Dr Diana Andreea Popescu

Biography

I completed my PhD in Computer Science working on data centre networking, network measurement and cluster scheduling, in the Systems Research Group at the University of Cambridge, Computer Laboratory, Department of Computer Science and Technology. During my PhD I was a Marie Curie Early-Stage Researcher working in the EU-funded METRICS project. During my postdoc in the Systems Research Group, I worked on network management for IoT devices, applied machine learning to IoT network traffic, and federated learning, in the context of the DADA project.

I studied for my undergraduate degree in Computer Science at the University Politehnica of Bucharest, Romania, where I graduated first of my class, and for my master degree in Computer Science at École Polytechnique Fédérale de Lausanne, Switzerland. I am an alumna of Churchill College. I served as postgraduate representative and media chair on the women@CL committee. I was awarded a Women Techmakers Scholarship by Google in 2017.

My PhD thesis can be found here. My Google Scholar profile can be found here.

Publications

• Network latency, application performance-aware cluster scheduling in data centres
  Diana Andreea Popescu (University of Cambridge), Andrew W. Moore (University of Cambridge)
  IEEE Network, Special Issue on New Network Architectures, Protocols and Algorithms for Time-Sensitive Applications
• Human-centred home network security
  Derek McAuley, Jiahong Chen, Tom Lodge (University of Nottingham), Richard Mortier (University of Cambridge), Stanislaw Piasecki (University of Nottingham), Diana Andreea Popescu (University of Cambridge), Lachlan Urquhart (University of Edinburgh)
  Book chapter, Privacy by Design for the Internet of Things: Building accountability and security
• Revisiting IoT Device Identification
  Roman Kolcun (University of Cambridge), Diana Andreea Popescu (University of Cambridge), Vadim Safronov (University of Cambridge), Poonam Yadav (University of York), Anna Maria Mandalari (Imperial College London), Richard Mortier (University of Cambridge), Hamed Haddadi (Imperial College London)
  IFIP Network Traffic Measurement and Analysis Conference (TMA conference), September 14-15, Virtual
• Measuring Network Conditions in Data Centres using the Precision Time Protocol
  Diana Andreea Popescu (University of Cambridge), Andrew W. Moore (University of Cambridge)
  IEEE Transactions on Network and Service Management, Journal paper
• The Case for Retraining of ML Models for IoT Device Identification at the Edge
  Roman Kolcun (Imperial College London), Diana Andreea Popescu (University of Cambridge), Vadim Safronov (University of Cambridge), Poonam Yadav (University of York), Anna Maria Mandalari (Imperial College London), Yiming Xie (Imperial College London), Richard Mortier (University of Cambridge), Hamed Haddadi (Imperial College London)
  
