Proceedings of the ACM on Measurement and Analysis of Computing Systems

In-band Network Telemetry (INT) enhances real-time, high-resolution network monitoring capabilities by incorporating fine-grained internal state information into packets. Utilizing INT for network-wide visualization can significantly bolster network management and operation. Although existing studies have made significant contributions, they have not been able to simultaneously meet the following two objectives: 1) Comprehensive visibility of the network's internal state, which refers to obtaining information on all switch ports and their corresponding link states; 2) Low measurement overhead, which involves reducing measurement bandwidth overhead and minimizing the impact of the measurement process on the network. We designed INT-MC to meet both of these objectives. INT-MC is an efficient and cost-effective network-wide telemetry solution based on matrix completion. By modeling link metadata as a matrix and leveraging its low-rank property, INT-MC selectively measures certain links. It then employs matrix completion algorithms to infer information about unmeasured links, thereby achieving low-overhead network-wide telemetry. Designing paths to cover the target links involves an NP-complete problem, and the high computational complexity may delay the measurement tasks. To circumvent this, we propose an improved scheme based on Eulerian digraph decomposition, transforming the path calculation problem into a high-information path selection problem, significantly reducing computational costs. We have implemented an INT-MC prototype within the NSFNet topology, consisting of 14 Tofino switches and 10 end-hosts, and conducted extensive experiments and evaluations. The results indicate that INT-MC incurs only 16% of the measurement overhead compared to existing network-wide telemetry solutions, while achieving nearly identical accuracy. Even under high-frequency measurements of 20 times per second, the bandwidth overhead of INT-MC is approximately 0.075% of the total bandwidth, exerting minimal impact on the network.

Xin Zeng

Kun Xie

Yanbiao Li

Ke Xu

Gaogang Xie

Jigang Wen

Yinchuan Cong

Wei Liang

INT-MC: Low-Overhead In-Band Network-Wide Telemetry Based on Matrix Completion

INT-MC：基于矩阵补全的低开销带内全网络遥测

Inferring the quality of streaming video applications is important for Internet service providers, but the fact that most video streams are encrypted makes it difficult to do so. We develop models that infer quality metrics (\ie, startup delay and resolution) for encrypted streaming video services. Our paper builds on previous work, but extends it in several ways. First, the models work in deployment settings where the video sessions and segments must be identified from a mix of traffic and the time precision of the collected traffic statistics is more coarse (\eg, due to aggregation). Second, we develop a single composite model that works for a range of different services (\ie, Netflix, YouTube, Amazon, and Twitch), as opposed to just a single service. Third, unlike many previous models, our models perform predictions at finer granularity (\eg, the precise startup delay instead of just detecting short versus long delays) allowing to draw better conclusions on the ongoing streaming quality. Fourth, we demonstrate the models are practical through a 16-month deployment in 66 homes and provide new insights about the relationships between Internet "speed'' and the quality of the corresponding video streams, for a variety of services; we find that higher speeds provide only minimal improvements to startup delay and resolution.

Francesco Bronzino

Paul Schmitt

Sara Ayoubi

Guilherme Martins

Renata Teixeira

Nick Feamster

Proc. ACM Meas. Anal. Comput. Syst.

Inferring Streaming Video Quality from Encrypted Traffic

O-RAN systems and their deployment in virtualized general-purpose computing platforms (O-Cloud) constitute a paradigm shift expected to bring unprecedented performance gains. However, these architectures raise new implementation challenges and threaten to worsen the already-high energy consumption of mobile networks. This paper presents first a series of experiments which assess the O-Cloud's energy costs and their dependency on the servers' hardware, capacity and data traffic properties which, typically, change over time. Next, it proposes a compute policy for assigning the base station data loads to O-Cloud servers in an energy-efficient fashion; and a radio policy that determines at near-real-time the minimum transmission block size for each user so as to avoid unnecessary energy costs. The policies balance energy savings with performance, and ensure that both of them are dispersed fairly across the servers and users, respectively. To cater for the unknown and time-varying parameters affecting the policies, we develop a novel online learning framework with fairness guarantees that apply to the entire operation horizon of the system (long-term fairness). The policies are evaluated using trace-driven simulations and are fully implemented in an O-RAN compatible system where we measure the energy costs and throughput in realistic scenarios.

