Nina Taft is a both a manager and a senior research scientist at Intel Research Berkeley. Nina is currently managing about 1/3 of Intel's Research Lab in Berkeley.
Her research is geared towards making the Internet a safer place, and thus she works on security solutions for
both network infrastructure and for mobile end-hosts (such as laptops). She is interested in improving security through the smart use of measurement and inference technologies.
Nina recently started doing research in technologies for disaster recovery, focusing on the design of disaster-mode functionality for consumer devices and communication infrastructures
to help citizens with media-rich post-disaster communications. Prior to joining Intel, Nina worked at Sprint Labs for 5 years. There, she worked on ISP traffic engineering problems such
as traffic matrix estimation, routing, backbone traffic characterization and capacity planning. Prior to Sprint, she worked at SRI International in Menlo Park, CA, and conducted research
on congestion control and QoS routing. Nina received her PhD from UC Berkeley.
My colleagues and I have been designing monitoring tools that run directly on endhosts (laptops and desktops) for a couple of years now, and we are currently designing our 3rd generation tool (in collaboration with the University of Paris). These tools collect data on network traffic patterns, application usage, idle times, OS processes, keyboard clicks, and so on. As the tools have evolved, they gradually collect more data, and do so less and less obtrusively. Our current tool has been carefully designed to have minimal impact on the performance of user’s laptop, while simultaneously meeting the privacy needs of individuals. (For more information, see our short paper in ACM CCR April 2010 [PDF]). If you are willing to participate in our upcoming data collection effort, please
The rich data sets we have gathered have fueled research on a multiplicity of topics, including, security, performance diagnosis, and energy-saving proxies. We are currently engaged in technology transfer to Intel IT of a botnet detector for laptops that is based on behavioral detection methods (i.e. it builds rich user profiles, and employs sophisticated data analysis). We have also carried out research on how to combine network data with OS process data in order to determine the cause of a performance problem when one occurs.
One of the key barriers to adoption of data mining technologies is privacy. A popular data mining technique, called Support Vector Machines (SVM), is proving to be useful for a range of applications including security, profile building of web users in data centers, and so on. We are working on privacy preserving mechanisms for SVM algorithms. Many applications that make use of an SVM capability analyze data from a large community of users. We try to make it very hard for users, and even for IT managers, to deduce the values of data belonging to other users.
In the last few years, we have witnessed the blossoming of new social networking applications that have proved to be very useful to victims during times of disaster such as earthquakes, floods and fires.
These applications allow victims to collectively log their status (via text messages, or photos, or annotated maps) and share this information with the larger public. However the underlying network support for these emerging situational awareness
applications needs a great deal of research. We are focusing on 3 key challenges in this area. The first focuses on how to provide some form of network transport inside the disaster area when the usual infrastructure is unavailable or intermittently
available (e.g., due to major power outages caused by the disaster). We are looking into how to employ Delay Tolerant Networks (DTNs) in this context. The second challenge involves dealing with congestion in these networks, and the third is focused
on supporting situational awareness applications in which we can verify the integrity of the data.
This is of paramount importance when decisions regarding life-threatening situations, are being made based upon data collected from ordinary citizens. For more information on our Disaster Recovery Communications project, look
2010
ASTUTE: Detecting a Different Class of Traffic Anomalies
F. Silveira, C. Diot, N. Taft and R. Govindan.
ACM SIGCOMM. August 2010. [
PDF ]
HostView: Annotating End-host Performance Measurements with User Feedback.
D. Joumblatt, R. Teixeira, J. Chandrashekar and N. Taft.
HotMetrics, ACM Sigmetrics Workshop June 2010. [
PDF ]
Detecting Traffic Anomalies Using an Equilibrium Property
F. Silveira, C. Diot, N. Taft and R. Govindan.
ACM Sigmetrics. Extended Abstract.
