubscribed queues, which can successfully isolate the queues showing unideal behavior. The proposed scheme reduces memory accesses by up to three orders of magnitude for one million queues sharing a 100Gbps interface of the router while maintaining stability under stressful scenarios.",
author = "Taehyun Kim and Jongbum Lim and Jinku Kim and Cho, {Woo Cheol} and Eui-Young Chung and Lee, {Hyuk Jun}",
year = "2017",
month = "5",
day = "1",
doi = "10.1145/3065926",
language = "English",
volume = "22",
journal = "ACM Transactions on Design Automation of Electronic Systems",
issn = "1084-4309",
publisher = "Association for Computing Machinery (ACM)",
number = "4",

}

Scalable bandwidth shaping scheme via adaptively managed parallel heaps in manycore-based network processors. / Kim, Taehyun; Lim, Jongbum; Kim, Jinku; Cho, Woo Cheol; Chung, Eui-Young; Lee, Hyuk Jun.

In: ACM Transactions on Design Automation of Electronic Systems, Vol. 22, No. 4, 59, 01.05.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Scalable bandwidth shaping scheme via adaptively managed parallel heaps in manycore-based network processors

AU - Kim, Taehyun

AU - Lim, Jongbum

AU - Kim, Jinku

AU - Cho, Woo Cheol

AU - Chung, Eui-Young

AU - Lee, Hyuk Jun

PY - 2017/5/1

Y1 - 2017/5/1

N2 - Scalability of network processor-based routers heavily depends on limitations imposed by memory accesses and associated power consumption. Bandwidth shaping of a flow is a key function, which requires a token bucket per output queue and abuses memory bandwidth. As the number of output queues increases, managing token buckets becomes prohibitively expensive and limits scalability. In this work, we propose a scalable software-based token bucket management scheme that can reduce memory accesses and power consumption significantly. To satisfy real-Time and low-cost constraints, we propose novel parallel heap data structures running on a manycore-based network processor. By using cache locking, the performance of heap processing is enhanced significantly and is more predictable. In addition, we quantitatively analyze the performance and memory footprint of the proposed software scheme using stochastic modeling and the Lyapunov central limit theorem. Finally, the proposed scheme provides an adaptive method to limit the size of heaps in the case of oversubscribed queues, which can successfully isolate the queues showing unideal behavior. The proposed scheme reduces memory accesses by up to three orders of magnitude for one million queues sharing a 100Gbps interface of the router while maintaining stability under stressful scenarios.

AB - Scalability of network processor-based routers heavily depends on limitations imposed by memory accesses and associated power consumption. Bandwidth shaping of a flow is a key function, which requires a token bucket per output queue and abuses memory bandwidth. As the number of output queues increases, managing token buckets becomes prohibitively expensive and limits scalability. In this work, we propose a scalable software-based token bucket management scheme that can reduce memory accesses and power consumption significantly. To satisfy real-Time and low-cost constraints, we propose novel parallel heap data structures running on a manycore-based network processor. By using cache locking, the performance of heap processing is enhanced significantly and is more predictable. In addition, we quantitatively analyze the performance and memory footprint of the proposed software scheme using stochastic modeling and the Lyapunov central limit theorem. Finally, the proposed scheme provides an adaptive method to limit the size of heaps in the case of oversubscribed queues, which can successfully isolate the queues showing unideal behavior. The proposed scheme reduces memory accesses by up to three orders of magnitude for one million queues sharing a 100Gbps interface of the router while maintaining stability under stressful scenarios.

UR - http://www.scopus.com/inward/record.url?scp=85019663587&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019663587&partnerID=8YFLogxK

U2 - 10.1145/3065926

DO - 10.1145/3065926

M3 - Article

VL - 22

JO - ACM Transactions on Design Automation of Electronic Systems

JF - ACM Transactions on Design Automation of Electronic Systems

SN - 1084-4309

IS - 4

M1 - 59

ER -