Cloud Computing
Our research explores the design of data centers, develops systems to monitor cloud computing infrastructure, and understand the reliability and manageability of hyperscalars.
- SIGCOMMNear-optimal Online Traffic EngineeringArvin Ghavidel, Pooria Namyar, Nikolai Matni, and 2 more authorsIn Proceedings of the ACM SIGCOMM 2026 Conference, 2026
Most deployed WAN Traffic Engineering (TE) systems use a logically centralized controller that periodically gathers traffic demands, runs a TE optimization or heuristic, and then programs the network. At scale, these solutions are often suboptimal and can take minutes to react to demand changes or failures. In this paper, we introduce OnlineTE, a system that reacts immediately to demand changes and failures and delivers near-optimal solutions within seconds of a change. OnlineTE builds on the theory of optimization decomposition to devise scalable, near-optimal, distributed TE solvers for path-based MLU and Max-Flow problems. In OnlineTE, switches each solve a local subproblem, and a central coordinator coordinates their convergence. As such, a switch can trigger a re-optimization as soon as it detects a demand change or failure, enabling high reactivity. OnlineTE scales to large WANs, and its computational requirements are well within the capabilities of modern WAN switches. It also enables a novel paradigm, edge-based TE, which can utilize resources more efficiently than today’s path-based approaches. On a testbed emulation of a 750-node WAN topology, OnlineTE outperforms the state-of-the-art by up to an order of magnitude.
@inproceedings{ghavidel2026, author = {Ghavidel, Arvin and Namyar, Pooria and Matni, Nikolai and Willinger, Walter and Govindan, Ramesh}, title = {Near-optimal Online Traffic Engineering}, booktitle = {Proceedings of the ACM SIGCOMM 2026 Conference}, year = {2026}, series = {SIGCOMM '26}, pages = {1024–1045}, address = {New York, NY, USA}, publisher = {Association for Computing Machinery}, isbn = {9798400724671}, url = {https://doi.org/10.1145/3789240.3829123}, doi = {10.1145/3789240.3829123}, numpages = {22}, keywords = {traffic engineering, wide-area networks, network optimization, distributed and parallel algorithms}, location = {Colorado Convention Center, Denver, CO, USA}, topics = {dcn,measure} } - SIGCOMMZENITH: Towards A Formally Verified Highly-Available Control PlanePooria Namyar, Arvin Ghavidel, Mingyang Zhang, and 4 more authorsIn Proceedings of the ACM SIGCOMM 2025 Conference, 2025
Today, large-scale software-defined networks use microservice-based controllers. Bugs in these controllers can reduce network availability by making the data plane state inconsistent with the high-level intent. To recover from such inconsistencies, modern controllers periodically reconcile the state of all the switches with the desired intent. However, periodic reconciliation limits the availability and performance of the network at scale. We introduce Zenith, a microservice-based controller that avoids inconsistencies by design rather than always relying on recovery mechanisms. We have formally verified Zenith’s specifications and have proved that it ensures the network state will eventually be consistent with intent. We automatically generate Zenith’s code from its specification to minimize the likelihood of errors in the final implementation. Zenith’s guarantees and abstractions also enable developers to independently verify SDN applications and ensure end-to-end safety and correctness. Zenith resolves inconsistencies 5\texttimes faster than today’s designs and significantly improves availability.