• CoLearn: Enabling Federated Learning in MUD compliant IoT Edge Networks
  Angelo Feraudo (University of Bologna), Poonam Yadav (University of York), Vadim Safronov (University of Cambridge), Diana Andreea Popescu (University of Cambridge), Richard Mortier (University of Cambridge), Shiqiang Wang (IBM Research), Paolo Bellavista (University of Bologna), Jon Crowcroft (University of Cambridge)
  The 3rd International Workshop on Edge Systems, Analytics and Networking (EdgeSys 2020), in conjuction with ACM EuroSys 2020, 27th April 2020, Heraklion, Crete, Greece
• Enabling Event-Triggered Data Plane Monitoring
  Jan Kučera (CESNET), Diana Andreea Popescu (University of Cambridge), Han Wang (Barefoot Networks), Andrew W. Moore (University of Cambridge), Jan Kořenek (Brno University of Technology), Gianni Antichi (Queen Mary University of London)
  ACM SIGCOMM Symposium on SDN Research (ACM SOSR 2020), 3 March 2020, San Jose, CA
• Latency-driven performance in data centres
  Diana Andreea Popescu
  Doctoral thesis, Technical Report UCAM-CL-TR-937, https://doi.org/10.17863/CAM.38843, June 2019
• No Delay: Latency-Driven, Application Performance-Aware, Cluster Scheduling
  Diana Andreea Popescu, Andrew W. Moore
  arXiv preprint:1903.07114, March 2019
• Sensing the IoT Network: Ethical Capture of Domestic IoT Network Traffic
  Diana Andreea Popescu, Vadim Safronov, Poonam Yadav (University of Cambridge), Roman Kolcun, Anna Maria Mandalari, Hamed Haddadi (Imperial College London), Derek McAuley (University of Nottingham), Richard Mortier (University of Cambridge)
  17th ACM Conference on Embedded Networked Sensor Systems (SenSys 2019), November 10-13 2019, New York, USA
• A First Look At Data Center Network Conditions Through The Eyes Of PTPmesh
  Diana Andreea Popescu, Andrew W. Moore
  The 2nd International Network Traffic Measurement and Analysis (TMA 2018), 26-29 June 2018, Vienna, Austria
  Dataset and software available here.
• Seek and Push: Detecting Large Traffic Aggregates in the Dataplane
  Jan Kučera, Diana Andreea Popescu, Gianni Antichi, Jan Kořenek, Andrew W. Moore
  arXiv preprint, May 2018
• Characterizing the impact of network latency on cloud-based applications’ performance
  Diana Andreea Popescu, Noa Zilberman, Andrew W. Moore
  Technical Report, Number 914, UCAM-CL-TR-914, ISSN 1476-2986, November 2017, Computer Laboratory, University of Cambridge, UK
• PTPmesh: Data Center Network Latency Measurements Using PTP
  Diana Andreea Popescu, Andrew W. Moore
  The 25th IEEE International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS 2017), Banff, AB, Canada, September 20-22, 2017
  Dataset and software available here.
• Enabling Fast Hierarchical Heavy Hitter Detection using Programmable Data Planes
  Diana Andreea Popescu, Gianni Antichi, Andrew W. Moore
  ACM Symposium on SDN Research (SOSR) 2017, 3-4 April 2017, Santa Clara, CA
• Where Has My Time Gone?
  Noa Zilberman, Matthew Grosvenor, Diana Andreea Popescu, Neelakandan Manihatty-Bojan, Gianni Antichi, Marcin Wojcik, Andrew W. Moore
  The 18th International Conference on Passive and Active Measurement (PAM 2017), Sydney, Australia, March 30-31, 2017
Dataset and software available here.
• Multivariate Polynomial Multiplication on GPU
  Diana Andreea Popescu (University of Cambridge), Rogelio Tomas Garcia (CERN)
  The 16th International Conference on Computational Science (ICCS 2016), 6-8 June 2016, San Diego, California
  Code available here.
• Reproducing Network Experiments in a Time-controlled Emulation Environment
  Diana Andreea Popescu, Andrew W. Moore
  The 8th International Workshop on Traffic Monitoring and Analysis (TMA 2016), 7-8 April 2016, Louvain-La-Neuve, Belgium
• Parallel Computing Methods For Particle Accelerator Design
  Diana Andreea Popescu (École Polytechnique Fédérale de Lausanne and CERN)
  Master Thesis August 2013
• An Application-Assisted Checkpoint-Restart Mechanism for Java Applications
  Diana Andreea Popescu, E.D. Tirsa, M.I. Andreica, V. Cristea
  IEEE 12th International Symposium on Parallel and Distributed Computing, 2013

Posters

• Performance-aware cluster scheduling
  Diana Andreea Popescu, Andrew W. Moore
  ACM EuroSys 2020, April 27-30 2020, Heraklion, Crete, Greece
• Latency-driven cluster scheduling
  Diana Andreea Popescu, Andrew W. Moore
  16th USENIX Symposium on Networked Systems Design and Implementation (NSDI), February 27th 2019, Boston, MA
• Characterizing Network Latency Impact on Cloud Applications' Performance
  Diana Andreea Popescu, Noa Zilberman, Andrew W. Moore
  ACM Internet Measurement Conference (IMC), November 1st 2017, London, UK
• Characterizing Network Latency Impact on Cloud Applications' Performance, Best Poster Award
  Diana Andreea Popescu, Noa Zilberman, Andrew W. Moore
  N2Women Workshop, ACM SIGCOMM, 20 August 2017, Los Angeles, CA
• Network Latency Impact on Cloud Applications Performance
  Diana Andreea Popescu, Noa Zilberman, Andrew W. Moore
  Google Networking Research Summit, 7-8 February 2017, Sunnyvale, CA
• Omniscient: Towards realizing near real-time data center network traffic maps
  Diana Andreea Popescu, Andrew W. Moore
  ACM CoNEXT Student Workshop 2015, 1 December 2015, Heidelberg, Germany

Projects

• Latency-driven performance in data centres (PhD thesis)
• Network measurements in the data plane using programmable switches
• Defence Against Threats in IoT Smart Homes

