Select a calendar:
Filter November Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for November
-
CS Colloquium: Adam Wierman (Caltech): Algorithmic challenges for greening data centers
Tue, Nov 06, 2012 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Adam Weirman, Caltech
Talk Title: Algorithmic challenges for greening data centers
Series: CS Colloquium
Abstract: Given the significant energy consumption of data centers, improving their energy efficiency is an important social problem.
However, energy efficiency is necessary but not sufficient for sustainability, which demands reduced usage of energy from fossil fuels. In this talk, I will describe some recent work highlighting the algorithmic challenges associated with "greening" data centers. We will focus on two applications:(i) dynamic resizing within a data center; and (ii) geographical load balancing across an Internet-scale system. In both contexts I will present our new algorithms, which provide significantly improved performance guarantees when compared with the "standard" approaches using Receding Horizon Control.
Additionally, if time allows, I will briefly discuss the our recent progress toward the implementation and evaluation of these algorithms in HP data centers.
Biography: Adam Wierman is a Professor in the Department of Computing and Mathematical Sciences at the California Institute of Technology, where he is a member of the Rigorous Systems Research Group (RSRG). His research interests center around resource allocation and scheduling decisions in computer systems and services. He received the ACM SIGMETRICS Rising Star award in 2011, and has been co-recipient of best paper awards at ACM SIGMETRICS, IEEE INFOCOM, IFIP Performance, the IEEE Green Computing Conference, and ACM GREENMETRICS. He was named a Seibel Scholar, received an Okawa Foundation grant, and received an NSF CAREER grant. Additionally, he has received multiple teaching awards, including the Associated Students of the California Institute of Technology (ASCIT) Teaching Award.
Host: Minlan Yu
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
George A Bekey Keynote Lecture: Dr. Jon Kleinberg
Tue, Nov 13, 2012 @ 04:00 PM - 07:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Jon Kleinberg, Cornell University
Talk Title: Computational Perspectives on Social Phenomena in On-Line Networks
Series: CS Keynote Series
Abstract: With an increasing amount of social interaction taking place in the digital domain, and often in public on-line settings, we are accumulating enormous amounts of data about phenomena that were once essentially invisible to us: the collective behavior and social interactions of hundreds of millions of people, recorded at unprecedented levels of scale and resolution. Analyzing this data computationally offers new insights into the design of on-line applications, as well as a new perspective on fundamental questions in the social sciences. We will review some of the basic issues around these developments; these include the problem of designing information systems in the presence of complex social feedback effects, and the emergence of a growing research interface between computing and the social sciences, facilitated by the availability of large new datasets on human interaction.
Biography: Dr. Kleinberg's research focuses on issues at the interface of networks and information, with an emphasis on the social and information networks that underpin the Web and other on-line media. His work has been supported by an NSF Career Award, an ONR Young Investigator Award, a MacArthur Foundation Fellowship, a Packard Foundation Fellowship, a Sloan Foundation Fellowship, and grants from Google, Yahoo!, and the NSF. He is a member of the National Academy of Sciences, the National Academy of Engineering, and the American Academy of Arts and Sciences.
Host: David Kempe
More Info: http://www.cs.usc.edu/calendar/csevents.asp?date=11%2F13%2F2012
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
Event Link: http://www.cs.usc.edu/calendar/csevents.asp?date=11%2F13%2F2012
-
Sanjoy Dasgupta: Cluster trees, Near-neighbor Graphs, and Continuum Percolation
Thu, Nov 15, 2012 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Sanjoy Dasgupta, UC San Diego
Talk Title: Cluster trees, Near-neighbor Graphs, and Continuum Percolation
Series: CS Colloquium
Abstract: What information does the clustering of a finite data set reveal about the underlying distribution from which the data were sampled? This basic question has proved elusive even for the most widely-used clustering procedures. One natural criterion is to seek clusters that converge (as the data set grows) to regions of high density. When all possible density levels are considered, this is a hierarchical clustering problem where the sought limit is called the "cluster tree". We give a simple algorithm for estimating this tree that implicitly constructs a multiscale hierarchy of near-neighbor graphs on the data points. We show that the procedure is consistent, answering an open problem of Hartigan. We also obtain rates of convergence, using a percolation argument that gives insight into how near-neighbor graphs should be constructed.
Biography: Sanjoy Dasgupta is a Professor in the Department of Computer Science and Engineering at UC San Diego. He received his PhD from Berkeley in 2000, and spent two years at AT&T Research Labs before joining UCSD.
His area of research is algorithmic statistics, with a focus on unsupervised and minimally supervised learning. He is the author of a textbook, "Algorithms" (with Christos Papadimitriou and Umesh Vazirani), that appeared in 2006.
