Free-scaling your data center

Laszlo Gyarmati, András Gulyás, Balázs Sonkoly, Tuan A. Trinh, Gergely Biczók

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The increasing popularity of both small and large private clouds and expanding public clouds poses new requirements to data center (DC) architectures. First, DC architectures should be incrementally scalable allowing the creation of DCs of arbitrary size with consistent performance characteristics. Second, initial DC deployments should be incrementally expandable supporting small-scale upgrades without decreasing operation efficiency. A DC architecture possessing both properties satisfies the requirement of free-scaling. Recent work in DC design focuses on traditional performance and scalability characteristics, therefore resulting in symmetric topologies whose upgradability is coarse-grained at best. In our earlier work we proposed Scafida, an asymmetric, scale-free network inspired DC topology which scales incrementally and has favorable structural characteristics. In this paper, we build on Scafida and propose a full-fledged DC architecture achieving free-scaling called FScafida. Our main contribution is threefold. First, we propose an organic expansion algorithm for FScafida; this combined with Scafida's flexible original design results in a freely scalable architecture. Second, we introduce the Effective Source Routing mechanism that provides near-shortest paths, multi-path and multicast capability, and low signaling overhead by exploiting the benefits of the FScafida topology. Third, we show based on extensive simulations and a prototype implementation that FScafida is capable of handling the traffic patterns characteristic of both enterprise and cloud data centers, tolerates network equipment failures to a high degree, and allows for high bisection bandwidth.

Original languageEnglish
Pages (from-to)1758-1773
Number of pages16
JournalComputer Networks
Volume57
Issue number8
DOIs
Publication statusPublished - 4 Jun 2013
Externally publishedYes

Fingerprint

Topology
Complex networks
Scalability
Bandwidth
Industry

Keywords

  • Asymmetric topology
  • Data center
  • Fault-tolerance
  • Routing
  • Scale-free

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Gyarmati, L., Gulyás, A., Sonkoly, B., Trinh, T. A., & Biczók, G. (2013). Free-scaling your data center. Computer Networks, 57(8), 1758-1773. https://doi.org/10.1016/j.comnet.2013.03.005

Free-scaling your data center. / Gyarmati, Laszlo; Gulyás, András; Sonkoly, Balázs; Trinh, Tuan A.; Biczók, Gergely.

In: Computer Networks, Vol. 57, No. 8, 04.06.2013, p. 1758-1773.

Research output: Contribution to journalArticle

Gyarmati, L, Gulyás, A, Sonkoly, B, Trinh, TA & Biczók, G 2013, 'Free-scaling your data center', Computer Networks, vol. 57, no. 8, pp. 1758-1773. https://doi.org/10.1016/j.comnet.2013.03.005
Gyarmati L, Gulyás A, Sonkoly B, Trinh TA, Biczók G. Free-scaling your data center. Computer Networks. 2013 Jun 4;57(8):1758-1773. https://doi.org/10.1016/j.comnet.2013.03.005
Gyarmati, Laszlo ; Gulyás, András ; Sonkoly, Balázs ; Trinh, Tuan A. ; Biczók, Gergely. / Free-scaling your data center. In: Computer Networks. 2013 ; Vol. 57, No. 8. pp. 1758-1773.
@article{975dddc450404f8b81aa57f3043f50ff,
title = "Free-scaling your data center",
abstract = "The increasing popularity of both small and large private clouds and expanding public clouds poses new requirements to data center (DC) architectures. First, DC architectures should be incrementally scalable allowing the creation of DCs of arbitrary size with consistent performance characteristics. Second, initial DC deployments should be incrementally expandable supporting small-scale upgrades without decreasing operation efficiency. A DC architecture possessing both properties satisfies the requirement of free-scaling. Recent work in DC design focuses on traditional performance and scalability characteristics, therefore resulting in symmetric topologies whose upgradability is coarse-grained at best. In our earlier work we proposed Scafida, an asymmetric, scale-free network inspired DC topology which scales incrementally and has favorable structural characteristics. In this paper, we build on Scafida and propose a full-fledged DC architecture achieving free-scaling called FScafida. Our main contribution is threefold. First, we propose an organic expansion algorithm for FScafida; this combined with Scafida's flexible original design results in a freely scalable architecture. Second, we introduce the Effective Source Routing mechanism that provides near-shortest paths, multi-path and multicast capability, and low signaling overhead by exploiting the benefits of the FScafida topology. Third, we show based on extensive simulations and a prototype implementation that FScafida is capable of handling the traffic patterns characteristic of both enterprise and cloud data centers, tolerates network equipment failures to a high degree, and allows for high bisection bandwidth.",
keywords = "Asymmetric topology, Data center, Fault-tolerance, Routing, Scale-free",
author = "Laszlo Gyarmati and Andr{\'a}s Guly{\'a}s and Bal{\'a}zs Sonkoly and Trinh, {Tuan A.} and Gergely Bicz{\'o}k",
year = "2013",
month = "6",
day = "4",
doi = "10.1016/j.comnet.2013.03.005",
language = "English",
volume = "57",
pages = "1758--1773",
journal = "Computer Networks",
issn = "1389-1286",
publisher = "Elsevier",
number = "8",

