We will be holding our next AMPLab retreat at the Chaminade Resort in sunny Santa Cruz, CA. The official schedule starts first thing in the morning on Thursday May 17 and runs until 3pm on Friday May 18th. … Continue reading →
Title: Energy Debugging in Smartphones Speaker: Y. Charlie Hu (Purdue) Date & Time: Monday 30 April 2012, 1:00pm Location: 373 Soda Hall Abstract: Despite the incredible market penetration of smartphones and exponential growth of the app market, utility of smartphones … Continue reading →
Title: Demystifying and controlling the performance of data center networks Speaker: Theophilus A. Benson (UW-Madison) Date & Time: Monday 30 April 2012, 11:00am Location: 380 Soda Hall Abstract: Although there is tremendous interest in designing improved networks for data centers, very little is known … Continue reading →
The next installment of the Spark User Meetup will be a detailed discussion of Shark, the port of Apache Hive onto Spark that the AMP Lab is developing. Shark lets users run SQL queries at interactive speeds against existing Hive data … Continue reading →
When: Tuesday, 4/17, 11:30am Where: AmpLab conference room (465 Soda Hall) Who: Mirek Riedewald, Northeastern U. Title: Interactive Search Queries for Online Communities Abstract: We introduce the problem of interactive search queries for online communities (ISQ-problem). Users often try to … Continue reading →
Title: Scalable Causal Consistency for Wide-Area Storage with COPS (SOSP 11) Speaker: Wyatt Lloyd, Princeton University Date & Time: Monday 9 April 2012, 11:00am Location: 380 Soda Hall Abstract: Geo-replicated, distributed data stores that support complex online applications, such as social networks, must provide … Continue reading →
Archana Ganapathi and others from Splunk will share with us their architecture for semi-structured time series database. Most modern systems generate abundant and diverse log data. With dwindling storage costs, there are fewer reasons to summarize or discard data. However, the … Continue reading →
Title: Scalable Label Assignment in Data Center Networks Speaker: Meg Walraed-Sullivan, University of California, San Diego Date & Time: Monday 2 April 2012, 11:00am Location: 380 Soda Hall Abstract: Modern data centers can consist of hundreds of thousands of servers and millions of virtualized … Continue reading →
2nd Annual NorCal DB Day to be held at UC Berkeley on Monday April 16, 2012 This is a single-day, workshop-style event where participants from academia and industry in Northern California meet to present ideas and discuss their research and … Continue reading →
Title: Achieving Rapid Response Times in Large Online Services Speaker: Jeffrey Dean, Google Date & Time: Monday 26 March 2012, 11:00am Location: Wozniak Lounge (Soda Hall 430) Abstract: Today’s large-scale web services provide rapid responses to interactive requests by applying … Continue reading →
Fred Reiss, IBM When: Tuesday, 3/20, 11:30am; 465 Soda Hall Title: Practical Entity Resolution at Massive Scale At the Infrastructure for Intelligent Information Systems group at IBM Almaden, we build systems that address a particular class of real-world customer problems. … Continue reading →
Title: Hogwild for Machine Learning on Multicore Speaker: Benjamin Recht, University of Wisconsin Date & Time: Wednesday 14 March 2012, 11:00am Location: 380 Soda Hall Abstract: Stochastic Gradient Descent (SGD) is a popular optimization algorithm for solving data-driven machine learning problems such as classification, … Continue reading →
Title: Running the largest HDFS cluster Speaker: Hairong Kuang, Facebook Date & Time: Monday 19 March 2012, 11:00am Location: 380 Soda Hall Abstract: HDFS is a highly scalable fault-tolerant distributed file system designed for running on low-cost commodity hardware. HDFS … Continue reading →
Markus Weimer, Yahoo Time: 11:30am 3/6/2012 Location: AmpLab conference room (465H Soda Hall) Title: Declarative Systems for Machine Learning Functional programming models like MapReduce raise the level of abstraction. They allow the programmer to focus on implementing her algorithm in … Continue reading →
Title: HULL: A High-bandwidth Ultra-Low Latency Data Center Fabric Architecture (NSDI 12) Speaker: Mohammad Alizadeh, Stanford University Date & Time: Monday 12 March 2012, 11:00am Location: 380 Soda Hall Abstract: An increasing number of applications and platforms in the data … Continue reading →
Title: CORFU: A Shared-Log Design for Network-Attached Flash (NSDI 12) Speaker: Dahlia Malkhi, Microsoft Research (SVC) Date & Time: Monday 5 March 2012, 11:00am Location: 380 Soda Hall Abstract: CORFU* is a novel storage cluster design that pools a farm … Continue reading →
PBS: Probabilistically Bounded Staleness Peter Bailis, UC Berkeley Wednesday 2/21/12 11:30am, 465 Soda Hall Eventually consistent semantics provide almost no guarantees regarding the recency of data returned (unbounded staleness of versions). Despite these weak guarantees, many data store users opt for eventual consistency in practice–why? It’s … Continue reading →
Title: Programming and Debugging Large-Scale Data Processing Workflows Speaker: Christopher Olston, Google Date & Time: Monday 27 February 2012, 11:00am Location: 380 Soda Hall Abstract: This talk gives an overview of my former team`s work on large-scale data processing at Yahoo! Research. The talk … Continue reading →
This week’s DB Seminar will feature our own Tim Kraska, talking about his work on transactional consistency in cloud and multi-data center environments. The talk will be held in 465 Soda.
