Contact The DL Team Contact Us | Switch to tabbed view

top of pageABSTRACT

Recently, much attention has been devoted to the question of whether/when traditional network protocol design, which relies on the application of algorithmic insights by human experts, can be replaced by a data-driven (i.e., machine learning) approach. We explore this question in the context of the arguably most fundamental networking task: routing. Can ideas and techniques from machine learning (ML) be leveraged to automatically generate "good" routing configurations? We focus on the classical setting of intradomain traffic engineering. We observe that this context poses significant challenges for data-driven protocol design. Our preliminary results regarding the power of data-driven routing suggest that applying ML (specifically, deep reinforcement learning) to this context yields high performance and is a promising direction for further research. We outline a research agenda for ML-guided routing.

top of pageSOURCE MATERIALS

AVAILABLE FOR DOWNLOAD
Buy this ArticleBuy this Article
PDFPDF (618 KB)  
Mp4Mp4 (19:45) (772.58 MB)  
AVAILABLE via STREAMING
Streaming Service
Play Stream (19:45)   
Buy this ArticleBuy this Article
APPEARS IN
Networking

top of pageAUTHORS



Author image not provided  Asaf Valadarsky

No contact information provided yet.

Bibliometrics: publication history
Publication years2014-2019
Publication count8
Citation Count67
Available for download6
Downloads (6 Weeks)1,244
Downloads (12 Months)1,888
Downloads (cumulative)4,395
Average downloads per article732.50
Average citations per article8.38
View colleagues of Asaf Valadarsky


Author image not provided  Michael Schapira

No contact information provided yet.

Bibliometrics: publication history
Publication years2005-2019
Publication count78
Citation Count947
Available for download59
Downloads (6 Weeks)1,529
Downloads (12 Months)3,724
Downloads (cumulative)20,414
Average downloads per article346.00
Average citations per article12.14
View colleagues of Michael Schapira


Author image not provided  Dafna Shahaf

No contact information provided yet.

Bibliometrics: publication history
Publication years2006-2019
Publication count25
Citation Count235
Available for download18
Downloads (6 Weeks)282
Downloads (12 Months)2,435
Downloads (cumulative)25,631
Average downloads per article1,423.94
Average citations per article9.40
View colleagues of Dafna Shahaf


Author image not provided  Aviv Tamar

No contact information provided yet.

Bibliometrics: publication history
Publication years2011-2018
Publication count18
Citation Count78
Available for download8
Downloads (6 Weeks)57
Downloads (12 Months)494
Downloads (cumulative)1,041
Average downloads per article130.13
Average citations per article4.33
View colleagues of Aviv Tamar

top of pageREFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
DeepMind AI Reduces Google Data Centre Cooling Bill by 40%. https://goo.gl/QTdU2T.
 
2
IBM ILOG CPLEX Optimizer. http://www-01.ibm.com/software/commerce/optimization/cplex-optimizer/index.html.
 
3
4
 
5
M. Allalouf and Y. Shavitt. Maximum flow routing with weighted max-min fairness. In Quality of Service in the Emerging Networking Panorama.
 
6
D. Andersen, H. Balakrishnan, F. Kaashoek, and R. Morris. Resilient overlay networks. 2001.
 
7
Y. Azar, E. Cohen, A. Fiat, H. Kaplan, and H. Racke. Optimal oblivious routing in polynomial time. STOC, 2003.
 
8
D. Bahdanau, K. Cho, and Y. Bengio. Neural machine translation by jointly learning to align and translate. arXiv, 2014.
 
9
10
 
11
T. W. S. Chow and C. T. Leung. Nonlinear autoregressive integrated neural network model for short-term load forecasting. IEE Proceedings - Generation, Transmission and Distribution, 1996.
 
12
 
13
14
 
15
 
16
 
17
18
19
20
 
21
K. Janocha and W. M. Czarnecki. On loss functions for deep neural networks in classification. CoRR, 2017.
 
22
W. S. Jewell. Multi-commodity Network Solutions. 1966.
 
23
J. Jiang, V. Sekar, I. Stoica, and H. Zhang. Unleashing the potential of data-driven networking. COMSNETS, 2017.
 
24
S. Kandula, D. Katabi, B. Davie, and A. Charny. Walking the tightrope: Responsive yet stable traffic engineering. 2005.
25
 
26
S. Knight, H. Nguyen, N. Falkner, R. Bowden, and M. Roughan. The internet topology zoo. IEEE Journal on Selected Areas in Communications, 2011.
 
