Saved in:
Bibliographic Details
Main Authors: Chang, Brian, Mogul, Jeffrey C., Wang, Rui, Zhang, Mingyang, Akella, Aditya
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.06004
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909382548127744
author Chang, Brian
Mogul, Jeffrey C.
Wang, Rui
Zhang, Mingyang
Akella, Aditya
author_facet Chang, Brian
Mogul, Jeffrey C.
Wang, Rui
Zhang, Mingyang
Akella, Aditya
contents Applications that run in large-scale data center networks (DCNs) rely on the DCN's ability to deliver application requests in a performant manner. DCNs expose a complex design and operational space, and network designers and operators care how different options along this space affect application performance. One might run controlled experiments and measure the corresponding application-facing performance, but such experiments become progressively infeasible at a large scale, and simulations risk yielding inaccurate or incomplete results. Instead, we show that we can predict application-facing performance through more easily measured network metrics. For example, network telemetry metrics (e.g., link utilization) can predict application-facing metrics (e.g., transfer latency). Through large-scale measurements of production networks, we study the correlation between the two types of metrics, and construct predictive, interpretable models that serve as a suggestive guideline to network designers and operators. We show that no single network metric is universally the best predictor (even though some prior work has focused on a single predictor). We found that simple linear models often have the lowest error, while queueing-based models are better in a few cases.
format Preprint
id arxiv_https___arxiv_org_abs_2411_06004
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Do Data Center Network Metrics Predict Application-Facing Performance?
Chang, Brian
Mogul, Jeffrey C.
Wang, Rui
Zhang, Mingyang
Akella, Aditya
Networking and Internet Architecture
Applications that run in large-scale data center networks (DCNs) rely on the DCN's ability to deliver application requests in a performant manner. DCNs expose a complex design and operational space, and network designers and operators care how different options along this space affect application performance. One might run controlled experiments and measure the corresponding application-facing performance, but such experiments become progressively infeasible at a large scale, and simulations risk yielding inaccurate or incomplete results. Instead, we show that we can predict application-facing performance through more easily measured network metrics. For example, network telemetry metrics (e.g., link utilization) can predict application-facing metrics (e.g., transfer latency). Through large-scale measurements of production networks, we study the correlation between the two types of metrics, and construct predictive, interpretable models that serve as a suggestive guideline to network designers and operators. We show that no single network metric is universally the best predictor (even though some prior work has focused on a single predictor). We found that simple linear models often have the lowest error, while queueing-based models are better in a few cases.
title Do Data Center Network Metrics Predict Application-Facing Performance?
topic Networking and Internet Architecture
url https://arxiv.org/abs/2411.06004