The topic of this year’s Berkeley EECS Annual Research Symposium is “Big Data at Berkeley”. AMPLab will be playing a large role in this event, including featured talks by Mike Jordan and Michael Armbrust. We will also be hosting a lunch … Continue reading →
Title: Learning about the world through social media Speaker: Emre Kiciman Date & Time: Monday 13 February 2012, 11:00am Location: 380 Soda Hall Abstract: Social media, namely Twitter and Facebook, provides a finer-grained, larger-scale digital record of human interactions than … Continue reading →
Database Seminar – 465 Soda Hall Tuesday 2/7, Alexandra Meliou, University of Washington Why and How – A Reverse Perspective on Data Management Abstract: Current trends have seen data grow larger, more intertwined, and more diverse, as more and more users … Continue reading →
Michael Bernstein, MIT, “Crowd Powered Systems” 07 February 2012, 17:00 to 18:30 BCNM Commons (340 Moffitt) Crowd-powered systems combine computation with human intelligence, drawn from large groups of people connecting and coordinating online. These hybrid systems enable applications and experiences … Continue reading →
Title: Science in the Clouds and Beyond Speaker: Lavanya Ramakrishnan, LBNL Date & Time: Monday 6 February 2012, 11:00am Location: 380 Soda Hall Abstract: Cloud computing has served the needs of web applications for the last few years. Scientific applications are increasingly investigating cloud … Continue reading →
Internet Landscape Of China https://www.facebook.com/events/153148911467871/ 2pm – 3:30pm, Tuesday, Jan 31, 2012 Woz Lounge (430-438 Soda Hall) In this talk, Alex Cheng, Executive Director at Baidu, will give an overview of the China internet landscape and based on his experience, how Baidu … Continue reading →
Title: Science in the Clouds and Beyond Speaker: Lavanya Ramakrishnan, LBNL Date & Time: Monday 6 February 2012, 11:00am Location: 380 Soda Hall Abstract: Cloud computing has served the needs of web applications for the last few years. Scientific applications … Continue reading →
Title: Storm: distributed and fault-tolerant realtime computation Speaker: Nathan Marz, Twitter.com Date & Time: Monday 30 January 2012, 11:00am Location: 380 Soda Hall Abstract: Storm makes it easy to write and scale complex realtime computations on a cluster of computers, … Continue reading →
On January 31st, we’re hosting the first Spark User Meetup at Klout in San Francisco. Spark is a cluster programming framework that provides in-memory computing for iterative and interactive analytics and a high-level programming interface in the Scala language. This … Continue reading →
Title: Systems challenges in global scale map rendering Speaker: Yatin Chawathe, Google Date & Time: Friday 27 January 2012, 11:00am Location: 380 Soda Hall Abstract: I will discuss the evolution of the Google Maps rendering infrastructure from a simple static … Continue reading →
Title: Naiad: Iterative and Incremental Data-Parallelism using Differential Dataflow Speaker: Michael Isard, Microsoft Research (SVC) Date & Time: Monday 23 January 2012, 11:00am Location: 380 Soda Hall Abstract: Naiad is a new framework for data-parallel computing, … Continue reading →
When: Wednesday, Jan 18, 12-1pm Where: 373 Soda Hall Title: Online Testing of Distributed Systems Speaker: Dejan Kostic, EPFL Abstract: It is notoriously difficult to make distributed systems reliable. This becomes even harder in the case of the widely-deployed systems … Continue reading →
A three-day, intensive update and discussion of recent results, on-going work and future plans. For lab members, application partners, and sponsors only.