Host: Shaddin Dughmi
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Matt Dreyer (VMWare): Software Defined Data Center: From Best Effort to 99%
Thu, Nov 15, 2012 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Matt Dreyer , VMWare
Talk Title: Software Defined Data Center: From Best Effort to 99%
Series: CS Colloquium
Abstract: For over a decade enterprises have been deploying a predictable design pattern for datacenter infrastructure and the applications residing there. Unfortunately this approach has failed to deliver the scale and velocity required by a highly competitive and global economy. The software defined datacenter holds the promise of unlocking application scale and deployment velocity for the next generation of applications.
Biography: Matt Dreyer is a Group Product Line Manager at VMware with responsibilities for Cloud Consumption products. Prior to VMware Mr. Dreyer was a Product Line Manager at Cisco where he lead the definition and go to market for the ground breaking Cisco ASA 5580 and ASA 5585 data center security appliances. Mr. Dreyer also managed the transformation of the security management product portfolio including Cisco Security Manager and the next generation Cisco Prime Security Manager. Prior to Cisco, Mr. Dreyer held product management positions at Sonicwall and ServGate where he pioneered "Unified Threat Management" firewalls. Mr. Dreyer graduated in from Colorado State University in 1996 with a BSEE degree. Mr. Dreyer has presented security talks at a number of VMware customer and partner events on topics including network security, application security, and security management.
Host: Minlan Yu
Location: Social Sciences Building (SOS) - B37
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Lixia Zhang: Evolving Internet into the Future via Named Data Networking
Mon, Nov 26, 2012 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Lixia Zhang, UCLA
Talk Title: Evolving Internet into the Future via Named Data Networking
Series: CS Colloquium
Abstract: While the Internet has succeeded far beyond expectations, the success has also stretched its initial design assumptions. Since applications operate in terms of data and more end points become mobile, it becomes increasingly difficult and inefficient to satisfy IP's requirement of determining exactly where (at which IP address) to find desired data.
The Named Data Networking project aims to carry the Internet into the future through a conceptually simple yet transformational architecture shift, from today's focus on where -- addresses and hosts -- to what -- the data that users and applications care about. In this talk I will present the basic design of NDN and our progress over the last couple years.
Biography: I joined the faculty of UCLA Computer Science Department in 1995/1996. My research at UCLA started with the design of a global scale web caching system, Adaptive Web Caching (AWC) funded by DARPA (joint work with Van Jacobson and Sally Floyd) and the Internet Distance Map Service funded by NSF (joint work with Paul Francis and Sugih Jamin). A direct follow-up to AWC was GRAB, "Reliable and Robust Sensor Data Collection by Gradient Broadcast" funded by DARPA. In parallel, we also did a number of initial IPv6 development projects. Our group was among the first on the 6Bone and implemented the first IPv6 multicast routing protocol, as well as porting vat and sdr to IPv6.
Since 1998 much of our focus has been on the deployed global Internet infrastructure. My students and I are currently tackling resiliency and security issues in the global routing system and Domain Name System (DNS), and the system challenges in deploying cryptographic protections in global scale open systems such as the Internet. My group has developed several useful tools that are widely used by the Internet research and operational communities, among them are Internet Topology Collection, Link Rank, Cyclops, SecSpider, and the latest addition EyeP, an IPv4 address allocation and usage visualization tool.
I coined the phrase "middlebox" in 1999, referring to the new components that were not in the original IP architecture but popped up in many places (web proxies, firewalls, NAT boxes). Much to my own surprise, the word was quickly picked up by the community and it is now used everywhere. In 2008 IEEE Network dedicated a special issue on the "Implications and Control of Middleboxes in the Internet".
I consider myself fortunate to join Internet research early on. During my 8 years of graduate school at MIT, my adviser Dr. David Clark taught me how to think architecturally. My career goal is to help the Internet grow. I am currently leading a 12-campus research project on the development of a new Internet architecture called Named Data Networking (NDN).
