Search arXivSearch

arXiv · 2202.03157

T-Plots: A Novel Approach to Network Design

Abstract

It is accepted wisdom that changes in the traffic matrix entail capacity over-provisioning, but there is no simple measure of just how much over-provisioning can buy. In this Thesis, we aim to provide the network designer with a simple view of the network robustness to traffic matrix changes. We first present the Traffic Load Distribution Plots, or T-Plots, a class of plots illustrating the percentage of traffic matrices that can be serviced as a function of the capacity over-provisioning. For instance, from a simple look at their T- Plots, network designers can guarantee that their network services all admissible traffic matrices, or 99% of permutation traffic matrices, or all traffic matrices with ingress/egress load at most half the maximum. We further show that, unfortunately, in the general case plotting T-Plots is #P-Complete, i.e., that it is impossible to plot a T-plot in a polynomial time by the noon tools. However, we show that T-Plots can sometimes be closely modeled as Gaussian, thus only using two values (mean and variance) to quantify the robustness of a capacity allocation to traffic matrix changes. We further utilize these Gaussian T-Plots to provide a more robust capacity allocation. Finally, we demonstrate the benefits of using T-Plots by showing results of extensive Monte Carlo simulations in a real backbone network. This Thesis was submitted in 2007. Since then, the results that appeared in it were applied in various networking environments. In this newer version, we revisit the results 13 years later and explain their relevance to state-of-the-art problems in network design.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Itamar Cohen. 2021-12-09. T-Plots: A Novel Approach to Network Design. https://arxiv.org/abs/2202.03157

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Accelerating Stateful Network Applications with Performance Prediction on SoC SmartNICs

Offloading stateful network functions to multi-threaded SoC SmartNICs promises significant performance and cost benefits. However, realizing this potential is hindered by two fundamental challenges. First, without performance guidance, developers are forced into a slow, manual trial-and-error cycle of deploying and testing to find a feasible resource allocation. Second, sustaining performance under changing traffic requires adapting state residency and handling overload within memory layouts fixed at compile time. This paper introduces Vela, a framework that addresses both challenges through a model-driven, compile-time/runtime co-design. Its core is a predictive compiler that replaces the manual tuning loop with fast, automated analysis, using a novel, state-centric analytical model to estimate the throughput ceiling of any given resource allocation plan. This is complemented by a lightweight runtime that dynamically manages cache contents within the compiled memory layout and mitigates overload. We implement and evaluate Vela on Netronome Agilio and NVIDIA BlueField 3 SmartNICs across four NFs. Vela reduces host CPU or on-board Arm core usage by 62.9%-91.9% relative to the best baseline achieving the same throughput. For the NATLB workload, vela also improves throughput by 15.2%-120.8% over the best baseline at the same host/Arm core count.

cs.NI

MonkeyTree: Near-Minimal Congestion for Multi-tenant Training via Migration

We present MonkeyTree, the first system to mitigate network congestion in multi-tenant GPU clusters through job-migration based defragmentation rather than network-layer techniques. As cloud operators co-locate ML training jobs on shared, oversubscribed networks, congestion degrades training throughput for over a third of jobs. Prior approaches either rely on routing and flow scheduling--which we show have fundamental limits when traffic exceeds capacity--or require costly full-bisection bandwidth topologies with packet spraying. MonkeyTree exploits characteristics of ML training traffic: ring-based collectives generate exactly one cross-rack flow per rack a job spans, making congestion-free placements achievable. The sparse constraint structure admits abundant valid configurations, making them easy to reach with few migrations. Once reached, low fragmentation is self-reinforcing, as new arrivals disturb only a few racks. MonkeyTree formulates defragmentation as an integer linear program that minimizes worker movements, subject to per-rack fragmentation bounds. We prove a tight bound showing any placement can be defragmented to at most two cross-rack fragments per ToR, and extend the formulation to hybrid parallelism with multiple rings per server. Migration is implemented via in-memory checkpoint-and-restore over RDMA, incurring only 9.02 seconds of system overhead end-to-end per worker. We evaluate MonkeyTree using a custom simulator modeling clusters of up to 2,048 H200 GPUs and prototype on a five-node A100 testbed. MonkeyTree improves average job completion time by 14 percent over the next best baseline on a cluster of 1,024 GPUs with a 4:1 oversubscription. With a high 16:1 oversubscription ratio and 2,048 GPUs, MonkeyTree keeps p99 job completion time within 5 percent of ideal.

cs.NI

The Operable Pareto Front: Distilling Offline Search into Run-Time Control for Multi-Objective UAV Edge-Computing Scheduling

A UAV mobile edge computing (MEC) fleet trades energy against delay, and its schedules form a Pareto front; we call a scheduler operable when the fleet can be asked for any point on that front at run time. We propose PrefDT, to the best of our knowledge the first preference-conditioned Decision Transformer for the problem of joint trajectory, association and offloading scheduling. Its idea comes from language modeling: we hand the model the desired trade-off as an input, such that a single model only needs to be trained once offline to return any desired point on the curve in one rollout. The fleet's state is summarized by attention pooling with a per-user bypass, so the scheduler keeps working when user reports are lost. The energy target is a running budget decremented by what the fleet actually spends. As a result, when wind or load pushes consumption off the plan, the policy can track the difference and hold its budget. Because no corpus of preference-labeled flights exists, we design a distillation pipeline and build the corpus by ourselves. In simulation against 26 method variants, PrefDT produces the best trade-off curve of any learned method and holds its energy budget to within 0.6% when propulsion cost rises by half in mid-flight.

cs.NI