Fatih Aslan

George Iosifidis

Jose A. Ayala-Romero

Andres Garcia-Saavedra

Xavier Costa-Perez

Fair Resource Allocation in Virtualized O-RAN Platforms

虚拟化O-RAN平台中的公平资源分配

Datasets of labeled network traces are essential for a multitude of machine learning (ML) tasks in networking, yet their availability is hindered by privacy and maintenance concerns, such as data staleness. To overcome this limitation, synthetic network traces can often augment existing datasets. Unfortunately, current synthetic trace generation methods, which typically produce only aggregated flow statistics or a few selected packet attributes, do not always suffice, especially when model training relies on having features that are only available from packet traces. This shortfall manifests in both insufficient statistical resemblance to real traces and suboptimal performance on ML tasks when employed for data augmentation. In this paper, we apply diffusion models to generate high-resolution synthetic network traffic traces. We present NetDiffusion1, a tool that uses a finely-tuned, controlled variant of a Stable Diffusion model to generate synthetic network traffic that is high fidelity and conforms to protocol specifications. Our evaluation demonstrates that packet captures generated from NetDiffusion can achieve higher statistical similarity to real data and improved ML model performance than current state-of-the-art approaches (e.g., GAN-based approaches). Furthermore, our synthetic traces are compatible with common network analysis tools and support a myriad of network tasks, suggesting that NetDiffusion can serve a broader spectrum of network analysis and testing tasks, extending beyond ML-centric applications.

Xi Jiang

Shinan Liu

Aaron Gember-Jacobson

Arjun Nitin Bhagoji

NetDiffusion: Network Data Augmentation Through Protocol-Constrained Traffic Generation

NetDiffusion：通过协议约束的流量生成实现网络数据增强

Mobility-on-Demand platforms are a fast growing component of the urban transit ecosystem. Though a growing literature addresses the question of how to make individual MoD platforms more efficient, much less is known about the cost of market fragmentation, i.e., the impact on overall welfare due to splitting demand between multiple independent platforms. Our work aims to quantify how much platform fragmentation degrades the efficiency of the system. In particular, we focus on a setting where demand is exogenously split between multiple platforms, and study the increase in supply rebalancing costs incurred by each platform to meet this demand, vis-à-vis the cost incurred by a centralized platform serving the aggregate demand. We show under a large-market scaling, this Price-of-Fragmentation undergoes a phase transition, wherein, depending on the nature of the exogenous demand, the additional cost due to fragmentation either vanishes or grows unbounded. We provide conditions that characterize which regime applies to any given system, and discuss implications of these findings on how such platforms should be regulated.