@inproceedings{namyar.p.ghavidel.a.ea:zenith, author = {Namyar, Pooria and Ghavidel, Arvin and Zhang, Mingyang and Madhyastha, Harsha V. and Ravi, Srivatsan and Wang, Chao and Govindan, Ramesh}, title = {ZENITH: Towards A Formally Verified Highly-Available Control Plane}, booktitle = {Proceedings of the ACM SIGCOMM 2025 Conference}, year = {2025}, series = {SIGCOMM '25}, pages = {409–433}, address = {New York, NY, USA}, publisher = {Association for Computing Machinery}, isbn = {9798400715242}, url = {https://doi.org/10.1145/3718958.3750533}, doi = {10.1145/3718958.3750533}, numpages = {25}, keywords = {software defined networking, formal methods, availability}, topics = {dcn}, location = {S\~{a}o Francisco Convent, Coimbra, Portugal} } - SIGCOMMFirefly: Scalable, Ultra-Accurate Clock Synchronization for DatacentersPooria Namyar, Yuliang Li, Weitao Wang, and 10 more authorsIn Proceedings of the ACM SIGCOMM 2025 Conference, 2025
Cloud-based financial exchanges require sub-10ns device-to-device clock synchronization accuracy while adhering to Coordinated Universal Time (UTC). Existing clock sync techniques struggle to meet this demand at scale and are vulnerable to clock drift, jitter, and path asymmetries. Firefly, a software-driven datacenter clock sync system, scalably, cost-effectively, and reliably achieves very high clock sync accuracy. It employs a distributed consensus algorithm on a random overlay graph to rapidly converge to a common time while applying gradual adjustments to device hardware clocks. To realize consistent sync-to-UTC (external sync) across devices while maintaining a stable device-to-device internal sync, Firefly uses a novel technique, layered synchronization, that decouples internal and external syncs. In a 248-machine Clos network, Firefly achieves sub-10ns device-to-device and ≤1μs device-to-UTC sync, and is resilient to time server failure and unstable clocks.
@inproceedings{namyar.p.li.y.ea:firefly, author = {Namyar, Pooria and Li, Yuliang and Wang, Weitao and Dukkipati, Nandita and Yap, Kk and Gong, Junzhi and Chen, Chen and Gao, Peixuan and Ray, Devdeep and Kumar, Gautam and Ma, Yidan and Govindan, Ramesh and Vahdat, Amin}, title = {Firefly: Scalable, Ultra-Accurate Clock Synchronization for Datacenters}, booktitle = {Proceedings of the ACM SIGCOMM 2025 Conference}, year = {2025}, series = {SIGCOMM '25}, pages = {434–452}, address = {New York, NY, USA}, publisher = {Association for Computing Machinery}, isbn = {9798400715242}, url = {https://doi.org/10.1145/3718958.3750502}, doi = {10.1145/3718958.3750502}, numpages = {19}, topics = {dcn}, keywords = {clock synchronization, financial exchange, UTC synchronization, path asymmetry, distributed consensus}, location = {S\~{a}o Francisco Convent, Coimbra, Portugal} } - SigmetricsTiered Cloud Routing: Methodology, Latency, and ImprovementShihan Lin, Yi Zhou, Xiao Zhang, and 3 more authorsProc. ACM Meas. Anal. Comput. Syst., Mar 2025
Large cloud providers including AWS, Azure, and Google Cloud offer two tiers of network services to their customers: one class uses the providers’ private wide area networks (WAN-transit) to carry a customer’s traffic as much as possible, and the other uses the public internet (inet-transit). Little is known about how each cloud provider configures its network to offer different transit services, how well these services work, and whether the quality of those services can be further improved. In this work, we conduct a large-scale study to answer these questions. Using RIPE Atlas probes as vantage points, we explore how traffic enters and leaves each cloud’s WAN. In addition, we measure the access latency of the WAN-transit and the inet-transit service of each cloud and compare it with that of an emulated performance-based routing strategy. Our study shows that despite the cloud providers’ intention to carry customers’ traffic on its WAN to the maximum extent possible, for about 12% (Azure) and 13% (Google) of our vantage points, traffic exits the cloud WAN early at cloud edges more than 5000km away from the vantage points’ nearest cloud edges. In contrast, more than 84% (AWS), 78% (Azure), and 81% (Google) of vantage points enter a cloud WAN within a 500km radius of their respective locations. Moreover, we find that cloud providers employ different routing strategies to implement the inet-transit service, leading to transit policies that may deviate from their advertised service descriptions. Finally, we find that a performance-based routing strategy can significantly reduce latencies in all three cloud providers for 4% to 85% of vantage point and cloud region pairs.