Today we’ll be doing dry runs of the talks for the upcoming AMPLab retreat. Feedback appreciated.
Title: Routing Along DAGs Speaker: Junda Liu Advisor: Scott Shenker Date: Thursday, December 15, 2011 Time: 9 – 10AM Room: 373 Soda Hall ABSTRACT: Routing is arguably the most fundamental aspect of networking, since it answers the basic question: how … Continue reading →
Speaker: Igor Ganichev Adviser: Scott Shenker Date: Monday, December 12, 2011 Time: 9 -10AM Room: 373 Soda Hall Abstract: While astonishingly successful, Internet is still less reliable than the phone system and supports very limited user choice and control. As … Continue reading →
What: CS294-42 Poster Session (Cloud Computing: Systems, Networking, and Frameworks) When: Wednesday, December 7, 9:30-11:30am Where: Wozniak Lounge List of posters: – Sparrow: Scheduling Interactive Jobs in Large Clusters – MegaPipe: A Clean-Slate Approach for Scalable Network I/O – Secure … Continue reading →
Title: MADDER and Self-Tuning Data Analytics on Hadoop with Starfish Speaker: Shivnath Babu, Duke University Date&Time: Friday, 9th of September, 12:30pm-1:30pm Location: Big RADLab meeting room Abstract: Timely and cost-effective analytics over “big data” is now a key ingredient for … Continue reading →
Title: Invited Lecture from Aaron T. Myers, Software Engineer, Cloudera Date & Time: Wednesday, 7th of September, 11-12pm Location: 293 Cory Hall
Title: Evolving the Internet with Declarative Networking Speaker: Boon Thau Loo, University of Pennsylvania Date & Time: Friday, 23th of September, 12:30pm-1:30pm (there will be lunch starting from 12:15) Location: Big RADLab meeting room (465H) Abstract: In this talk, I … Continue reading →
Title: Integrating Statistical Data-Analysis Tools into an RDBMS using Incremental Gradient Descent Speaker: Christopher RÈ, University of Wisconsin-Madison Date&Time: Friday, 7th of October, 12:30pm-1:30pm (there will be pizza from 12:15) Location: Soda Hall 310 (different room) Upcoming talks Abstract: The … Continue reading →
Michael Franklin will present an overview of the AMPLab at the South Bay Big Data Analytics Meet Up, at Adobe Headquarters in San Jose. Details and registration information are available at http://www.meetup.com/Big-Data-Analytics-Mobile-Social-and-Web/events/33585562/
Title: Realtime Apache Hadoop at Facebook Speaker: Dhruba Borthakur Date & Time: Wednesday, 5th of October, 11-12pm Location: 293 Cory Hall
Title: Introducing Druid: Real-Time Analytics at a Billion Rows per Second Speaker: Eric Tschetter, Metamarkets Date&Time: Friday, 14th of October, 12:30pm-1:30pm Location: Big RADLab meeting room (465H) Abstract: At Metamarkets, we have developed a distributed, in-memory OLAP store we call … Continue reading →
Title: PNUTS: Yahoo!’s hosted data serving platform Speaker: Raghu Ramakrishnan Date & Time: Wednesday, 10/17/11, 11am Abstract We describe PNUTS, a massively parallel and geographically distributed database system for Yahoo!’s web applications. PNUTS provides data storage organized as hashed or … Continue reading →
Title: Parallel Index and Query for Large Scale Data Analysis Speaker: John Wu, Lawrence Berkeley Lab Date&Time: Friday, October 21, 12:30pm-1:30pm (there will be lunch from 12:15) Location: Big RADLab meeting room (465H) Seminar page: http://db.cs.berkeley.edu/w/db-seminar/ Abstract: Modern scientific datasets … Continue reading →
Title: Implications of Non Volatile Memory on Software ArchitecturesSpeaker: Nisha Talagala (FusionIO)Date&Time: Wednesday, 19th of October, 3:00pm-4:00pmLocation: RadLab, 465H, Soda Hall Abstract: Flash based non volatile memory is revolutionizing data center architectures, improving application performance by bridging the gap … Continue reading →
Title: Scaling up database transactions Speaker: Anastasia Ailamaki, EPFL Date&Time: Friday, October 28, 12:30pm-1:30pm (there will be lunch from 12:15) Location: Big RADLab meeting room (465H) Seminar page Abstract: As the number of compute elements per chip increases exponentially, embarrassingly … Continue reading →
Title: Hadoop NextGen Speakers: Arun Murthy and Mahadev Konar, Hortonworks 1-2:30pm, Monday, October 31, 60 Evans Hall Abstract: The Apache Hadoop MapReduce framework has hit a scalability limit around 4,000 machines. We are developing the next generation of Apache Hadoop … Continue reading →