Thibault Séjournè

Samitha Samaranayake

Siddhartha Banerjee

The Price of Fragmentation in Mobility-on-Demand Services

按需移动服务中的碎片化成本

Graph Neural Networks (GNNs) are emerging models to analyze graph-structure data. GNN execution involves both compute-intensive and memory-intensive kernels. The latter kernels dominate execution time, because they are significantly bottlenecked by data movement between memory and processors. Processing-In-Memory (PIM) systems can alleviate this data movement bottleneck by placing simple processors near or inside to memory arrays. This work investigates the potential of PIM systems to alleviate the data movement bottleneck in GNNs, and introduces PyGim, an efficient and easy-to-use GNN library for real PIM systems. We propose intelligent parallelization techniques for memory-intensive kernels of GNNs tailored for real PIM systems, and develop an easy-to-use Python API for them. PyGim employs a cooperative GNN execution, in which the compute- and memory-intensive kernels are executed in processor-centric and memory-centric computing systems, respectively, to fully exploit the hardware capabilities. PyGim integrates a lightweight autotuner to tune the parallelization strategy of the memory-intensive kernel of GNNs and enable high programming ease. We extensively evaluate PyGim on a real-world PIM system that has 16 PIM DIMMs with 1992 PIM cores connected to a Host CPU. In GNN inference, we demonstrate that it outperforms prior state-of-the-art PIM works by on average 4.38× (up to 7.20×), and state-of-the-art PyTorch running on Host by on average 3.04× (up to 3.44×). PyGim improves energy efficiency by 2.86× (up to 3.68×) and 1.55× (up to 1.75×) over prior PIM and PyTorch Host schemes, respectively. In memory-intensive kernel of GNNs, PyGim provides 11.6× higher resource utilization in PIM system than that of PyTorch library (optimized CUDA implementation) in GPU systems. Our work provides useful recommendations for software, system and hardware designers. PyGim is publicly and freely available at https://github.com/CMU-SAFARI/PyGim facilitate the widespread use of PIM systems in GNNs.

Christina Giannoula

Peiming Yang

Ivan Fernandez

Jiacheng Yang

Sankeerth Durvasula

Yu Xin Li

Mohammad Sadrosadati

Juan Gomez Luna

Onur Mutlu

Gennady Pekhimenko

PyGim : An Efficient Graph Neural Network Library for Real Processing-In-Memory Architectures

PyGim：适用于真实存内处理架构的高效图神经网络库

Many applications, e.g., federated learning, require the aggregation of information across a large number of distributed nodes. In this paper, we explore efficient schemes to do this at scale leveraging aggregation at intermediate nodes across overlay trees. Files/updates are split into chunks which are in turn simultaneously aggregated over different trees. For a synchronous setting with homogeneous communications capabilities and deterministic link delays, we develop a delay optimal aggregation schedule. In the asynchronous setting, where delays are stochastic but i.i.d., across links, we show that for an asynchronous implementation of the above schedule, the expected aggregation delay is near-optimal. We then consider the impact that failures in the network have on the resulting Mean Square Error (MSE) for the estimated aggregates and how it can be controlled through the addition of redundancy, reattempts, and optimal failure-aware estimates for the desired aggregate. Based on the analysis of a natural model of failures, we show how to choose parameters to optimize the trade-off between aggregation delay and MSE. We present simulation results exhibiting the above mentioned tradeoffs. We also consider a more general class of correlated failures and demonstrate via simulation the applicability of our techniques in those settings as well.

Parikshit Hegde

Gustavo de Veciana

Optimal Aggregation via Overlay Trees: Delay-MSE Tradeoffs under Failures

通过覆盖树实现最优聚合：故障情况下的延迟-均方误差权衡

Confidential computing is gaining traction in the cloud, driven by the increasing security and privacy concerns across various industries. Recent trusted hardware advancements introduce Confidential Virtual Machines (CVMs) to alleviate the programmability and usability challenges of the previously proposed enclave-based trusted computing technologies. CVM hardware extensions facilitate secure, hardware-isolated encrypted VMs, promoting programmability and easier deployment in cloud infrastructures. However, differing microarchitectural features, interfaces, and security properties among hardware vendors complicate the evaluation of CVMs for different use cases. Understanding the performance implications, functional limitations, and security guarantees of CVMs is a crucial step toward their adoption. This paper presents a detailed empirical analysis of two leading CVM technologies: AMD Secure Encrypted Virtualization-Secure Nested Paging (SEV-SNP) and Intel Trust Domain Extensions (TDX). We review their microarchitectural components and conduct a thorough performance evaluation across various aspects, including memory management, computational performance, storage and network stacks, and attestation primitives. We further present a security analysis through a trusted computing base (TCB) evaluation and Common Vulnerabilities and Exposures (CVE) analysis. Our key findings demonstrate, among others, the effect of CVMs on boot time, memory management and I/O, and identify inefficiencies in their context switch mechanisms. We further provide insights into the performance implications of CVMs and highlight potential room for improvement.