@article{lin.s.zhou.y.ea:tiered, author = {Lin, Shihan and Zhou, Yi and Zhang, Xiao and Arnold, Todd and Govindan, Ramesh and Yang, Xiaowei}, title = {Tiered Cloud Routing: Methodology, Latency, and Improvement}, journal = {Proc. ACM Meas. Anal. Comput. Syst.}, year = {2025}, volume = {9}, number = {1}, month = mar, issue_date = {March 2025}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3711705}, doi = {10.1145/3711705}, articleno = {12}, numpages = {41}, topics = {dcn,measure}, keywords = {bgp, cloud routing, internet routing, wide area network} } - NSDIPreventing Network Bottlenecks: Accelerating Datacenter Services with Hotspot-Aware Placement for Compute and StorageHamid Bazzaz, Weiwu Pang, Yingjie Bi, and 10 more authorsIn 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25), Mar 2025
@inproceedings{bazzaz.h.pang.w.ea:preventing, author = {Bazzaz, Hamid and Pang, Weiwu and Bi, Yingjie and Yu, Minlan and Govindan, Ramesh and Cardwell, Neal and Dukkipati, Nandita and Tsai, Meng-Jung and DeForeest, Chris and Jin, Yuxue Jin and Carver, Charles and Kopański, Jan and Cheng, Liqun}, title = {Preventing Network Bottlenecks: Accelerating Datacenter Services with Hotspot-Aware Placement for Compute and Storage}, booktitle = {22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25)}, year = {2025}, topics = {dcn} } - HotOSGranular Resource Demand HeterogeneityYizhuo Liang, Ramesh Govindan, and Seo Jin ParkIn Proceedings of the 20th Workshop on Hot Topics in Operating Systems, Mar 2025
@inproceedings{liang.y.govindan.r.ea:granular, title = {Granular Resource Demand Heterogeneity}, author = {Liang, Yizhuo and Govindan, Ramesh and Park, Seo Jin}, booktitle = {Proceedings of the 20th Workshop on Hot Topics in Operating Systems}, year = {2025}, topics = {dcn} } - NSDIEnhancing Network Failure Mitigation with Performance-Aware RankingPooria Namyar, Arvin Ghavidel, Daniel Crankshaw, and 5 more authorsIn 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25), Mar 2025
@inproceedings{namyar.p.ghavidel.a.ea:enhancing, author = {Namyar, Pooria and Ghavidel, Arvin and Crankshaw, Daniel and Berger, Daniel S and Hsieh, Kevin and Kandula, Srikanth and Govindan, Ramesh and Arzani, Behnaz}, title = {Enhancing Network Failure Mitigation with Performance-Aware Ranking}, booktitle = {22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25)}, year = {2025}, topics = {dcn} } - HotCarbonLovelock: Towards Smart NIC-hosted ClustersSeo Jin Park, Ramesh Govindan, Kai Shen, and 4 more authorsIn In Proc. HotCarbon Workshop on Sustainable Computer Systems, Jul 2024
@inproceedings{park.sj.govindan.r.ea:lovelock--towards-smart-nic-hosted-clusters, author = {Park, Seo Jin and Govindan, Ramesh and Shen, Kai and Culler, David and Fatma, Ozcan and Kim, Geon-Woo and Levy, Hank}, title = {{Lovelock: Towards Smart NIC-hosted Clusters}}, booktitle = {In Proc. HotCarbon Workshop on Sustainable Computer Systems}, year = {2024}, topics = {dcn}, month = jul } - HotNetsEnd-to-End Performance Analysis of Learning-enabled SystemsPooria Namyar, Michael Schapira, Ramesh Govindan, and 4 more authorsIn Proceedings of the 23rd ACM Workshop on Hot Topics in Networks, Jul 2024
We propose a performance analysis tool for learning-enabled systems that allows operators to uncover potential performance issues before deploying DNNs in their systems. The tools that exist for this purpose require operators to faithfully model all components (a white-box approach) or do inefficient black-box local search. We propose a gray-box alternative, which eliminates the need to precisely model all the system’s components. Our approach is faster and finds substantially worse scenarios compared to prior work. We show that a state-of-the-art learning-enabled traffic engineering pipeline can underperform the optimal by 6\texttimes — a much higher number compared to what the authors found.
