Ramesh Govindan

@inproceedings{nokey,

title = {CloudCluster: Unearthing the Functional Structure of a Cloud Service},

author = {Weiwu Pang and Sourav Panda and Jehangir Amjad and Christophe Diot and Ramesh Govindan},

year  = {2022},

date = {2022-04-01},

urldate = {2022-04-01},

booktitle = {19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)},

publisher = {USENIX Association},

address = {Renton, WA},

keywords = {},

pubstate = {forthcoming},

tppubtype = {inproceedings}

}

@inproceedings{10.1145/3452296.3472913,

title = {A Throughput-Centric View of the Performance of Datacenter Topologies},

author = {Pooria Namyar and Sucha Supittayapornpong and Mingyang Zhang and Minlan Yu and Ramesh Govindan},

url = {https://nsl.usc.edu/wp-content/uploads/2021/09/throughput_sigcomm21.pdf, Paper

https://github.com/USC-NSL/TUB, Code

https://nsl.usc.edu/wp-content/uploads/2021/09/throughput_sigcomm21_short_slides.pdf, Short Slides

https://nsl.usc.edu/wp-content/uploads/2021/09/throughput_sigcomm21_long_slides.pdf, Long Slides},

doi = {10.1145/3452296.3472913},

isbn = {9781450383837},

year  = {2021},

date = {2021-08-09},

urldate = {2021-08-09},

booktitle = {Proceedings of the 2021 ACM SIGCOMM 2021 Conference},

pages = {349–369},

publisher = {Association for Computing Machinery},

address = {Virtual Event, USA},

series = {SIGCOMM '21},

abstract = {While prior work has explored many proposed datacenter designs, only two designs,

Clos-based and expander-based, are generally considered practical because they can

scale using commodity switching chips. Prior work has used two different metrics,

bisection bandwidth and throughput, for evaluating these topologies at scale. Little

is known, theoretically or practically, how these metrics relate to each other. Exploiting

characteristics of these topologies, we prove an upper bound on their throughput,

then show that this upper bound better estimates worst-case throughput than all previously

proposed throughput estimators and scales better than most of them. Using this upper

bound, we show that for expander-based topologies, unlike Clos, beyond a certain size

of the network, no topology can have full throughput, even if it has full bisection

bandwidth; in fact, even relatively small expander-based topologies fail to achieve

full throughput. We conclude by showing that using throughput to evaluate datacenter

performance instead of bisection bandwidth can alter conclusions in prior work about

datacenter cost, manageability, and reliability.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{10.1145/3472305.3472314,

title = {Tools for Disambiguating RFCs},

author = {Jane Yen and Ramesh Govindan and Barath Raghavan},

url = {https://doi.org/10.1145/3472305.3472314, Paper},

doi = {10.1145/3472305.3472314},

isbn = {9781450386180},

year  = {2021},

date = {2021-07-24},

urldate = {2021-07-24},

booktitle = {Proceedings of the Applied Networking Research Workshop},

pages = {85–91},

publisher = {Association for Computing Machinery},

address = {Virtual Event, USA},

series = {ANRW '21},

abstract = {For decades, drafting Internet protocols has taken significant amounts of human supervision 

due to the fundamental ambiguity of natural language. Given such ambiguity, it is 

also not surprising that protocol implementations have long exhibited bugs. This pain 

and overhead can be significantly reduced with the help of natural language processing 

(NLP).We recently applied NLP to identify ambiguous or under-specified sentences in 

RFCs, and to generate protocol implementations automatically when the ambiguity is 

clarified. However this system is far from general or deployable. To further reduce 

the overhead and errors due to ambiguous sentences, and to improve the generality 

of this system, much work remains to be done. In this paper, we consider what it would 

take to produce a fully-general and useful system for easing the natural-language 

challenges in the RFC process.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{RimIoTDI,

title = {Rim: Offloading Inference to the Edge},

author = {Yitao Hu and Weiwu Pang and Xiaochen Liu and Rajrup Ghosh and Bongjun Ko and Wei-Han Lee and Ramesh Govindan},

url = {https://conferences.computer.org/iotDI/2021/program.html},

year  = {2021},

date = {2021-05-18},

booktitle = {Proceedings of the 6th ACM/IEEE Conference on Internet of Things Design and Implementation, 2021},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{10.1145/3452296.3472910,

title = {Semi-Automated Protocol Disambiguation and Code Generation},

author = {Jane Yen and Tamás Lévai and Qinyuan Ye and Xiang Ren and Ramesh Govindan and Barath Raghavan},

url = {https://doi.org/10.1145/3452296.3472910, Paper

https://github.com/USC-NSL/sage, Code},

isbn = {9781450383837},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

booktitle = {Proceedings of the 2021 ACM SIGCOMM 2021 Conference},

pages = {272–286},

publisher = {Association for Computing Machinery},

address = {Virtual Event, USA},

series = {SIGCOMM '21},

abstract = {For decades, Internet protocols have been specified using natural language. Given 

the ambiguity inherent in such text, it is not surprising that protocol implementations 

have long exhibited bugs. In this paper, we apply natural language processing (NLP) 

to effect semi-automated generation of protocol implementations from specification 

text. Our system, Sage, can uncover ambiguous or under-specified sentences in specifications; 

once these are clarified by the author of the protocol specification, Sage can generate 

protocol code automatically.Using Sage, we discover 5 instances of ambiguity and 6 

instances of under-specification in the ICMP RFC; after fixing these, Sage is able 

to automatically generate code that interoperates perfectly with Linux implementations. 