27
 
28
 
29
P. Kumar, Y. Yuan, C. Yu, N. Foster, R. D. Kleinberg, and R. Soulé. Kulfi: Robust traffic engineering using semi-oblivious routing. CoRR, 2016.
 
30
Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 1998.
 
31
32
33
 
34
A. Medina, N. Taft, K. Salamatian, S. Bhattacharyya, and C. Diot. Traffic matrix estimation: Existing techniques and new directions. 2002.
 
35
 
36
 
37
V. Mnih, K. Kavukcuoglu, D. Silver, A. Graves, I. Antonoglou, D. Wierstra, and M. A. Riedmiller. Playing atari with deep reinforcement learning. CoRR, 2013.
 
38
V. Mnih, K. Kavukcuoglu, D. Silver, A. A. Rusu, J. Veness, M. G. Bellemare, A. Graves, M. Riedmiller, A. K. Fidjeland, G. Ostrovski, et al. Human-level control through deep reinforcement learning. Nature, 2015.
 
39
40
 
41
 
42
J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov. Proximal Policy Optimization Algorithms. ArXiv e-prints, 2017.
43
 
44
D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, G. van den Driessche, J. Schrittwieser, I. Antonoglou, V. Panneershelvam, M. Lanctot, S. Dieleman, D. Grewe, J. Nham, N. Kalchbrenner, I. Sutskever, T. Lillicrap, M. Leach, K. Kavukcuoglu, T. Graepel, and D. Hassabis. Mastering the game of Go with deep neural networks and tree search. Nature, 2016.
 
45
 
46
 
47
C. J. Watkins and P. Dayan. Q-learning. Machine Learning, 1992.
48
 
49
Y. Wu, E. Mansimov, S. Liao, R. B. Grosse, and J. Ba. Scalable trust-region method for deep reinforcement learning using kronecker-factored approximation. ArXiv e-prints, 2017.
 
50
D. Xu, M. Chiang, and J. Rexford. Link-state routing with hop-by-hop forwarding can achieve optimal traffic engineering. INFOCOM, 2008.
51

top of pageCITED BY

top of pageINDEX TERMS

Index Terms are not available

top of pagePUBLICATION

Title HotNets-XVI Proceedings of the 16th ACM Workshop on Hot Topics in Networks table of contents
Pages 185-191
Publication Date2017-11-30 (yyyy-mm-dd)
Funding Source Israeli Centers for Research Excellence
Sponsors SIGCOMM ACM Special Interest Group on Data Communication
CISCO
PublisherACM New York, NY, USA ©2017
ISBN: 978-1-4503-5569-8 doi>10.1145/3152434.3152441
Conference COMMACM SIGCOMM COMM logo
Paper Acceptance Rate 28 of 124 submissions, 23%
Overall Acceptance Rate 179 of 803 submissions, 22%
Year Submitted Accepted Rate
Hotnets '10 104 22 21%
HotNets '11 119 24 20%
HotNets-XI 120 23 19%
HotNets-XII 110 26 24%
HotNets-XIII 118 26 22%
HotNets '16 108 30 28%
HotNets-XVI 124 28 23%
Overall 803 179 22%

APPEARS IN
Networking

top of pageREVIEWS


Reviews are not available for this item
Computing Reviews logo

top of pageCOMMENTS

Be the first to comment To Post a comment please sign in or create a free Web account

top of pageTable of Contents

Proceedings of the 16th ACM Workshop on Hot Topics in Networks
Table of Contents
SESSION: Security, Privacy, and Censorship
DIY Hosting for Online Privacy
Shoumik Palkar, Matei Zaharia
Pages: 1-7
doi>10.1145/3152434.3152459
Full text: PDFPDF
Other formats: Mp4Mp4

Web users today rely on centralized services for applications such as email, file transfer and chat. Unfortunately, these services create a significant privacy risk: even with a benevolent provider, a single breach can put millions of users' data at ...
expand
Online Advertising under Internet Censorship
Hira Javaid, Hafiz Kamran Khalil, Zartash Afzal Uzmi, Ihsan Ayyub Qazi
Pages: 8-14
doi>10.1145/3152434.3152455
Full text: PDFPDF
Other formats: Mp4Mp4

