Abstract:Emerging reconfigurable data centers introduce unprecedented flexibility in how the physical layer can be programmed to adapt to current traffic demands. These reconfigurable topologies are commonly hybrid, consisting of static and reconfigurable links, enabled by e.g. an Optical Circuit Switch (OCS) connected to top-of-rack switches in Clos networks. Even though prior work has showcased the practical benefits of hybrid networks, several crucial performance aspects are not well understood. For example, many systems enforce artificial segregation of the hybrid network parts, leaving money on the table. In this paper, we study the algorithmic problem of how to jointly optimize topology and routing in reconfigurable data centers, in order to optimize a most fundamental metric, maximum link load. The complexity of reconfiguration mechanisms in this space is unexplored at large, especially for the following cross-layer network-design problem: given a hybrid network and a traffic matrix, jointly design the physical layer and the flow routing in order to minimize the maximum link load. We chart the corresponding algorithmic landscape in our work, investigating both un-/splittable flows and (non-)segregated routing policies. A topological complexity classification of the problem reveals NP-hardness in general for network topologies that are trees of depth at least two, in contrast to the tractability on trees of depth one. We moreover prove that the problem is not submodular for all these routing policies, even in multi-layer trees. However, networks that can be abstracted by a single packet switch (e.g., nonblocking Fat-Tree topologies) can be optimized efficiently, and we present optimal polynomial-time algorithms accordingly. We complement our theoretical results with trace-driven simulation studies, where our algorithms can significantly improve the network load in comparison to the state of the art.

Efficient Minimum Spanning Tree Algorithms on the Reconfigurable Mesh

Interconnect Estimation for Mesh-Based Reconfigurable Computing

Minimizing interference for the highway model in wireless ad-hoc and sensor networks

On-line Scheduling Mesh Jobs with Dependencies.

Temporal task partitioning algorithm for dynamically reconfigurable systems

A Parallel Local Reconnection Approach for Tetrahedral Mesh Improvement

Exact Algorithms to Minimize Interference in Wireless Sensor Networks

A Constant Time Optimal Parallel EXOR Algorithm on Reconfigurable Meshes and Its Applications

Spanning graph-based nonrectilinear steiner tree algorithms

A Constant-Time Algorithm for Computing the Euclidean Distance Transform on Reconfigurable Meshes

Efficient Fault-Tolerant Topology Reconfiguration Using a Maximum Flow Algorithm.

A Novel Approach Using a Minimum Cost Maximum Flow Algorithm for Fault-Tolerant Topology Reconfiguration in NoC Architectures.

Efficient Maintenance of Minimum Spanning Trees in Dynamic Weighted Undirected Graphs

Online Scheduling Algorithm with Reconfigurable Resource Management Based on Vertex Position Tree

Reconfiguring Shortest Paths in Graphs

Load-Optimization in Reconfigurable Data-Center Networks: Algorithms and Complexity of Flow Routing

Achieving Flexible Global Reconfiguration in NoCs Using Reconfigurable Rings.

Exact Algorithms for the Minimum Load Spanning Tree Problem

Fast Algorithms for Simulating the CRCW Shared-Memory Computer on Reconfigurable Meshes

Achieving Efficient Routing in Reconfigurable DCNs.

An 0(|E|loglog|V|) algorithm for finding minimum spanning trees