- talks - lists.cs.princeton.edu

Annie Edmundson will present her General Exam on January 21st, 2015 at 2pm in 302
by Nicki Gotsis 15 Jan '15

15 Jan '15

Annie Edmundson will present her General Exam on January 21st, 2015 at 2pm in 302. The members of her committee are Ed Felten, Arvind Narayanan and Cynthia Sturton (UNC-Chapel Hill). Everyone is invited to attend her talk, and those faculty wishing to remain for the oral exam following are welcome to do so. Her abstract and reading list are attached. Abstract: Malicious hardware provides an attacker a greater advantage than software-based attacks because it gives an attacker control over the entire software stack while providing a smaller chance of detection and prevention. The supply chain process for devices is extremely hard to secure in order to provide trusted microelectronic components; current methods for malicious circuit detection at design time usually look for circuitry that is never or rarely used. These methods include dynamic analysis during verification testing and boolean function analysis on a gate-level netlist. This work shows how to defeat the state-of-the-art algorithms in design time hardware trojan detection. We have implemented a backdoor in the LEON3 processor, which implements the SPARC V8 ISA; the malicious backdoor will allow a user-level program to enter supervisor mode with a specific sequence of instructions. We will also discuss fundamental problems of trojan detection, as well as possible areas for improvement. Reading List: Textbooks: 1. Anderson, Ross. Security engineering. John Wiley & Sons, 2008. 2. Harris, David Money, and Sarah L. Harris. Digital design and computer architecture. Elsevier, 2013. Chapters 1-5, 6.5, 6.6, 7.5, 7.6 Papers: 1. Agrawal, Dakshi, Selcuk Baktir, Deniz Karakoyunlu, Pankaj Rohatgi, and Berk Sunar. "Trojan detection using IC fingerprinting." In Security and Privacy, 2007. SP'07. IEEE Symposium on, pp. 296-310. IEEE, 2007. 2. Bilzor, Michael, Ted Huffmire, Cynthia Irvine, and Tim Levin. "Security checkers: Detecting processor malicious inclusions at runtime." In Hardware-Oriented Security and Trust (HOST), 2011 IEEE International Symposium on, pp. 34-39. IEEE, 2011. 3. Hicks, Matthew, Murph Finnicum, Samuel T. King, Milo Martin, and Jonathan M. Smith. "Overcoming an untrusted computing base: Detecting and removing malicious hardware automatically." In Security and Privacy (SP), 2010 IEEE Symposium on, pp. 159-172. IEEE, 2010. 4. King, Samuel T., Joseph Tucek, Anthony Cozzie, Chris Grier, Weihang Jiang, and Yuanyuan Zhou. "Designing and Implementing Malicious Hardware." LEET 8 (2008): 1-8. 5. Sturton, Cynthia, Matthew Hicks, David Wagner, and Samuel T. King. "Defeating UCI: Building stealthy and malicious hardware." In Security and Privacy (SP), 2011 IEEE Symposium on, pp. 64-77. IEEE, 2011. 6. Waksman, Adam, and Simha Sethumadhavan. "Silencing hardware backdoors." In Security and Privacy (SP), 2011 IEEE Symposium on, pp. 49-63. IEEE, 2011. 7. Waksman, Adam, and Simha Sethumadhavan. "Tamper evident microprocessors." In Security and Privacy (SP), 2010 IEEE Symposium on, pp. 173-188. IEEE, 2010. 8. Waksman, Adam, Matthew Suozzo, and Simha Sethumadhavan. "FANCI: identification of stealthy malicious logic using boolean functional analysis." In Proceedings of the 2013 ACM SIGSAC conference on Computer & communications security, pp. 697-708. ACM, 2013. 9. Zhang, Jie, Feng Yuan, Lingxiao Wei, Zelong Sun, and Qiang Xu. "VeriTrust: verification for hardware trust." In Proceedings of the 50th Annual Design Automation Conference, p. 61. ACM, 2013. 10. Zhang, Jie, Feng Yuan, and Qiang Xu. "DeTrust: Defeating hardware trust verification with stealthy implicitly-triggered hardware Trojans." In Proceedings of the 2014 ACM SIGSAC Conference on Computer and Communications Security, pp. 153-166. ACM, 2014.