Online advertising plays a critical role in enabling the free Web by allowing publishers to monetize their services. However, the rise in internet censorship events globally poses an economic threat to the advertising ecosystem. This paper studies this ...
expand
The Case For Secure Delegation
Dmitry Kogan, Henri Stern, Ashley Tolbert, David Mazières, Keith Winstein
Pages: 15-21
doi>10.1145/3152434.3152444
Full text: PDFPDF
Other formats: Mp4Mp4

Today's secure stream protocols, SSH and TLS, were designed for end-to-end security and do not include a role for semi-trusted third parties. As a result, users who wish to delegate some of their authority to third parties (e.g., to run SSH clients in ...
expand
Securing Ultra-High-Bandwidth Science DMZ Networks with Coordinated Situational Awareness
Vasudevan Nagendra, Vinod Yegneswaran, Phillip Porras
Pages: 22-28
doi>10.1145/3152434.3152460
Full text: PDFPDF
Other formats: Mp4Mp4

The Science DMZ (SDMZ) is a special purpose network infrastructure that is engineered to cater to the ultra-high bandwidth needs of the scientific and high performance computing (HPC) communities. These networks are isolated from stateful security devices ...
expand
SESSION: Wireless
Rethinking Congestion Control for Cellular Networks
Prateesh Goyal, Mohammad Alizadeh, Hari Balakrishnan
Pages: 29-35
doi>10.1145/3152434.3152437
Full text: PDFPDF
Other formats: Mp4Mp4

We propose Accel-Brake Control (ABC), a protocol that integrates a simple and deployable signaling scheme at cellular base stations with an endpoint mechanism to respond to these signals. The key idea is for the base station to enable each sender to ...
expand
Programmable Radio Environments for Smart Spaces
Allen Welkie, Longfei Shangguan, Jeremy Gummeson, Wenjun Hu, Kyle Jamieson
Pages: 36-42
doi>10.1145/3152434.3152456
Full text: PDFPDF
Other formats: Mp4Mp4

Smart spaces, such as smart homes and smart offices, are common Internet of Things (IoT) scenarios for building automation with networked sensors. In this paper, we suggest a different notion of smart spaces, where the radio environment is programmable ...
expand
Wi-Fly: Widespread Opportunistic Connectivity via Commercial Air Transport
Talal Ahmad, Ranveer Chandra, Ashish Kapoor, Michael Daum, Eric Horvitz
Pages: 43-49
doi>10.1145/3152434.3152458
Full text: PDFPDF
Other formats: Mp4Mp4

More than half of the world's population face barriers in accessing the Internet. A recent ITU study estimates that 2.6 billion people cannot afford connectivity and that 3.8 billion do not have access. Recent proposals for providing low-cost connectivity ...
expand
SESSION: Video
360° Innovations for Panoramic Video Streaming
Xing Liu, Qingyang Xiao, Vijay Gopalakrishnan, Bo Han, Feng Qian, Matteo Varvello
Pages: 50-56
doi>10.1145/3152434.3152443
Full text: PDFPDF

360-degree videos are becoming increasingly popular on commercial platforms. In this position paper, we propose a holistic research agenda aiming at improving the performance, resource utilization efficiency, and users' quality of experience (QoE) for ...
expand
How will Deep Learning Change Internet Video Delivery?
Hyunho Yeo, Sunghyun Do, Dongsu Han
Pages: 57-64
doi>10.1145/3152434.3152440
Full text: PDFPDF
Other formats: Mp4Mp4
SESSION: Refactoring Distributed Applications
Network Stack as a Service in the Cloud
Zhixiong Niu, Hong Xu, Dongsu Han, Peng Cheng, Yongqiang Xiong, Guo Chen, Keith Winstein
Pages: 65-71
doi>10.1145/3152434.3152442
Full text: PDFPDF
Other formats: Mp4Mp4

The tenant network stack is implemented inside the virtual machines in today's public cloud. This legacy architecture presents a barrier to protocol stack innovation due to the tight coupling between the network stack and the guest OS. In particular, ...
expand
The Barriers to Overthrowing Internet Feudalism
Tai Liu, Zain Tariq, Jay Chen, Barath Raghavan
Pages: 72-79
doi>10.1145/3152434.3152454
Full text: PDFPDF
Other formats: Mp4Mp4