Best Poster Award. June 2010. [
PDF ]
A Disruption-Tolerant Architecture for Secure and Efficient Disaster Response Communications
K. Fall, G. Iannaccone, J. Kannan, F. Silveira, and N. Taft
ISCRAM (Information Systems for Crisis Response and Management) May 2010. [
PDF ]
Perspectives on Tracing End-Hosts: A Survey Summary
D. Joumblatt, R. Teixeira, J. Chandrashekar and N. Taft.
ACM CCR. April 2010. [
PDF ]
2009
Macroscope: End-Point Approach to Networked Application Dependency Discovery
L. Popa, B.-G. Chun, I. Stoica, J. Chandrashekar and N. Taft.
ACM Conference on emerging Networking Experiments and Technologies (CoNEXT). December 2009. [
PDF ]
ANTIDOTE: Understanding and Defending against the Poisoning of Anomaly Detectors
B. Rubinstein, B. Nelson, L. Huang, A. Joseph, S. Lau, S. Rao, N. Taft and D. Tyger.
ACM Internet Measurement Conference (IMC). October 2009. [
PDF]
Exploiting Temporal Persistence to Detect Covert Botnet Channels
F. Giroire, J. Chandrashekar, N. Taft, E. Schooler and K. Papagiannaki.
Recent Advances in Intrusion Detection (RAID). September 2009. [
Electronic Paper Edition]
Impact of IT Monoculture on Behavioral End Host Intrusion Detection
D. Barman, J. Chandrashekar, N. Taft, M. Faloutsos, L. Huang and F. Giroire.
ACM SIGCOMM Workshop on Research on Enterprise Networks (WREN). August 2009. [
PDF]
Stealthy Poisoning Attacks on PCA-based Anomaly Detectors
B. Rubinstein, B. Nelson, L. Huang, A. Joseph, S. Lau, S. Rao, N. Taft and D. Tyger.
ACM Sigmetrics. Extended Abstract. June 2009.[
PDF]
Skilled in the Art of Being Idle: Reducing Energy Wasted in Networked Systems
S. Nedevschi, J. Chandrashekar, J. Liu, B. Nordman, S. Ratnasamy, and N. Taft.
Networked Systems Design and Implementation (NSDI). April 2009. [
PDF]
2008
Spectral Clustering with Perturbed Data
L. Huang, D. Yan, M. Jordan, and N. Taft.
Neural Information Processing Systems (NIPS). December, 2008. [
PDF]
How Healthy are Today's Enterprise Networks?
S. Guha, J. Chandrashekar, N. Taft and D. Papagiannaki.
Internet Measurement Conference (IMC), Greece. October, 2008.
[
PDF]
Evading Anomaly Detection Through Variance Injection Attacks on PCA
B. Rubinstein, B. Nelson, L. Huang, A. Joseph, S. Lau, N. Taft and D. Tyger.
Recent Advances in Intrusion Detection (RAID). (Extended Abstract). Best Poster Award. September, 2008.
[
PDF]
The Cubicle Vs. The Coffee Shop: Behavioral Modes in Enterprise End-Users
F. Giroire, J. Chandrashekar, G. Iannaccone, K. Papagiannaki, E. Schooler and N. Taft.
Passive and Active Measurement Workshop (PAM). April, 2008.
[
PDF]
Race Conditions in Coexisting Overlay Networks
R. Keralapura, C.-N. Chuah, N. Taft and G. Iannaccone.
IEEE/ACM Transactions on Networking. February, 2008.
[
PDF]
2007
Approximate Decision Making in Large-Scale Distributed Systems
L. Huang, A. Joseph, M. Garofalakis and N. Taft.
NIPS Workshop: Statistical Learning Techniques for Solving Systems
Problems (MLSys). (Extended Abstract). December, 2007. [
PDF]
IGP Link Weight Assignment for Operational Tier-1 Backbones
A. Nucci, S. Bhattacharyya, N. Taft and C. Diot.
IEEE/ACM Transactions on Networking. August, 2007. [
PDF]
Estimating Dynamic Traffic Matrices by using Viable Routing Changes
A. Soule, A. Nucci, R. Cruz, E. Leonardi and N. Taft.
IEEE/ACM Transactions on Networking. June, 2007 [
PDF]
Communication-Efficient Tracking of Distributed Cumulative Triggers
L. Huang, M. Garofalakis, A. Joseph and N. Taft.
International Conference on Distributed Computing Systems (ICDCS). Toronto, Canada. June, 2007.
Communication-Efficient Online Detection of Network-Wide Anomalies.