Masanori Misono

Dimitrios Stavrakakis

Nuno Santos

Pramod Bhatotia

Confidential VMs Explained: An Empirical Analysis of AMD SEV-SNP and Intel TDX

机密虚拟机解析：AMD SEV-SNP与Intel TDX的实证分析

The Proceedings of the ACM on Measurement and Analysis of Computing Systems (POMACS) focuses on the measurement and performance evaluation of computer systems and operates in close collaboration with the ACM Special Interest Group SIGMETRICS. All papers in this issue of POMACS will be presented at the ACM SIGMETRICS 2025 conference on June 9-13, 2025, in Stony Brook, New York, USA. The papers in this issue were selected during the summer submission round by the 93 members of the ACM SIGMETRICS 2025 program committee via a rigorous review process. Each paper was conditionally accepted (and shepherded), or allowed to be resubmitted to one of the subsequent two SIGMETRICS deadlines, or rejected (with resubmission allowed only after a year). POMACS is publishing 20 papers out of 110 submissions in this issue, of which 3 had previously received a one-shot revision decision. Submissions typically receive at least three reviews. While most papers were decided in the online discussion phase, borderline cases were extensively discussed during the online program committee meeting. Based on the indicated track(s), roughly 18% of the submissions were in the Theory track, 39% were in the Measurement &amp; Applied Modeling track, 27% were in the Systems track, and 16% were in the Learning track. Many individuals contributed to the success of this issue of POMACS. We thank the authors, who submitted their best work to SIGMETRICS/POMACS. We would also like to thank the program committee members who provided constructive feedback in their reviews to authors and participated in the online discussions and program committee meeting. Further, we thank the several external reviewers who provided their expert opinions on specific submissions that required additional input. Finally, we are also grateful to the ACM SIGMETRICS Chair, Mor Harchol-Balter, to the SIGMETRICS Organization Committee and to the SIGMETRICS Executive Committee for their ongoing efforts and initiatives for creating an exciting program for ACM SIGMETRICS 2025.

短名	Proc. ACM Meas. Anal. Comput. Syst.
Journal Impact	2.19
国际分区	COMPUTER SCIENCE, INFORMATION SYSTEMS(Q3)

ISSN	2476-1249
h-index	10

出版信息	出版商: Association for Computing Machinery (ACM)，出版周期: ，期刊类型: journal，开源期刊: 否
基本数据	创刊年份: 2021，原创研究文献占比: ，自引率:， Gold OA占比: 0.00

Proceedings of the ACM on Measurement and Analysis of Computing Systems

期刊引文格式

有 1 位以上作者的期刊

有 2 位作者的期刊

有 3 位作者的期刊

有 5 位以上作者的期刊

书籍引用格式

学位论文引用格式

网页引用格式

专利引用格式

手工熬夜修改参考文献？研飞自动匹配期刊，一键轻松成稿，支持 Word/WPS

投稿经验分享

INT-MC: Low-Overhead In-Band Network-Wide Telemetry Based on Matrix Completion

Inferring Streaming Video Quality from Encrypted Traffic

Fair Resource Allocation in Virtualized O-RAN Platforms

NetDiffusion: Network Data Augmentation Through Protocol-Constrained Traffic Generation

The Price of Fragmentation in Mobility-on-Demand Services

PyGim : An Efficient Graph Neural Network Library for Real Processing-In-Memory Architectures

INT-MC: Low-Overhead In-Band Network-Wide Telemetry Based on Matrix Completion

Optimal Aggregation via Overlay Trees: Delay-MSE Tradeoffs under Failures

Confidential VMs Explained: An Empirical Analysis of AMD SEV-SNP and Intel TDX

POMACS V8, N3, December 2024 Editorial