Today's Internet scarcely resembles the mythological image of it as a fundamentally democratic system. Instead, users are at the whims of a small number of providers who control nearly everything about users' experiences on the Internet. In response, ...
expand
SESSION: Measurement
FreeLab: A Free Experimentation Platform
Matteo Varvello, Diego Perino
Pages: 80-86
doi>10.1145/3152434.3152436
Full text: PDFPDF

As researchers, we are aware of how hard it is to obtain access to vantage points in the Internet. Experimentation platforms are useful tools, but they are also: 1) paid, either via a membership fee or by resource sharing, 2) unreliable, nodes come and ...
expand
Opportunities and Challenges of Ad-based Measurements from the Edge of the Network
Patricia Callejo, Conor Kelton, Narseo Vallina-Rodriguez, Rubén Cuevas, Oliver Gasser, Christian Kreibich, Florian Wohlfart, Ángel Cuevas
Pages: 87-93
doi>10.1145/3152434.3152895
Full text: PDFPDF
Other formats: Mp4Mp4

For many years, the research community, practitioners, and regulators have used myriad methods and tools to understand the complex structure and behavior of ISPs from the edge of the network. Unfortunately, the nature of these techniques forces the researcher ...
expand
Stick a fork in it: Analyzing the Ethereum network partition
Lucianna Kiffer, Dave Levin, Alan Mislove
Pages: 94-100
doi>10.1145/3152434.3152449
Full text: PDFPDF
Other formats: Mp4Mp4

As blockchain technologies and cryptocurrencies increase in popularity, their decentralization poses unique challenges in network partitions. In traditional distributed systems, network partitions are generally a result of bugs or connectivity failures; ...
expand
SESSION: Congestion Control
The Case for Moving Congestion Control Out of the Datapath
Akshay Narayan, Frank Cangialosi, Prateesh Goyal, Srinivas Narayana, Mohammad Alizadeh, Hari Balakrishnan
Pages: 101-107
doi>10.1145/3152434.3152438
Full text: PDFPDF
Other formats: Mp4Mp4

With Moore's law ending, the gap between general-purpose processor speeds and network link rates is widening. This trend has led to new packet-processing "datapaths" in endpoints, including kernel bypass software and emerging SmartNIC hardware. In addition, ...
expand
HotCocoa: Hardware Congestion Control Abstractions
Mina Tahmasbi Arashloo, Monia Ghobadi, Jennifer Rexford, David Walker
Pages: 108-114
doi>10.1145/3152434.3152457
Full text: PDFPDF
Other formats: Mp4Mp4

Congestion control in multi-tenant data centers is an active area of research because of its significant impact on customer experience, and, consequently, on revenue. Therefore, new algorithms and protocols are expected to emerge as the Cloud evolves. ...
expand
An Axiomatic Approach to Congestion Control
Doron Zarchy, Radhika Mittal, Michael Schapira, Scott Shenker
Pages: 115-121
doi>10.1145/3152434.3152445
Full text: PDFPDF
Other formats: Mp4Mp4

Recent years have witnessed a surge of interest in congestion control. Unfortunately, the overwhelmingly large design space along with the increasingly diverse range of application environments makes evaluating congestion control protocols a daunting ...
expand
Congestion-Control Throwdown
Michael Schapira, Keith Winstein
Pages: 122-128
doi>10.1145/3152434.3152446
Full text: PDFPDF
Other formats: Mp4Mp4

Congestion control is a perennial topic of networking research. In making decisions about who sends data when, congestion-control schemes prevent collapses and ultimately determine the allocation of scarce communications resources among contending users ...
expand
SESSION: The Control Plane
Integrating Verification and Repair into the Control Plane
Aaron Gember-Jacobson, Costin Raiciu, Laurent Vanbever
Pages: 129-135
doi>10.1145/3152434.3152439
Full text: PDFPDF
Other formats: Mp4Mp4

Network verification has made great progress recently, yet existing solutions are limited in their ability to handle specific protocols or implementation quirks or to diagnose and repair the cause of policy violations. In this positioning paper, we examine ...
expand
Low-Latency Routing on Mesh-Like Backbones
Nikola Gvozdiev, Stefano Vissicchio, Brad Karp, Mark Handley
Pages: 136-142
doi>10.1145/3152434.3152453
Full text: PDFPDF
Other formats: Mp4Mp4