L. Huang, X. Nguyen, M. Garofalakis, J. Hellerstein, M. Jordan, M. Joseph, and N. Taft.
IEEE Infocom . Alaska. May, 2007. [
PDF]
Profiling the End Host
T. Karagiannis, K. Papagiannaki, N. Taft and M. Faloutsos.
Passive and Active Measurement Workshop (PAM) . Belgium. April, 2007. [
PDF]
Public Health for the Internet: Towards A New Grand Challenge for Information Management
J. Hellerstein, T. Condie, M. Garofalakis, B.T. Loo, P. Maniatis, T. Roscoe, and N. Taft.
Conference on Innovative Data Systems Research (CIDR). January, 2007.[
PDF]
2006
In-Network PCA and Anomaly Detection
L Huang, X. Nguyen, M. Garofalakis, M. Jordon, A. Joseph and N. Taft.
Neural Information Processing Systems (NIPS). December, 2006. [
PDF]
An Independent Connection Model for Traffic Matrices.
V, Erramilli, M. Crovella and N. Taft.
In ACM Internet Measurement Conference (IMC). Rio de Janeiro, Brazil. October, 2006. [
PDF]
Toward Sophisticated Detection With Distributed Triggers
L. Huang, M. Garofalakis, J. Hellerstein, A. Joseph, and N. Taft.
Workshop on Mining Network Data (MineNet),
co-located at ACM SIGCOMM. September, 2006. [
PDF]
A Fast Lightweight Approach to Origin-Destination IP Traffic Estimation Using Partial Measurements
G. Liang, N. Taft and B. Yu.
Special joint issue of IEEE Transaction on Networking and IEEE Transactions on Information Theory. June, 2006. [
PDF]
Sleeping Coordination for Comprehensive Sensing Using Isotonic Regression and Domatic Partitions
F. Koushanfar, N. Taft and M. Potkonjak.
IEEE Infocom. Barcelona, Spain. March, 2006. [
PDF]
2005
Traffic Matrix Tracking Using Kalman Filters
A. Soule, K. Salamatian, A. Nucci and N. Taft.
ACM SIGMETRICS Performance Evaluation Review (PER) Vol. 33, Issue 3.
Special issue on the First ACM Sigmetrics Workshop on Large Scale Networks Inference (LSNI). December, 2005. [
PDF]
Can Coexisting Overlays Inadvertently Step on Each Other?
R. Keralapura, C.N. Chuah, N. Taft and G. Iannaconne.
IEEE Proceedings of International Conference of Network Protocols (ICNP). November 2005. [
PDF]
Combining Filtering and Statistical Methods for Anomaly Detection
A. Soule, K. Salamatian and N. Taft.
In ACM/Sigcomm Internet Measurement Conference (IMC). Berkeley, CA. October, 2005. [
PDF]
Long-Term Forecasting of Internet Backbone Traffic
K. Papagiannaki, N. Taft, Z.-L. Zhang and C. Diot.
IEEE Transactions on Neural Networks, Vol. 15. No. 5. September, 2005.
[PDF]
An earlier conference version of this paper appeared in
Proceedings of IEEE Infocom. San Francisco. April, 2003.
[PDF] [PS]
The Problem of Synthetically Generating IP Traffic Matrices: Initial Recommendations
A. Nucci, A. Sridharan and N. Taft.
ACM Computer Communications Review (CCR). July, 2005.[
PDF]
Traffic Matrices: Balancing Measurements, Inference and Modeling
A. Soule, A. Lakhina, N. Taft, K. Papagiannaki, K. Salamatian, A. Nucci, M. Crovella and C. Diot.
ACM Sigmetrics. June, 2005. [
PDF]
Increasing Link Utilization in IP over WDM Networks Using Availability as QoS
A. Nucci, N. Taft, P. Thiran, H. Zhang and C. Diot.
Photonic Network Communications Journal (special issue on Mesh Restoration). Kluwer Academic Publishers. Vol 9.1, pp 55-75. January, 2005.
[PDF]
[PS]
2004
Controlled Use of Excess Backbone Bandwidth for Providing New Services in IP-over-WDM Networks
A. Nucci, N. Taft, C. Barakat, and P. Thiran.
IEEE Journal on Selected Areas in Communications (JSAC). November 2004 [
PDF]
Can ISP's Take the Heat from Overlay Networks?
R. Keralapura, N. Taft, C.N. Chuah, and G. Iannaconne.
ACM HotNets Workshop. November, 2004. San Diego. [
PDF]
Can ISPs and Overlay Networks form a Synergistic Co-existance?
