Reducing Reconfiguration Time in Hybrid Optical-Electrical Datacenter Networks (Extended Abstract)

05/08/2023
by   Shuyuan Zhang, et al.
0

We study how to reduce the reconfiguration time in hybrid optical-electrical Datacenter Networks (DCNs). With a layer of Optical Circuit Switches (OCSes), hybrid optical-electrical DCNs can reconfigure its logical topologies to better match the on-going traffic patterns, but the reconfiguration time may directly affect the benefits of reconfigurability. The reconfiguration time consists of the topology solver running time and the network convergence time after triggering reconfiguration. However, existing topology solvers either incur high algorithmic complexity or fail to minimize the reconfiguration overhead. In this paper, we propose a novel algorithm that combines the ideas of bipartition and Minimum Cost Flow (MCF) to reduce the overall reconfiguration time. For the first time, we formulate the topology solving problem as a MCF problem with piecewise-linear cost, which strikes a better balance between solver complexity and solution optimality. Our evaluation shows that our algorithm can significantly reduce the network convergence time while consuming less topology solver running time, making its overall performance superior to existing algorithms. Our code and test cases are available at a public repository.

READ FULL TEXT

page 1

page 2

research
09/06/2019

A quasilinear complexity algorithm for the numerical simulation of scattering from a two-dimensional radially symmetric potential

Standard solvers for the variable coefficient Helmholtz equation in two ...
research
09/30/2020

COUDER: Robust Topology Engineering for Optical Circuit Switched Data Center Networks

Many optical circuit switched data center networks (DCN) have been propo...
research
02/01/2022

Self-Adjusting Ego-Trees Topology for Reconfigurable Datacenter Networks

State-of-the-art topologies for datacenters (DC) and high-performance co...
research
07/08/2023

SDT: A Low-cost and Topology-reconfigurable Testbed for Network Research

Network experiments are essential to network-related scientific research...
research
06/10/2020

Empirical Time Complexity of Generic Dijkstra Algorithm

Generic Dijkstra is a novel algorithm for finding the optimal shortest p...
research
09/07/2022

Constructing Optimal Contraction Trees for Tensor Network Quantum Circuit Simulation

One of the key problems in tensor network based quantum circuit simulati...
research
02/05/2021

Network Support for High-performance Distributed Machine Learning

The traditional approach to distributed machine learning is to adapt lea...

Please sign up or login with your details

Forgot password? Click here to reset