Computationally Efficient Worst-Case Analysis of Flow-Controlled Networks with Network Calculus

TL;DR

This paper introduces new computational techniques to efficiently analyze worst-case delays in flow-controlled networks using Network Calculus, enabling exact results for complex multi-hop scenarios that were previously infeasible.

Contribution

It presents novel algorithms and methods to reduce data complexity and computation time in Network Calculus analysis of flow-controlled networks, overcoming existing intractability issues.

Findings

Significant speedup in delay analysis computations

Ability to analyze larger, more complex network scenarios

Exact results without approximation

Abstract

Networks with hop-by-hop flow control occur in several contexts, from data centers to systems architectures (e.g., wormhole-routing networks on chip). A worst-case end-to-end delay in such networks can be computed using Network Calculus (NC), an algebraic theory where traffic and service guarantees are represented as curves in a Cartesian plane. NC uses transformation operations, e.g., the min-plus convolution, to model how the traffic profile changes with the traversal of network nodes. NC allows one to model flow-controlled systems, hence one can compute the end-to-end service curve describing the minimum service guaranteed to a flow traversing a tandem of flow-controlled nodes. However, while the algebraic expression of such an end-to-end service curve is quite compact, its computation is often intractable from an algorithmic standpoint: data structures tend to grow quickly to unfeasibly large sizes, making operations intractable, even with as few as three hops. In this paper, we propose computational and algebraic techniques to mitigate the above problem. We show that existing techniques (such as reduction to compact domains) cannot be used in this case, and propose an arsenal of solutions, which include methods to mitigate the data representation space explosion as well as computationally efficient algorithms for the min-plus convolution operation. We show that our solutions allow a significant speedup, enable analysis of previously unfeasible case studies, and -- since they do not rely on any approximation -- still provide exact results.