R. Keralapura, N. Taft, G. Iannaconne, and C.-N. Chuah.
IFIP/IEEE Distributed Systems: Operations and Management (DSOM). (extended abstract) November 2004
[PDF]
A Distributed Approach to Measure IP Traffic Matrices
K. Papagiannaki, N. Taft and A. Lakhina.
ACM Internet Measurement Conference (IMC). October 2004. [
PDF]
Maximum Entropy Models: Convergence Rates and Applications in Dynamic System Monitoring
G. Liang, B. Yu and N. Taft.
IEEE International Symposium on Information Theory (ISIT). June 2004.
Extended abstract version [
PDF],
Short paper version [
PDF].
How to Identify and Estimate the Largest Traffic Matrix Elements in a Dynamic Environment
A. Soule, A. Nucci, E. Leonardi, R. Cruz, and N. Taft.
ACM Sigmetrics. June 2004.
[PDF]
Structural Analysis of Network Traffic Flows.
A. Lakhina, K. Papagiannaki, M. Crovella, C. Diot, E. Kolacyzk, and N. Taft. ;
ACM Sigmetrics. June 2004.
[PDF]
Flow Classification Using Histograms or How To Go on Safari on the Internet
A. Soule, K. Salamatian, N. Taft, R. Emilion and K. Papagiannaki.
ACM Sigmetrics. June 2004.
[PDF]
Design of IGP Link Weights for Estimation of Traffic Matrices
A. Nucci, R. Cruz, N. Taft and C. Diot.
Proceedings of IEEE Infocom. Hong Kong. March, 2004.
[PDF]
Impact of Flow Dynamics on Traffic Engineering Principles
K. Papagiannaki, N. Taft and C. Diot.
Proceedings of IEEE Infocom. Hong Kong. March, 2004.
[PDF]
2003
IGP Link Weight Assignment for Transient Link Failures
A. Nucci, B. Schroeder, S. Bhattacharyya, N. Taft and C. Diot.
International Teletraffic Congress (ITC). September 2003.
[PDF] [PS]
Network Availability Based Service Differentiations
M. Durvy, C. Diot, N. Taft and P. Thiran.
Proceedings of International Workshop on Quality-of-Service (IWQoS). June 2003.
[PDF]
An Approach to Alleviate Link Overload as Observed on an IP Backbone
S. Iyer, S. Bhattacharyya, N. Taft and C. Diot.
Proceedings of IEEE Infocom. April, 2003.
[PDF]
Increasing the Robustness of IP Backbones in the Absence of Optical Level Protection
F. Giroire, A. Nucci, N. Taft and C. Diot.
Proceedings of IEEE Infocom. April, 2003.
[PDF]
2002
Traffic Matrix Estimation: Existing Techniques and New Directions
A. Medina, N. Taft, K. Salamatian, S. Bhattacharyya, C. Diot.
ACM SIGCOMM. August 2002.
[PS]
A Taxonomy of IP Traffic Matrices
A. Medina, C. Fraleigh, N. Taft, S. Bhattacharyya and C. Diot.
SPIE ITCOM: Scalability and Traffic Control in IP Networks II. August 2002.
[PS]
Increasing Link Utilization without Penalty in IP over WDM Networks
A. Nucci, N. Taft, P. Thiran, H. Zhang, C. Diot.
SPIE Opticomm. August 2002.
[PDF]
A Pragmatic Definition of Elephants in Internet Backbone Traffic
K. Papagiannaki, N. Taft, S. Bhattacharyya, P. Thiran, K. Salamatian and C. Diot.
ACM Sigcomm Internet Measurement Workshop (IMW) (extended abstract). May 2002.
[PDF]
Geographical and Temporal Characteristics of Inter-POP Flows: View from a Single POP
S. Bhattacharyya, C. Diot, J. Jetcheva and N. Taft.
European Transactions on Telecommunications. February 2002.
[PS]
Tutorial
The Basics of BGP Routing and It's Performance in Today's Internet
Nina Taft.
Presented at RHDM (Reseaux Haut Debit et Multimedia) High Speed Multimedia Networks. French National Summer School. Corsica, France. May 2001.
[PPT]
Book Chapter
Neural Network Methods for Call Admission Control
Richard Ogier and Nina Taft-Plotkin.
Computational Intelligence in Telecommunications Networks. Chapter 2. Editor: Witold Pedrycz. CRC Press. 2001.
[PS]
Nina Taft