Resource Allocation Model for Hybrid Switching Networks
draft-sun-nmrg-hybrid-switching-00

Document Type Active Internet-Draft (individual)
Last updated 2019-12-30
Stream (None)
Intended RFC status (None)
Formats plain text pdf htmlized bibtex
Stream Stream state (No stream defined)
Consensus Boilerplate Unknown
RFC Editor Note (None)
IESG IESG state I-D Exists
Telechat date
Responsible AD (None)
Send notices to (None)
Network Working Group                                             W. Sun
Internet-Draft                                                   J. Shao
Intended status: Informational                                     W. Hu
Expires: July 3, 2020                                               SJTU
                                                            Dec 31, 2019

        Resource Allocation Model for Hybrid Switching Networks
                 draft-sun-nmrg-hybrid-switching-00.txt

Abstract

   The fast increase in traffic volumn within and outside Datacenters is
   placing an unprecendented challenge on the underline network, in both
   the capacity it can provide, and the way it delivers traffic.  When a
   large portion of network traffic is contributed by large flows,
   providing high capacity and slow to change optical circuit switching
   along side fine-granular packet services may potentially improve
   network utility and reduce both CAPEX and OpEX.  This gives rise to
   the concept of hybrid switching - a paradigm that seeks to make the
   best of packet and circuit switching.

   However, the full potential of hybrid switching networks (HSNs) can
   only be realized when such a network is optimally designed and
   operated, in the sense that "an appropriate amount of resource is
   used to handle an appropriate amount of traffic in both switching
   planes."  The resource allocation problem in HSNs is in fact complex
   ineractions between three components: resource allocation between the
   two switching planes, traffic partitioning between the two switching
   planes, and the overall cost or performance constraints.

   In this memo, we explore the challenges of planning and operating
   hybrid switching networks, with a particular focus on the resource
   allocation problem, and provide a high-level model that may guide
   resource allocation in future hybrid switching networks.

Status of This Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at https://datatracker.ietf.org/drafts/current/.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any

Sun, et al.               Expires July 3, 2020                  [Page 1]
Internet-Draft     Resource Allocation Model for HSNs           Dec 2019

   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   This Internet-Draft will expire on July 3, 2020.

Copyright Notice

   Copyright (c) 2019 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (https://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.

1.  Introduction

   In facing rapid increase of network traffic [Gantz12], as well as the
   number of servers in cloud data centers [Cisco15], new architectures
   and operation models of Data Center Networks (DCNs) gained wide
   interests.  One concept that attracted considerable and lasting
   attention is the introduction of optical switching technologies into
   DCNs, hoping that bypassing some of the traffic without performing
   per-packet electronic processing will help reducing the Operational
   Cost (OpEx), as well as the Capital Expenditure (CapEx) of DCNs.
   This concept of combining electronic packet switching (EPS) and
   optical switching (often optical circuit switching, OCS), is called
   hybrid switching [Zukerman89].  In recent years, many hybrid
   switching schemes have been proposed [Gauger06], and it is reasonable
   to believe that when a DCN grows beyond a certain scale, the benefit
   of introducing optical switching will emerge and become more evident
   as the size of the DC continues to increase.

   On the other hand, achieving the benefits of hybrid switching
   requires careful design at the planning stage, and proper operation
   during runtime.  This poses challenges that goes far beyond the
   topological or architectural aspects.  For instance, at the planning
   stage, one has to decide how much to invest in the two switching
   planes, such that each could be fully utilized when the network
   becomes operational.  Under cases when dynamic resource allocation
   between the two planes are possible, one has to decide how resource
   is allocated between the two planes, and how traffic should be
   directed to each of them, such that performance constraints can be
Show full document text