1 0

Colloquium Speaker Adam Wierman, Today, 11am
by Nicole E. Wagenblast 15 Jan '15

15 Jan '15

Colloquium Speaker Adam Wierman, CalTech Data Centers, Energy, and Online Optimization Thursday, January 15, 11:00am Computer Science 105 This talk will tell two stories, one about designing sustainable data centers and one about the underlying algorithmic challenges, which fall into the context of online convex optimization. Story 1: The typical message surrounding data centers and energy is an extremely negative one: Data centers are energy hogs. This message is pervasive in both the popular press and academia, and it certainly rings true. However, the view of data centers as energy hogs is too simplistic. One goal of this talk is to highlight that, yes, data centers use a lot of energy, but data centers can also be a huge benefit in terms of integrating renewable energy into the grid and thus play a crucial role in improving the sustainability of our energy landscape. In particular, I will highlight a powerful alternative view: data centers as demand response opportunities. Story 2: Typically in online convex optimization it is enough to exhibit an algorithm with low (sub-linear) regret, which implies that the algorithm can match the performance of the best static solution in retrospect. However, what if one additionally wants to maintain performance that is nearly as good as the dynamic optimal, i.e., a good competitive ratio? In this talk, I'll highlight that it is impossible for an online algorithm to simultaneously achieve these goals. Luckily though, in practical settings (like data centers), noisy predictions about the future are often available, and I will show that, under a general model of prediction noise, even very limited predictions about the future are enough to overcome the impossibility result. Adam Wierman is a Professor in the Department of Computing and Mathematical Sciences at the California Institute of Technology, where he is a founding member of the Rigorous Systems Research Group (RSRG) and maintains a popular blog called Rigor + Relevance. His research interests center around resource allocation and scheduling decisions in computer systems and services. He received the 2011 ACM SIGMETRICS Rising Star award, the 2014 IEEE Communications Society William R. Bennett Prize, and has been coauthor on papers that received of best paper awards at ACM SIGMETRICS, IEEE INFOCOM, IFIP Performance (twice), IEEE Green Computing Conference, IEEE Power & Energy Society General Meeting, and ACM GREENMETRICS.

1 0

Santiago Cuellar will present his General Exam on Wednesday, January 21st, 2015 in Room 402 at 2pm.
by Nicki Gotsis 14 Jan '15

14 Jan '15

Santiago Cuellar will present his General Exam on Wednesday, January 21st, 2015 in Room 402 at 2pm. The members of his committee are Andrew W. Appel, David Walker, and Aarti Gupta. Everyone is invited to attend his talk, and those faculty wishing to remain for the oral exam following are welcome to do so. His/her abstract and reading list follow below. ================================================ Abstract: The first verified separate compiler for C, Compositional CompCert, was recently introduced with an operational model supporting language-independent linking and machine-checked proofs of vertical and horizontal composition, strengthening Leroy’s original CompCert proofs. The key techniques for its verification are structured simulations: logical relations between runtime states that enforce safety invariants in the presence of compiler-managed memory regions (private) and addressable portions of the stack frames that might be leaked to foreign modules through external function calls (public). However, the invariants are too strict to allow merging of stack frame as needed, for example, in function inlining optimizations. Moreover, changing the simulations is challenging since they strike a delicate balance that satisfies relies and guarantees among the compiled modules. In this work we present a refinement of the structured simulations that admits a larger number of compiler transformations (e.g. Function inlining) while maintaining compatibility with the operational model and the semantics preservation proof of the full compiler. The relaxation of the simulation works together with a strengthening of the proof for the composition of compiler phases (transitivity). The development has been written and verified in the proof assistant Coq. ======================================================== Reading List: Text books: * A. W. Appel. Modern Compiler Implementation in ML. Cambridge University Press, 1998 * Part I Fundamentals of Compilation * 13. Garbage Collection * 15. Functional Programming Languages (?) * 17. Dataflow Analysis * 19. Static Single-Assignment Form B. Pierce. Types and Programming Languages. 2002. * Part I Untyped Systems * Part II Simple Types * Part IV Recursive Types * Part V Polymorphism * 22. Type Reconstruction * 23. Universal Types * 24. Existencial Types * 25. An ML Implementation of system F Papers * [A. W. Appel] VeriSmall: Veriﬁed Smallfoot Shape Analysis. In First International Conf. on Certiﬁed Programs and Proofs Dec. 2011 * *[L. Beringer, G. Stewart, and A. W. Appel] Verified compilation for shared-memory C. In Esop’ 14: The 23rd European Symposium on Programming, 2014. * *[Leroy 2009] X. Leroy. Formal verification of a realistic compiler. Communications of the ACM, 52(7):107–115, 2009. * *[James T. Perconti and Amal Ahmed] Verifying an Open Compiler Using Multi-Language Semantics. Programming Languages and Systems 2014 * [Reynolds 2002] J.C. Reynolds. Separation Logic: a Logic for Shared Mutable Data Structures. In LICS’02: Proceedings of the 17th Annual IEEE Symposium on Logic in Computer Science, pages 55 - 74, 2002. * [Lerner 2003] S. Lerner, T. Millstein, and C. Chambers. Automatically Proving the Correctness of Compiler Optimizations. In PLDI’03: Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation, 2003. * [Leroy and Blazy 2008] X. Leroy and S. Blazy. Formal verification of a C-like memory model and its uses for verifying program transformations. J. Automated Reasoning, 41(1):1–31, 2008. * [G. Morrisett and D. Walker] From System F to Typed Assembly Language