@inproceedings{namyar.p.schapira.m.ea:end-to-end, author = {Namyar, Pooria and Schapira, Michael and Govindan, Ramesh and Segarra, Santiago and Beckett, Ryan and Kakarla, Siva Kesava Reddy and Arzani, Behnaz}, title = {End-to-End Performance Analysis of Learning-enabled Systems}, booktitle = {Proceedings of the 23rd ACM Workshop on Hot Topics in Networks}, year = {2024}, series = {HOTNETS '24}, pages = {86–94}, address = {New York, NY, USA}, publisher = {Association for Computing Machinery}, isbn = {9798400712722}, url = {https://doi.org/10.1145/3696348.3696875}, doi = {10.1145/3696348.3696875}, numpages = {9}, keywords = {Machine Learning for Systems, Performance Analysis}, location = {Irvine, CA, USA}, topics = {dcn} } - NSDISolving Max-Min Fair Resource Allocations Quickly on Large GraphsPooria Namyar, Behnaz Arzani, Srikanth Kandula, and 5 more authorsIn 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 24), Jul 2024
We consider the max-min fair resource allocation problem. The best-known solutions use either a sequence of optimizations or waterfilling, which only applies to a narrow set of cases. These solutions have become a practical bottleneck in WAN traffic engineering and cluster scheduling, especially at larger problem sizes. We improve both approaches: (1) we show how to convert the optimization sequence into a single fast optimization, and (2) we generalize waterfilling to the multi-path case. We empirically show our new algorithms Pareto-dominate prior techniques: they produce faster, fairer, and more efficient allocations. Some of our allocators also have theoretical guarantees: they trade off a bounded amount of unfairness for faster allocation. We have deployed our allocators in Azure’s WAN traffic engineering pipeline, where we preserve solution quality and achieve a roughly 3× speedup.
@inproceedings{namyar.p.arzani.b.ea:solving, author = {Namyar, Pooria and Arzani, Behnaz and Kandula, Srikanth and Segarra, Santiago and Crankshaw, Daniel and Krishnaswamy, Umesh and Govindan, Ramesh and Raj, Himanshu}, title = {{S}olving {M}ax-{M}in {F}air {R}esource {A}llocations {Q}uickly on {L}arge {G}raphs}, booktitle = {21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 24)}, year = {2024}, topics = {dcn}, } - NSDIFinding Adversarial Inputs for Heuristics using Multi-level OptimizationPooria Namyar, Behnaz Arzani, Ryan Beckett, and 5 more authorsIn 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 24), Jul 2024
Production systems use heuristics because they are faster or scale better than their optimal counterparts. Yet, practitioners are often unaware of the performance gap between a heuristic and the optimum or between two heuristics in realistic scenarios. We present MetaOpt, a system that helps analyze heuristics. Users specify the heuristic and the optimal (or another heuristic) as input, and MetaOpt automatically encodes these efficiently for a solver to find performance gaps and their corresponding adversarial inputs. Its suite of built-in optimizations helps it scale its analysis to practical problem sizes. To show it is versatile, we used MetaOpt to analyze heuristics from three domains (traffic engineering, vector bin packing, and packet scheduling). We found a production traffic engineering heuristic can require 30% more capacity than the optimal to satisfy realistic demands. Based on the patterns in the adversarial inputs MetaOpt produced, we modified the heuristic to reduce its performance gap by 12.5×. We examined adversarial inputs to a vector bin packing heuristic and proved a new lower bound on its performance.