Title: I-files: Handling Intermediate Data In Parallel Dataflow Graphs Speaker: Sriram Rao (Yahoo! Research) Date & Time: Wednesday (tomorrow), 2nd of November, 3:00pm-4:00pm Location: 465H, RadLab big conference room, Soda Hall Over the past few years parallel dataflow graph frameworks … Continue reading →
Title: Query Languages & Systems for Continuous Analytics and Data Mining: the Stream Mill Experience Speaker: Carlo Zaniolo,UCLA Date&Time: Friday, November 04, 12:30pm-1:30pm (there will be lunch from 12:15) Location: Big RADLab meeting room (465H Soda Hall) Abstract: Database-inspired query … Continue reading →
Title: Scale Out Networking in the Data Center Speaker: Amin Vahdat (Google/UCSD) Date & Time: Monday, Nov 7, 1-2:30pm Location: 60 Evans Hall Abstract: Scale-out architectures supporting flexible, incremental growth in capacity are common for computing and storage. However, the … Continue reading →
Title: Complex and Continuous Analytics with Distributed Arrays Speaker: Indrajit Roy (HP labs) Date&Time: Wednesday, 9th of November, 3:00pm-4:00pm Location: 465H, RadLab big conference room, Soda Hall Abstract: Presto is a distributed programming model for continuously analyzing data. Continuous analytics, … Continue reading →
Title: Thoughts on building a private cloud Speaker: Jase McCarty, EMC Corporation Date&Time: Wednesday, 16th of November, 3:00pm-4:00pm Location: Soda Hall 380 Abstract: Real-world insight & experience on gathering requirements, designing the solution, and some pitfalls to avoid. This talk … Continue reading →
Title: Implications of Evolving Hardware Trends: Towards Energy-Conscious Main-Memory Data Processing Speaker: Jignesh Patel, University of Wisconsin-Madison Date&Time: Friday, November 18, 12:30pm-1:30pm Location: Big RADLab meeting room (465H Soda Hall) Abstract: The internals of data management systems have largely made … Continue reading →
Title: Making device access less peripheral Speaker: Michael Swift (U. of Wisconsin, Madison) Date & Time: 11/21/11, 2:30-3:30 Location: Wozniak Lounge Abstract: Supporting peripheral devices, once a motivation for creating operating systems, has received little attention from the research community. … Continue reading →
Microsoft Research Silicon Valley Distinguished Speaker Series Title: AMPLab: Making Sense of Big Data with Algorithms, Machines and People Speaker: Michael Franklin, UC Berkeley When: November 17th, 3-4p, informal reception from 2:30-3p Where: Microsoft SVC 6 – Titan Conference Room … Continue reading →
Title: ASTERIX: Towards a Scalable, Semistructured Data Platform for Evolving World Models Speaker: Michael Carey, UC Irvine Date&Time: Friday, December 2, 12:30pm-1:30pm (there will be lunch from 12:15) Location: Big RADLab meeting room (465H Soda Hall) Abstract: ASTERIX is a … Continue reading →
Title: Managing Descheduling Risk in the Google Cloud Speaker: Walfredo Cirne, Google Inc. Date&Time: Wednesday (tomorrow), 30th of November, 3:00pm-4:00pm Location: 465H, RadLab big conference room, Soda HallAbstract: Increasing utilization is goal we all strive for. However, too … Continue reading →
What: Cluster management at Google Who: John Wilkes When: Monday, November 28, 2011, 1-2:30pm Where: 60 Evans Hall Abstract: Cluster management is the term that Google uses to describe how we control the computing infrastructure in our datacenters that supports almost all … Continue reading →
Title: Implications of Evolving Hardware Trends: Towards Energy-Conscious Main-Memory Data Processing Speaker: Jignesh Patel, University of Wisconsin-Madison Date&Time: Friday, November 18, 12:30pm-1:30pm Location: Big RADLab meeting room (465H Soda Hall) Abstract: The internals of data management systems have largely made … Continue reading →
Microsoft Research Silicon Valley Distinguished Speaker Series Title: AMPLab: Making Sense of Big Data with Algorithms, Machines and People Speaker: Michael Franklin, UC Berkeley When: November 17th, 3-4p, informal reception from 2:30-3p Where: Microsoft SVC 6 – Titan Conference Room … Continue reading →