1 0

Colloquium Speaker Adam Wierman, Thursday, Jan 15, 11am
by Nicole E. Wagenblast 14 Jan '15

14 Jan '15

Colloquium Speaker Adam Wierman, CalTech Data Centers, Energy, and Online Optimization Thursday, January 15, 11:00am Computer Science 105 This talk will tell two stories, one about designing sustainable data centers and one about the underlying algorithmic challenges, which fall into the context of online convex optimization. Story 1: The typical message surrounding data centers and energy is an extremely negative one: Data centers are energy hogs. This message is pervasive in both the popular press and academia, and it certainly rings true. However, the view of data centers as energy hogs is too simplistic. One goal of this talk is to highlight that, yes, data centers use a lot of energy, but data centers can also be a huge benefit in terms of integrating renewable energy into the grid and thus play a crucial role in improving the sustainability of our energy landscape. In particular, I will highlight a powerful alternative view: data centers as demand response opportunities. Story 2: Typically in online convex optimization it is enough to exhibit an algorithm with low (sub-linear) regret, which implies that the algorithm can match the performance of the best static solution in retrospect. However, what if one additionally wants to maintain performance that is nearly as good as the dynamic optimal, i.e., a good competitive ratio? In this talk, I'll highlight that it is impossible for an online algorithm to simultaneously achieve these goals. Luckily though, in practical settings (like data centers), noisy predictions about the future are often available, and I will show that, under a general model of prediction noise, even very limited predictions about the future are enough to overcome the impossibility result. Adam Wierman is a Professor in the Department of Computing and Mathematical Sciences at the California Institute of Technology, where he is a founding member of the Rigorous Systems Research Group (RSRG) and maintains a popular blog called Rigor + Relevance. His research interests center around resource allocation and scheduling decisions in computer systems and services. He received the 2011 ACM SIGMETRICS Rising Star award, the 2014 IEEE Communications Society William R. Bennett Prize, and has been coauthor on papers that received of best paper awards at ACM SIGMETRICS, IEEE INFOCOM, IFIP Performance (twice), IEEE Green Computing Conference, IEEE Power & Energy Society General Meeting, and ACM GREENMETRICS.

1 0

1-3pm Thu Jan 15 poster session for COS 561 (grad networks) in the Friend Convocation Room
by Jennifer Rexford 14 Jan '15