@inproceedings{namyar.p.arzani.b.ea:finding, author = {Namyar, Pooria and Arzani, Behnaz and Beckett, Ryan and Segarra, Santiago and Raj, Himanshu and Krishnaswamy, Umesh and Govindan, Ramesh and Kandula, Srikanth}, title = {{F}inding {A}dversarial {I}nputs for {H}euristics using {M}ulti-level {O}ptimization}, booktitle = {21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 24)}, year = {2024}, topics = {dcn}, } - ToNOptimal Oblivious Routing With Concave Objectives for Structured NetworksK. Chitavisutthivong, S. Supittayapornpong, P. Namyar, and 3 more authorsIEEE/ACM Transactions on Networking, Apr 2023
@article{chitavisutthivong.k.supittayapornpong.s.ea:optimal, author = {Chitavisutthivong, K. and Supittayapornpong, S. and Namyar, P. and Zhang, M. and Yu, M. and Govindan, R.}, journal = {IEEE/ACM Transactions on Networking}, title = {Optimal Oblivious Routing With Concave Objectives for Structured Networks}, year = {2023}, volume = {}, number = {01}, issn = {1558-2566}, pages = {1-13}, keywords = {routing;topology;network topology;servers;throughput;optimization;ieee transactions}, doi = {10.1109/TNET.2023.3264632}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, month = apr, topics = {dcn} } - NSDICloudCluster: Unearthing the Functional Structure of a Cloud ServiceWeiwu Pang, Sourav Panda, Jehangir Amjad, and 2 more authorsIn 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22), Apr 2022
@inproceedings{pang.w.panda.s.ea:cloudcluster, author = {Pang, Weiwu and Panda, Sourav and Amjad, Jehangir and Diot, Christophe and Govindan, Ramesh}, title = {{CloudCluster}: Unearthing the Functional Structure of a Cloud Service}, booktitle = {19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)}, year = {2022}, isbn = {978-1-939133-27-4}, address = {Renton, WA}, pages = {1213--1230}, url = {https://www.usenix.org/conference/nsdi22/presentation/pang}, publisher = {USENIX Association}, topics = {dcn} } - SoCCQuadrant: A Cloud-Deployable NF Virtualization PlatformJianfeng Wang, Tamás Lévai, Zhuojin Li, and 3 more authorsIn SoCC ’22: Proceedings of the ACM Symposium on Cloud Computing, Apr 2022
Network Functions (NFs) now touch a significant fraction of Internet traffic. The hope has been that software-based NF Virtualization (NFV) would enable rapid development of new NFs by vendors and leverage the power and economics of commodity computing infrastructure for NF deployment. To date, no cloud NFV systems achieve NF chaining, isolation, SLO-adherence, and scaling together with existing cloud computing infrastructure and abstractions, all while achieving generality, speed, and ease of deployment; these properties are taken for granted in other cloud contexts but unavailable for NF processing. We present Quadrant, an efficient and secure cloud-deployable NFV system, and show that Quadrant’s approach of adapting existing cloud infrastructure to support packet processing can achieve NF chaining, isolation, generality, and performance in NFV. Quadrant reuses common cloud infrastructure such as Kubernetes, cloud functions, the Linux kernel, NIC hardware, and switches. It enables easy NFV deployment while delivering up to double the performance per core compared to the state of the art.
@inproceedings{wang.j.lévai.t.ea:quadrant, title = {Quadrant: A Cloud-Deployable NF Virtualization Platform}, author = {Wang, Jianfeng and Lévai, Tamás and Li, Zhuojin and Vieira, Marcos A. M. and Govindan, Ramesh and Raghavan, Barath}, url = {https://dl.acm.org/doi/10.1145/3542929.3563471}, talk = {https://youtu.be/70_bUua5870}, year = {2022}, date = {2022-11-07}, booktitle = {SoCC '22: Proceedings of the ACM Symposium on Cloud Computing}, keywords = {NFV, cloud computing, SLO}, pubstate = {to appear}, tppubtype = {inproceedings}, topics = {dcn} } - InfocomOptimal Oblivious Routing for Structured NetworksSucha Supittayapornpong, Pooria Namyar, Mingyang Zhang, and 2 more authorsIn IEEE INFOCOM 2022 - IEEE Conference on Computer Communications, Apr 2022