}

TY - JOUR

T1 - Free-scaling your data center

AU - Gyarmati, Laszlo

AU - Gulyás, András

AU - Sonkoly, Balázs

AU - Trinh, Tuan A.

AU - Biczók, Gergely

PY - 2013/6/4

Y1 - 2013/6/4

N2 - The increasing popularity of both small and large private clouds and expanding public clouds poses new requirements to data center (DC) architectures. First, DC architectures should be incrementally scalable allowing the creation of DCs of arbitrary size with consistent performance characteristics. Second, initial DC deployments should be incrementally expandable supporting small-scale upgrades without decreasing operation efficiency. A DC architecture possessing both properties satisfies the requirement of free-scaling. Recent work in DC design focuses on traditional performance and scalability characteristics, therefore resulting in symmetric topologies whose upgradability is coarse-grained at best. In our earlier work we proposed Scafida, an asymmetric, scale-free network inspired DC topology which scales incrementally and has favorable structural characteristics. In this paper, we build on Scafida and propose a full-fledged DC architecture achieving free-scaling called FScafida. Our main contribution is threefold. First, we propose an organic expansion algorithm for FScafida; this combined with Scafida's flexible original design results in a freely scalable architecture. Second, we introduce the Effective Source Routing mechanism that provides near-shortest paths, multi-path and multicast capability, and low signaling overhead by exploiting the benefits of the FScafida topology. Third, we show based on extensive simulations and a prototype implementation that FScafida is capable of handling the traffic patterns characteristic of both enterprise and cloud data centers, tolerates network equipment failures to a high degree, and allows for high bisection bandwidth.

AB - The increasing popularity of both small and large private clouds and expanding public clouds poses new requirements to data center (DC) architectures. First, DC architectures should be incrementally scalable allowing the creation of DCs of arbitrary size with consistent performance characteristics. Second, initial DC deployments should be incrementally expandable supporting small-scale upgrades without decreasing operation efficiency. A DC architecture possessing both properties satisfies the requirement of free-scaling. Recent work in DC design focuses on traditional performance and scalability characteristics, therefore resulting in symmetric topologies whose upgradability is coarse-grained at best. In our earlier work we proposed Scafida, an asymmetric, scale-free network inspired DC topology which scales incrementally and has favorable structural characteristics. In this paper, we build on Scafida and propose a full-fledged DC architecture achieving free-scaling called FScafida. Our main contribution is threefold. First, we propose an organic expansion algorithm for FScafida; this combined with Scafida's flexible original design results in a freely scalable architecture. Second, we introduce the Effective Source Routing mechanism that provides near-shortest paths, multi-path and multicast capability, and low signaling overhead by exploiting the benefits of the FScafida topology. Third, we show based on extensive simulations and a prototype implementation that FScafida is capable of handling the traffic patterns characteristic of both enterprise and cloud data centers, tolerates network equipment failures to a high degree, and allows for high bisection bandwidth.

KW - Asymmetric topology

KW - Data center

KW - Fault-tolerance

KW - Routing

KW - Scale-free

UR - http://www.scopus.com/inward/record.url?scp=84878113582&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878113582&partnerID=8YFLogxK

U2 - 10.1016/j.comnet.2013.03.005

DO - 10.1016/j.comnet.2013.03.005

M3 - Article

VL - 57

SP - 1758

EP - 1773

JO - Computer Networks

JF - Computer Networks

SN - 1389-1286

IS - 8

ER -