14 Jan '15

The students in COS 561 (graduate computer networks) are having a poster session to present their final course projects. The poster session will be 1-3pm Thursday January 15 in the Friend Convocation Room. The 25 projects will be divided into two one-hour periods, each including about half of the posters. We'll have snacks and beverages. Everyone is welcome! The project titles and authors are listed below. -- Jen FIRST HOUR (1-2pm) Managing Fairness in Equilibrium Strategies for Routing Among Selfish Nodes in an Ad Hoc Network Andrew Grasso Global Internet Reachability: A Longitudinal Study Anne Edmundson, Elba Garza, Caroline Trippel Improving the Performance of Path Query in Software Defined Networks Mina Tahmasbi Arashloo Proxy Rotation with Bitcoin Micropayments Miles Carlsten, Harry Kalodner, Paul Ellenbogen TIP: An IP Protocol for Detecting BGP Hijacking Attacks Collin Stedman Ravana: Transparent Controller Fault Tolerance in Software-Defined Networks Haoyu Zhang Latency Equalized Routing for Interactive Networks Heemin Seog and Andrew Tran Imagining a Clock-Synchronized Internet Gregory Owen and Jonathan Frankle Statistics Query Support for CoVisor Jennifer Gossels Detecting BGP Man-in-the-Middle Attacks via Control Plane Anomaly Detection Cornellius K. Metto, Laura M. Roberts, Elena Sizikova Fast BGP Convergence with the Supercharged Router: Electrifying Traditional Routers with SDN Michael Alan Chang Compressing Cached Rules in Software-Defined Networks Yatin A. Manerkar, Adi Fuchs, Alexey Lavrov SECOND HOUR (2-3pm) Upgrading HTTPS in Mid-Air: An Empirical Study of Strict Transport Security and Key Pinning Michael Kranch Prediction Based Mechanism for efficient scheduling of Cellular Data usage Themistoklis Melissaris, Nayana Prasad Nagendra Towards Temporal Network Synthesis Ryan Beckett, Qinxiang Cao, Olivier Savary Bélanger Multi-Network Cellular Access Parishad Karimi SDN-Enabled IPv4 Multicast Protocol Design Robert Macdavid, Sergiy Popovych, Hansen Zhang A Visualization Environment for Network Information in Pyretic Nora Coler, Ruth Dannenfelser, and Nevin Li Inter-domain path diversity through virtual peering Shreyasee Mukherjee MITM Detection with P2P Networks Steven Englehardt, Steven Goldfeder, Maciej Halber, Peter Zimmerman Evaluating Path Queries in Software-Defined Networks with Different Forwarding Policies Violeta Ilieva Dynamic Service Chaining Tengyu Ma, Amy Tai, Kelvin Zou Customizable Adaptive Streaming over Cellular Network Yichen Chen, Yixin Sun Exploring Scalability in Hierarchical SDN Controllers Disney Y. Lam Towards remote music collaboration: forecasting mallet motion via networks Huiwen Chang, Zeyu Jin and Shuran Song

1 0

Elba Garza will present her Generals on Monday, January 19, 2015 at 8:30 am in Rm 401
by Nicki Gotsis 12 Jan '15