Early in in the Internet's history, routing within a single provider's WAN centered on placing traffic on the shortest path. More recent traffic engineering efforts aim to reduce congestion and/or increase utilization within the status quo of greedy ...
expand
Run, Walk, Crawl: Towards Dynamic Link Capacities
Rachee Singh, Monia Ghobadi, Klaus-Tycho Foerster, Mark Filer, Phillipa Gill
Pages: 143-149
doi>10.1145/3152434.3152451
Full text: PDFPDF
Other formats: Mp4Mp4

Fiber optic cables are the workhorses of today's Internet services. Operators spend millions of dollars to purchase, lease and maintain their optical backbone, making the efficiency of fiber essential to their business. In this work, we make a case for ...
expand
SESSION: Data Centers
In-Network Computation is a Dumb Idea Whose Time Has Come
Amedeo Sapio, Ibrahim Abdelaziz, Abdulla Aldilaijan, Marco Canini, Panos Kalnis
Pages: 150-156
doi>10.1145/3152434.3152461
Full text: PDFPDF
Other formats: Mp4Mp4

Programmable data plane hardware creates new opportunities for infusing intelligence into the network. This raises a fundamental question: what kinds of computation should be delegated to the network? In this paper, we discuss the opportunities and challenges ...
expand
Granular Computing and Network Intensive Applications: Friends or Foes?
Arjun Singhvi, Sujata Banerjee, Yotam Harchol, Aditya Akella, Mark Peek, Pontus Rydin
Pages: 157-163
doi>10.1145/3152434.3152450
Full text: PDFPDF
Other formats: Mp4Mp4

Computing/infrastructure as a service continues to evolve with bare metal, virtual machines, containers and now serverless granular computing service offerings. Granular computing enables developers to decompose their applications into smaller logical ...
expand
Tolerating Faults in Disaggregated Datacenters
Amanda Carbonari, Ivan Beschasnikh
Pages: 164-170
doi>10.1145/3152434.3152447
Full text: PDFPDF
Other formats: Mp4Mp4

Recent research shows that disaggregated datacenters (DDCs) are practical and that DDC resource modularity will benefit both users and operators. This paper explores the implications of disaggregation on application fault tolerance. We expect that resource ...
expand
Stop Rerouting!: Enabling ShareBackup for Failure Recovery in Data Center Networks
Yiting Xia, Xin Sunny Huang, T. S. Eugene Ng
Pages: 171-177
doi>10.1145/3152434.3152452
Full text: PDFPDF
Other formats: Mp4Mp4

This paper introduces sharable backup as a novel solution to failure recovery in data center networks. It allows the entire network to share a small pool of backup devices. This proposal is grounded in three key observations. First, the traditional rerouting-based ...
expand
SESSION: Machine Learning
Harvesting Randomness to Optimize Distributed Systems
Mathias Lecuyer, Joshua Lockerman, Lamont Nelson, Siddhartha Sen, Amit Sharma, Aleksandrs Slivkins
Pages: 178-184
doi>10.1145/3152434.3152435
Full text: PDFPDF
Other formats: Mp4Mp4

We view randomization through the lens of statistical machine learning: as a powerful resource for offline optimization. Cloud systems make randomized decisions all the time (e.g., in load balancing), yet this randomness is rarely used for optimization ...
expand
Learning to Route
Asaf Valadarsky, Michael Schapira, Dafna Shahaf, Aviv Tamar
Pages: 185-191
doi>10.1145/3152434.3152441
Full text: PDFPDF
Other formats: Mp4Mp4

Recently, much attention has been devoted to the question of whether/when traditional network protocol design, which relies on the application of algorithmic insights by human experts, can be replaced by a data-driven (i.e., machine learning) approach. ...
expand
Biases in Data-Driven Networking, and What to Do About Them
Mihovil Bartulovic, Junchen Jiang, Sivaraman Balakrishnan, Vyas Sekar, Bruno Sinopoli
Pages: 192-198
doi>10.1145/3152434.3152448
Full text: PDFPDF
Other formats: Mp4Mp4

Recent efforts highlight the promise of data-driven approaches to optimize network decisions. Many such efforts use trace-driven evaluation; i.e., running offline analysis on network traces to estimate the potential benefits of different policies before ...
expand

Powered by The ACM Guide to Computing Literature


The ACM Digital Library is published by the Association for Computing Machinery. Copyright © 2019 ACM, Inc.
Terms of Usage   Privacy Policy   Code of Ethics   Contact Us