We show that Sage generalizes to sections of BFD, IGMP, and NTP and identify additional 

conceptual components that Sage needs to support to generalize to complete, complex 

protocols like BGP and TCP.},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{Liu2020,

title = {Grab: Fast and Accurate Sensor Processing for Cashier-Free Shopping},

author = {Xiaochen Liu and Yurong Jiang and Kyu-Han Kim and Ramesh Govindan},

url = {https://arxiv.org/abs/2001.01033},

year  = {2020},

date = {2020-01-07},

publisher = {arXiv preprint arXiv:2001.01033},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{246280,

title = {CarMap-Fast 3D Feature Map Updates for Automobiles},

author = {Fawad Ahmad and Hang Qiu and Ray Eells and Fan Bai and Ramesh Govindan},

url = {https://nsl.usc.edu/wp-content/uploads/2020/02/nsdi20-paper-ahmad.pdf, Paper

https://github.com/USC-NSL/CarMap, Code

https://www.youtube.com/watch?v=SlG4QGq5ypk, Video Demo

https://nsl.usc.edu/wp-content/uploads/2020/02/nsdi20_slides_ahmad.pdf, Slides

},

year  = {2020},

date = {2020-01-01},

booktitle = {17th USENIX Symposium on Networked Systems Design and Implementation (NSDI 20)},

publisher = {USENIX Association},

address = {Santa Clara, CA},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ghosh2020iros,

title = {Persistent Connected Power Constrained Surveillance with Unmanned Aerial Vehicles},

author = {Pradipta Ghosh and Paulo Tabuada and Ramesh Govindan and Gaurav S Sukhatme},

year  = {2020},

date = {2020-01-01},

booktitle = {Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{ghosh2020rapid,

title = {Rapid Top-Down Synthesis of Large-Scale IoT Networks},

author = {Pradipta Ghosh and Jonathan Bunton and Dimitrios Pylorof and Marcos Vieira and Kevin Chan and Ramesh Govindan and Gaurav Sukhatme and Paulo Tabuada and Gunjan Verma},

url = {https://arxiv.org/abs/2002.04244},

year  = {2020},

date = {2020-01-01},

booktitle = {Proceedings of the IEEE International Conference on Computer Communications and Networks (ICCCN)},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{10.1145/3386367.3431292,

title = {Meeting SLOs in Cross-Platform NFV},

author = {Jane Yen and Jianfeng Wang and Sucha Supittayapornpong and Marcos A M Vieira and Ramesh Govindan and Barath Raghavan},

url = {https://doi.org/10.1145/3386367.3431292, Paper

https://github.com/USC-NSL/Lemur, Code

https://nsl.usc.edu/wp-content/uploads/2020/11/CoNEXT20-Lemur-slides.pdf, Slides},

isbn = {9781450379489},

year  = {2020},

date = {2020-01-01},

booktitle = {Proceedings of the 16th International Conference on Emerging Networking EXperiments and Technologies},

pages = {509–523},

publisher = {Association for Computing Machinery},

address = {Barcelona, Spain},

series = {CoNEXT '20},

abstract = {Network Functions (NFs) perform on-path processing of network traffic. ISPs are deploying NF Virtualization (NFV) with software NFs run on commodity servers. ISPs aim to ensure that NF chains, directed acyclic graphs of NFs, do not violate Service Level Objectives (SLOs) promised by the ISP to its customers. To meet SLOs, NFV systems sometimes leverage on-path hardware (such as programmable switches and smart NICs) to accelerate NF execution.Lemur places and executes NF chains across heterogeneous hardware while meeting SLOs. Lemur's novel placement algorithm yields an SLO-satisfying NF placement while weighing many constraints: hardware memory and processing stages, server cores, link capacity, NF profiles, and NF chain interactions. Lemur's metacompiler automatically generates code and rules (in P4, Python, eBPF, C++, and OpenFlow) to stitch cross-platform NF chain execution while also optimizing resource usage. Our experiments show that Lemur is alone among competing strategies in meeting SLOs for canonical NF chains while maximizing marginal throughput (the traffic rate in excess of the service-level objective).},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{avic,

title = {AViC: A Cache for Adaptive Bitrate Video},

author = {Zahaib Akhtar and Yaguang Li and Ramesh Govindan and Emir Halepovic and Yan Liu and Shuai Hao and Subhabrata Sen},

url = {https://nsl.usc.edu/wp-content/uploads/2019/10/ATTVideoCache.pdf},

year  = {2019},

date = {2019-12-09},

booktitle = {15th ACM Conference on emerging Networking EXperiments and Technologies (CoNEXT) Orlando, Florida, U.S. December 9-12, 2019},

keywords = {},

pubstate = {published},

tppubtype = {inproceedings}

}