12 Jan '15

Elba Garza will present her Generals on Monday, January 19, 2015 at 8:30 am in Rm 401. The members of her committee are David August, Margaret Martonosi, Kelly Shaw (University of Richmond). Everyone is invited to attend her talk, and those faculty wishing to remain for the oral exam following are welcome to do so. His/her abstract and reading list follow below. Abstract The emerging popularity of heterogeneous architectures results in a necessary and complex selection of runtime hardware/software parameters. As these heterogeneous systems have grown to offload computation to graphics processing units (GPUs), various Application-specific integrated circuits (ASICs) and other compute units, there is a higher need to optimize these units’ software and hardware settings to reach better power, performance, and area goals. Exhaustive exploration techniques are impractical since the design spaces grow exponentially with the addition of new parameters. Design spaces can easily reach thousands or millions of configurations to simulate or test. Manual and inference-based information can be flawed due to individual knowledge and bias, which may lead to unsearched design areas. Using statistical learning techniques, we have been able to optimize design space exploration for parameters of heterogeneous architecture components such as GPUs and Accelerator ASICs. These techniques allow for efficient and tractable design space assessment and can output optimal or near-optimal parameter settings for given applications automatically. Reading List [Cas+10] C. Cascaval et al. “A taxonomy of accelerator architectures and their programming models”. In: IBM Journal of Research and Development 54.5 (Sept. 2010), 5:1–5:10. issn: 0018-8646. doi: 10. 1147/JRD.2010.2059721. [Con+14] Jason Cong et al. “Accelerator-Rich Architectures: Opportunities and Progresses”. In: Proceedings of the 51st Annual Design Automation Conference. DAC ’14. San Francisco, CA, USA: ACM, 2014, 180:1–180:6. isbn: 978-1-4503-2730-5. doi: 10 . 1145 / 2593069 . 2596667. url: http://doi.acm.org/10.1145/2593069.2596667. [Esm+11] Hadi Esmaeilzadeh et al. “Dark Silicon and the End of Multicore Scaling”. In: Proceedings of the 38th Annual International Symposium on Computer Architecture. ISCA ’11. San Jose, California, USA: ACM, 2011, pp. 365–376. isbn: 978-1-4503-0472-6. doi: 10. 1145/2000064.2000108. url: http://doi.acm.org/10.1145/ 2000064.2000108. [Lyo+12] Michael J. Lyons et al. “The Accelerator Store: A Shared Memory Framework for Accelerator-based Systems”. In: ACM Trans. Archit. Code Optim. 8.4 (Jan. 2012), 48:1–48:22. issn: 1544-3566. doi: 10. 1145/2086696.2086727. url: http://doi.acm.org/10.1145/ 2086696.2086727. [Sha+14] Yakun Sophia Shao et al. “Aladdin: A Pre-RTL, Power-performance Accelerator Simulator Enabling Large Design Space Exploration of Customized Architectures”. In: Proceeding of the 41st Annual International Symposium on Computer Architecuture. ISCA ’14. Minneapolis, Minnesota, USA: IEEE Press, 2014, pp. 97–108. isbn: 1lba Garza January 19, 2015 978-1-4799-4394-4. url: http://dl.acm.org/citation.cfm?id= 2665671.2665689. [Ham+10] Rehan Hameed et al. “Understanding Sources of Inefficiency in General-purpose Chips”. In: Proceedings of the 37th Annual International Symposium on Computer Architecture. ISCA ’10. SaintMalo, France: ACM, 2010, pp. 37–47. isbn: 978-1-4503-0053-7. doi: 10.1145/1815961.1815968. url: http://doi.acm.org/10.1145/ 1815961.1815968. [JSM12] Wenhao Jia, Kelly A. Shaw, and Margaret Martonosi. “Stargazer: Automated Regression-based GPU Design Space Exploration”. In: Proceedings of the 2012 IEEE International Symposium on Performance Analysis of Systems & Software. ISPASS ’12. Washington, DC, USA: IEEE Computer Society, 2012, pp. 2–13. isbn: 978-1- 4673-1143-4. doi: 10 . 1109 / ISPASS . 2012 . 6189201. url: http : //dx.doi.org/10.1109/ISPASS.2012.6189201. [JSM13] Wenhao Jia, Kelly A. Shaw, and Margaret Martonosi. “Starchart: Hardware and Software Optimization Using Recursive Partitioning Regression Trees”. In: Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques. PACT ’13. Edinburgh, Scotland, UK: IEEE Press, 2013, pp. 257– 268. isbn: 978-1-4799-1021-2. url: http://dl.acm.org/citation. cfm?id=2523721.2523757. [JVT06a] P. J. Joseph, K. Vaswani, and Matthew J. Thazhuthaveetil. “Construction and use of linear regression models for processor performance analysis”. In: High-Performance Computer Architecture, 2006. The Twelfth International Symposium on. Feb. 2006, pp. 99– 108. doi: 10.1109/HPCA.2006.1598116. [JVT06b] P. J. Joseph, Kapil Vaswani, and Matthew J. Thazhuthaveetil. “A Predictive Performance Model for Superscalar Processors”. In: Proceedings of the 39th Annual IEEE/ACM International Symposium on Microarchitecture. MICRO 39. Washington, DC, USA: IEEE Computer Society, 2006, pp. 161–170. isbn: 0-7695-2732-9. doi: 10.1109/MICRO.2006.6. url: http://dx.doi.org/10.1109/ MICRO.2006.6. [LB06] Benjamin C. Lee and David M. Brooks. “Accurate and Efficient Regression Modeling for Microarchitectural Performance and Power Prediction”. In: SIGPLAN Not. 41.11 (Oct. 2006), pp. 185–194. issn: 0362-1340. doi: 10 . 1145 / 1168918 . 1168881. url: http : //doi.acm.org/10.1145/1168918.1168881. 2lba Garza January 19, 2015 Book List [HP12] J.L. Hennessy and D.A. Patterson. Computer Architecture: A Quantitative Approach. Morgan Kaufmann. Morgan Kaufmann/Elsevier, 2012. isbn: 9780123838728. Nicki Gotsis Graduate Coordinator Room 310, Computer Science (609) 258-5387 ngotsis(a)cs.princeton.edu