Invited Talks

• Performance in the age of cloud computing and edge computing, ACM-W UK, July 31 2020, Virtual Seminar (video)
• Defence Against Threats in IoT Smart Homes, Women@CL Talklets, Computer Laboratory, University of Cambridge, June 12, 2020, Cambridge, UK
• Latency-driven performance in data centres, Systems Research Group Seminar, Computer Laboratory, University of Cambridge, June 4, 2020, Cambridge, UK (video)
• Performance in the age of cloud computing, VMware Research, February 19, 2020, Palo Alto, CA
• Performance in the age of cloud computing and edge computing, Nokia Bell Labs, January 24, 2020, Cambridge, UK
• Latency-driven performance in data centres, ETH Zurich, June 19, 2019, Zurich, Switzerland
• Network Latency in Data Centres: Measurement, Impact and Mitigation, Next Generation Networks Workshop, July 6, 2018, Abingdon, Oxfordshire, UK
• A First Look At Data Center Network Conditions Through They Eyes Of PTPmesh, 2nd International Conference on Traffic Measurement and Analysis (TMA 2018), 26-29 June, 2018, Vienna, Austria
• Network Latency in Data Centres: Measurement, Impact and Mitigation, Churchill College, CHUTalks, June 13, 2018
• Network Latency in Data Centres: Measurement, Impact and Mitigation, Conference on Everything, Churchill College, April 28, 2018
• Network Latency in Data Centres: Measurement and Impact, Oxbridge Women in Computer Science Conference, University of Cambridge Computer Laboratory, March 15, 2018
• A Look at Network Latency in Data Centres and How It Affects Application Performance, Women@CL Talklet, University of Cambridge Computer Laboratory, November 24, 2017
• PTPmesh: Data Center Network Latency Measurements Using PTP, IEEE MASCOTS, Banff, Canada, September 20, 2017
• PTPmesh: Data Center Network Latency Measurements Using PTP, Systems Research Group, University of Cambridge Computer Laboratory, September 7, 2017
• Network Latency Within Data Centers, University of California, Berkeley, August 9, 2017
• Network Latency Within Data Centers, Google, Bandwidth Enforcer team, Sunnyvale, July 31, 2017
• Network Latency Within Data Centers, Facebook, Menlo Park, July 21, 2017
• Where Has My Time Gone?, The 18th International Conference on Passive and Active Measurement (PAM 2017), Sydney, Australia, March 31, 2017
• Where Has My Time Gone?, Exablaze, Sydney, Australia, March 29, 2017
• Where Has My Time Gone?, Systems Research Group, University of Cambridge Computer Laboratory, March 16, 2017
• Can we use PTP to detect network congestion/measure network latency?, Systems Research Group/NetOS meeting, University of Cambridge Computer Laboratory, 1st November 2016
• Providing Network Latency Guarantees in Data Centres, Cambridge Festival of Ideas, Cambridge, UK, 22nd October 2016
• Network Measurements for Data Centers, UK Academic Meeting on Systems and Networks, Coseners 7-8th July 2016
• Reproducing Network Experiments in a Time-controlled Emulation Environment, The 8th International Workshop on Traffic Monitoring and Analysis (TMA 2016), 7-8th April 2016, Louvain-La-Neuve, Belgium
• Omniscient: Towards realizing near real-time data center network traffic maps, ACM CoNEXT Student Workshop, Heidelberg, Germany, 1st December 2015
• Using parallel computing methods at CERN, Systems Research Group/NetOS meeting, University of Cambridge Computer Laboratory, 22nd July 2014

Service

• TPC member for TMA 2022, PAM 2022, AIChallengeIoT 2021, CoNEXT Artefact Evaluation Committee 2021, PAM 2021, ANRW 2021, TMA 2021, AIChallengeIoT 2020, ACM EdgeSys 2020, ACM SenSys-ML 2020
• Reviewer for IEEE/ACM Transactions on Networking, IEEE Open Journal of Communications Society, IEEE Communications Letters
• Session chair for Performance at PAM 2021

Work Experience

• Software Engineering Intern at Google, Sunnyvale, USA, June 2017 - August 2017
  I worked in the Network Infrastructure team on optimising Bandwidth Enforcer's traffic delegation system to improve performance when used by specific application architectures.
• Software Engineering Intern at Google, Mountain View, USA, June 2015 - September 2015
  I worked in the Platforms Networking team on improving a network monitoring system used in Google's data centers.
• Site Reliability Engineer Intern at Google, Zurich, Switzerland, August 2013 - January 2014
  I worked in the Social Storage SRE team on measuring the CPU usage of individual user queries in order to estimate the cost of a query.
• Master Thesis Intern at CERN, Geneva, Switzerland, February 2013 - July 2013
  I worked for my master thesis on researching parallelisation techniques for the library used in the design of the Final Focus System of the future accelerators, the International Linear Collider (ILC) and the Compact Linear Collider (CLIC).
• Intern at CERN openlab, Geneva, Switzerland, July 2010 - August 2010
  Performance monitoring of the software frameworks for the Large Hadron Collider experiments.

Supervisions

I supervised students for the following courses:

• Programming in C/C++, Michaelmas 2015
• Operating Systems, Lent 2017
• Computer Networking, Lent 2019

Contact

University of Cambridge Computer Laboratory
William Gates Building, 15 JJ Thomson Avenue
Cambridge CB3 0FD, UK
dap53 (at) cam.ac.uk
+44 (0)1223 763607
Linkedin