Host: Minlan Yu
Location: GFS 106
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: George Porter: Towards Balanced, Data-intensive Scalable Computing
Tue, Nov 27, 2012 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: George Porter, UC San Diego
Talk Title: Towards Balanced, Data-intensive Scalable Computing
Series: CS Colloquium
Abstract: While many interesting systems are able to scale linearly with additional servers, per-server performance can lag behind per-server capacity by more than an order of magnitude. In this talk, we will present Themis, a runtime supporting highly-efficient data-intensive computing. As an initial challenge application for this runtime, we built TritonSort, a highly efficient, scalable sorting system. It is designed to process large datasets with very high throughput (and has been evaluated against as much as 100 TB of input data spread across 832 disks in 52 nodes at a rate of 0.938 TB/min). It is also the winner of the 100TB "Indy" and "Daytona"
JouleSort benchmarks. In this talk, we will give an overview of the hardware and software architecture necessary to drive this level of efficiency. We then discuss how we have subsequetly generalized our system to support Map/Reduce programming. We believe the work holds a number of lessons for balanced system design and for scale-out architectures in general. Bridging the gap between high scalability and high performance will enable either significantly cheaper systems that are able to do the same work, or provide the ability to address significantly larger problem sets with the same infrastructure.
Biography: George Porter is a Research Scientist in the Center for Networked Systems and a member of the Systems and Networking group at UC San Diego. He received his B.S. from the University of Texas at Austin, and his Ph.D. from the University of California, Berkeley.
Host: Minlan Yu
Location: SSL 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Josh Reich: Modular Programming for Software Defined Networks
Thu, Nov 29, 2012 @ 02:00 PM - 03:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Josh Reich, Princeton University
Talk Title: Modular Programming for Software Defined Networks
Series: CS Colloquium
Abstract: In Software Defined Networks (SDN), a network application comprising many disparate tasks must be converted to a single set of packet-processing rules on network switches. Such rules are specified in terms of low-level hardware-centric operations. Thus, without additional logic, the rules comprising multiple independently written components cannot be composed as these will likely conflict.
Moreover, such rules inherently encode the details of the particular network topology for which they are written. Today's SDN platforms neither automate the process of composing independently written software components, nor provide abstractions for decoupling irrelevant topological features from core software logic.
Consequently, network programmers are forced to write monolithic programs tied to particular topologies - an expensive, error-prone process that produces code that is neither portable, nor reusable.
I will discuss our work to resolve both of these problems. I will introduce novel techniques for synthesizing a set of independent software components into a single coherent network application by providing both series and parallel composition operators. The talk will then turn to decoupling core software component logic from irrelevant topology specifics by providing transformed views of the network to software components. I will demonstrate two such mechanisms: a lightweight labeling technique that enriches the network view with abstracted attributes and a heavier-weight technique capable of hiding topologic detail by presenting views comprising virtualized switches and links.
Biography: Dr. Joshua Reich is a postdoctoral researcher working with Professor Jen Rexford at the Department of Computer Science, Princeton University. He designs and then builds systems to utilize networked infrastructure more easily and efficiently - currently focusing on Software Defined Networks. His other research interests include cloud computing, mobile/wireless systems, and green networks, Joshua was selected as an NSF/CRA Computing Innovation Fellow in the 2011 Cohort. He received Best Student Demo at ACM MobiCom/MobiHoc 2007 and was an ACM SRC Finalist at SIGCOMM 2010. Joshua's dissertation work on scalable P2P virtual machine streaming led to the founding of a startup, Silver Lining Systems.
Joshua earned his BA (Magna Cum Laude, Mathematics, 2002), MS (Computer Science, 2004), and PhD (Computer Science, 2011) from Columbia University, where he was co-advised by Professors Vishal Misra and Dan Rubenstein. During his PhD studies, Joshua interned with Sandia National Labs (NM and CA), Microsoft Research (Bangalore and Redmond) and Technicolor (Paris).
Host: Minlan Yu
Location: SOS B37
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Alexander Sherstov (UCLA): Limits of Multiparty Communication
Thu, Nov 29, 2012 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Alexander Sherstov, UCLA
Talk Title: Limits of Multiparty Communication
Series: CS Colloquium
Abstract: Communication complexity theory studies the following question: how many bits of communication are required to compute a Boolean function f whose arguments are distributed among several parties, possibly with overlap? Apart from being a natural subject of study in its own right, communication complexity sheds light on various questions in the theory of computing that do not seem to involve communication in any way.
A function f of basic importance in the area is the so called disjointness function, which evaluates to true when its arguments are sets with empty intersection. The multiparty communication requirements of this function have been actively studied since the late 1980s, with only very partial results available. In this work, we essentially resolve the question in its entirety.
PAPER URL: http://eccc.hpi-web.de/report/2011/145
Biography: Alexander Sherstov completed his Ph.D. at the University of Texas at Austin, under the direction of Adam Klivans. After a two-year postdoc at Microsoft Research, Sherstov joined the Computer Science Department at UCLA last year as an assistant professor. He has broad research interests in theoretical computer science, including computational complexity, computational learning, and quantum computing.
Host: Shaddin Dughmi
Location: SSL 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