1 0

Colloquium speaker: Ben Raphael, Today, 4:30pm
by Nicole E. Wagenblast 12 Jan '15

12 Jan '15

Colloquium Speaker Ben Raphael, Brown University Monday, January 12, 4:30pm Computer Science 105 Understanding the molecular alterations that cause cancer is one of the greatest challenges in biology and medicine. Recent advances in DNA sequencing technology now enable large-scale measurement of mutations in cancer genomes and provide an unprecedented opportunity to address this challenge. However, translating this information into deeper insight about the mutational process that drives cancer demands novel computational approaches. In this talk, I will describe algorithms that we have developed to address key problems in cancer genomics. First, I will discuss algorithms that deconvolve DNA sequencing data from a single tumor and determine subpopulations of cancer cells harboring different mutations. These algorithms exploit integrality constraints on copy number aberrations and phylogenetic tree constraints that relate subpopulations. Next, I will describe algorithms to identify combinations of mutations that perturb cellular signaling and regulatory networks. One algorithm employs a heat diffusion process to identify subnetworks of a genome-scale interaction network that are recurrently mutated across samples. A second algorithm finds combinations of mutations that optimize a measure of mutual exclusivity across samples. I will illustrate applications of these approaches to multiple cancer types in The Cancer Genome Atlas (TCGA), including a recent Pan-Cancer study of >3000 samples from 12 cancer types. Ben Raphael is an Associate Professor in the Department of Computer Science and Director of the Center for Computational Molecular Biology (CCMB) at Brown University. His research focuses on the design of algorithms for genome sequencing and interpretation. Recent interests include structural variation in human and cancer genomes, and network/pathway analysis of genetic variants. He received an S.B. in Mathematics from MIT, a Ph.D. in Mathematics from the University of California, San Diego (UCSD), and completed postdoctoral training in Bioinformatics and Computer Science at UCSD. He is the recipient of an NSF CAREER award, a Career Award from the Burroughs Wellcome Fund, and a Sloan Research Fellowship.

1 0

Colloquium speaker: Ben Raphael, Monday January 12, 4:30pm
by Nicole E. Wagenblast 11 Jan '15

11 Jan '15

Colloquium Speaker Ben Raphael, Brown University Monday, January 12, 4:30pm Computer Science 105 Understanding the molecular alterations that cause cancer is one of the greatest challenges in biology and medicine. Recent advances in DNA sequencing technology now enable large-scale measurement of mutations in cancer genomes and provide an unprecedented opportunity to address this challenge. However, translating this information into deeper insight about the mutational process that drives cancer demands novel computational approaches. In this talk, I will describe algorithms that we have developed to address key problems in cancer genomics. First, I will discuss algorithms that deconvolve DNA sequencing data from a single tumor and determine subpopulations of cancer cells harboring different mutations. These algorithms exploit integrality constraints on copy number aberrations and phylogenetic tree constraints that relate subpopulations. Next, I will describe algorithms to identify combinations of mutations that perturb cellular signaling and regulatory networks. One algorithm employs a heat diffusion process to identify subnetworks of a genome-scale interaction network that are recurrently mutated across samples. A second algorithm finds combinations of mutations that optimize a measure of mutual exclusivity across samples. I will illustrate applications of these approaches to multiple cancer types in The Cancer Genome Atlas (TCGA), including a recent Pan-Cancer study of >3000 samples from 12 cancer types. Ben Raphael is an Associate Professor in the Department of Computer Science and Director of the Center for Computational Molecular Biology (CCMB) at Brown University. His research focuses on the design of algorithms for genome sequencing and interpretation. Recent interests include structural variation in human and cancer genomes, and network/pathway analysis of genetic variants. He received an S.B. in Mathematics from MIT, a Ph.D. in Mathematics from the University of California, San Diego (UCSD), and completed postdoctoral training in Bioinformatics and Computer Science at UCSD. He is the recipient of an NSF CAREER award, a Career Award from the Burroughs Wellcome Fund, and a Sloan Research Fellowship.