@inproceedings{supittayapornpong.s.namyar.p.ea:optimal, author = {Supittayapornpong, Sucha and Namyar, Pooria and Zhang, Mingyang and Yu, Minlan and Govindan, Ramesh}, booktitle = {IEEE INFOCOM 2022 - IEEE Conference on Computer Communications}, title = {Optimal Oblivious Routing for Structured Networks}, year = {2022}, pages = {1988-1997}, doi = {10.1109/INFOCOM48880.2022.9796682}, topics = {dcn} } - SIGCOMMA throughput-centric view of the performance of datacenter topologiesPooria Namyar, Sucha Supittayapornpong, Mingyang Zhang, and 2 more authorsIn ACM SIGCOMM 2021 Conference, Virtual Event, USA, August 23-27, 2021, Apr 2021
@inproceedings{namyar.p.supittayapornpong.s.ea:throughput-centric, author = {Namyar, Pooria and Supittayapornpong, Sucha and Zhang, Mingyang and Yu, Minlan and Govindan, Ramesh}, editor = {Kuipers, Fernando A. and Caesar, Matthew C.}, title = {A throughput-centric view of the performance of datacenter topologies}, booktitle = {{ACM} {SIGCOMM} 2021 Conference, Virtual Event, USA, August 23-27, 2021}, pages = {349--369}, publisher = {{ACM}}, year = {2021}, url = {https://doi.org/10.1145/3452296.3472913}, doi = {10.1145/3452296.3472913}, timestamp = {Wed, 11 Aug 2021 15:56:33 +0200}, topics = {dcn}, biburl = {https://dblp.org/rec/conf/sigcomm/NamyarS0YG21.bib} } - ArXivGemini: Practical Reconfigurable Datacenter Networks with Topology and Traffic EngineeringMingyang Zhang, Jianan Zhang, Rui Wang, and 3 more authorsCoRR, Apr 2021
@article{zhang.m.zhang.j.ea:gemini, author = {Zhang, Mingyang and Zhang, Jianan and Wang, Rui and Govindan, Ramesh and Mogul, Jeffrey C. and Vahdat, Amin}, title = {Gemini: Practical Reconfigurable Datacenter Networks with Topology and Traffic Engineering}, journal = {CoRR}, volume = {abs/2110.08374}, year = {2021}, eprinttype = {arXiv}, eprint = {2110.08374}, timestamp = {Thu, 03 Feb 2022 07:26:25 +0100}, biburl = {https://dblp.org/rec/journals/corr/abs-2110-08374.bib}, bibsource = {dblp computer science bibliography, https://dblp.org}, topics = {dcn} } - SIGCOMMTowards Highly Available Clos-Based WAN RoutersSucha Supittayapornpong, Barath Raghavan, and Ramesh GovindanApr 2019
@conference{supittayapornpong.s.raghavan.b.ea:towards, title = {Towards Highly Available Clos-Based WAN Routers}, author = {Supittayapornpong, Sucha and Raghavan, Barath and Govindan, Ramesh}, url = {https://dl.acm.org/citation.cfm?id=3342086}, year = {2019}, date = {2019-08-19}, booktitle = {ACM SIGCOMM Conference on Network Architectures and Protocols}, pubstate = {published}, topics = {dcn}, tppubtype = {conference}, } - NSDIUnderstanding Lifecycle Management Complexity of Datacenter TopologiesMingyang Zhang, Radhika Niranjan Mysore, Sucha Supittayapornpong, and 1 more authorIn 16th USENIX Symposium on Networked Systems Design and Implementation (NSDI), Apr 2019
@inproceedings{zhang.m.mysore.rn.ea:understanding, title = {Understanding Lifecycle Management Complexity of Datacenter Topologies}, author = {Zhang, Mingyang and Mysore, Radhika Niranjan and Supittayapornpong, Sucha and Govindan, Ramesh}, year = {2019}, date = {2019-02-26}, booktitle = {16th USENIX Symposium on Networked Systems Design and Implementation (NSDI)}, pubstate = {published}, topics = {dcn}, tppubtype = {conference}, } - SIGCOMMEvolve or Die: High-Availability Design Principles Drawn from Google’s Network InfrastructureRamesh Govindan, Ina Minei, Mahesh Kallahalla, and 2 more authorsIn Proceedings of the ACM Conference of the Special Interest Group on Data Communication (SIGCOMM ’16) , au 2016
@inproceedings{govindan.r.minei.i.ea:evolve-or-die--high-availability-design-principles-drawn-from-googles-network-infrastructure, author = {Govindan, Ramesh and Minei, Ina and Kallahalla, Mahesh and Koley, Bikash and Vahdat, Amin}, title = { {Evolve or Die: High-Availability Design Principles Drawn from Google’s Network Infrastructure} }, booktitle = { Proceedings of the ACM Conference of the Special Interest Group on Data Communication (SIGCOMM '16) }, month = au, location = { Florianópolis, Brazil }, topics = {dcn,measure}, year = {2016} }