1 0

11am-noon Thu Jan 15 talk on "Data Centers, Energy, and Online Optimization"
by Jennifer Rexford 09 Jan '15

09 Jan '15

Speaker: Adam Wierman, CalTech Title: Data Centers, Energy, and Online Optimization Date/time: 11am-noon Thursday January 15 Location: CS 105 (small auditorium) Abstract: This talk will tell two stories, one about designing sustainable data centers and one about the underlying algorithmic challenges, which fall into the context of online convex optimization. Story 1: The typical message surrounding data centers and energy is an extremely negative one: Data centers are energy hogs. This message is pervasive in both the popular press and academia, and it certainly rings true. However, the view of data centers as energy hogs is too simplistic. One goal of this talk is to highlight that, yes, data centers use a lot of energy, but data centers can also be a huge benefit in terms of integrating renewable energy into the grid and thus play a crucial role in improving the sustainability of our energy landscape. In particular, I will highlight a powerful alternative view: data centers as demand response opportunities. Story 2: Typically in online convex optimization it is enough to exhibit an algorithm with low (sub-linear) regret, which implies that the algorithm can match the performance of the best static solution in retrospect. However, what if one additionally wants to maintain performance that is nearly as good as the dynamic optimal, i.e., a good competitive ratio? In this talk, I'll highlight that it is impossible for an online algorithm to simultaneously achieve these goals. Luckily though, in practical settings (like data centers), noisy predictions about the future are often available, and I will show that, under a general model of prediction noise, even very limited predictions about the future are enough to overcome the impossibility result. Bio: Adam Wierman is a Professor in the Department of Computing and Mathematical Sciences at the California Institute of Technology, where he is a founding member of the Rigorous Systems Research Group (RSRG) and maintains a popular blog called Rigor + Relevance. His research interests center around resource allocation and scheduling decisions in computer systems and services. He received the 2011 ACM SIGMETRICS Rising Star award, the 2014 IEEE Communications Society William R. Bennett Prize, and has been coauthor on papers that received of best paper awards at ACM SIGMETRICS, IEEE INFOCOM, IFIP Performance (twice), IEEE Green Computing Conference, IEEE Power & Energy Society General Meeting, and ACM GREENMETRICS.

1 0

Colloquium speaker: Ben Raphael, Monday January 12, 4:30pm
by Nicole E. Wagenblast 06 Jan '15

06 Jan '15

Colloquium Speaker Ben Raphael, Brown University Monday, January 12, 4:30pm Computer Science 105 Understanding the molecular alterations that cause cancer is one of the greatest challenges in biology and medicine. Recent advances in DNA sequencing technology now enable large-scale measurement of mutations in cancer genomes and provide an unprecedented opportunity to address this challenge. However, translating this information into deeper insight about the mutational process that drives cancer demands novel computational approaches. In this talk, I will describe algorithms that we have developed to address key problems in cancer genomics. First, I will discuss algorithms that deconvolve DNA sequencing data from a single tumor and determine subpopulations of cancer cells harboring different mutations. These algorithms exploit integrality constraints on copy number aberrations and phylogenetic tree constraints that relate subpopulations. Next, I will describe algorithms to identify combinations of mutations that perturb cellular signaling and regulatory networks. One algorithm employs a heat diffusion process to identify subnetworks of a genome-scale interaction network that are recurrently mutated across samples. A second algorithm finds combinations of mutations that optimize a measure of mutual exclusivity across samples. I will illustrate applications of these approaches to multiple cancer types in The Cancer Genome Atlas (TCGA), including a recent Pan-Cancer study of >3000 samples from 12 cancer types. Ben Raphael is an Associate Professor in the Department of Computer Science and Director of the Center for Computational Molecular Biology (CCMB) at Brown University. His research focuses on the design of algorithms for genome sequencing and interpretation. Recent interests include structural variation in human and cancer genomes, and network/pathway analysis of genetic variants. He received an S.B. in Mathematics from MIT, a Ph.D. in Mathematics from the University of California, San Diego (UCSD), and completed postdoctoral training in Bioinformatics and Computer Science at UCSD. He is the recipient of an NSF CAREER award, a Career Award from the Burroughs Wellcome Fund, and a Sloan Research Fellowship.

1 0