Berkeley AMPLab Seminar

News :

Stay tuned for full Fall 2015 schedule!

AMPLab Seminar Schedule (Fall 2015)

Date	Speaker	Title	Abstract	Slides

September 3	Kimon Fountoulakis, UC Berkeley	Flexible block coordinate descent	[LINK]
September 17	Deepak Agarwal, LinkedIn	Scaling Machine Learning for Web Applications	[LINK]
September 24	Sailesh Krishnamurthy, Amazon	An Introduction to Amazon Aurora	[LINK]
October 8	Roxana Geambasu, Columbia University	Privacy in a Data-Driven World	[LINK]
October 15	Sunil Shah, Mesosphere	Powering the Internet with Apache Mesos	[LINK]
October 22	Marcus Peinado, Microsoft Research	Shielding applications from an untrusted cloud with Haven and VCCC	[LINK]
October 29	Jeff Mogul, Google	Under the covers: the networks of a cloud provider	[LINK]
November 5	Sham Kakade, University of Washington	Faster least squares and faster eigenvector computation: Improved algorithms for optimization and statistics in the big data regime.	[LINK]
November 12	Mehul Shah, Amazon	Amazon Redshift and the Case for Simpler Data Warehouses	[LINK]
November 19	Matthew Johnson, Harvard University	Analyzing Gaussian Hogwild Gibbs Sampling	[LINK]
November 26	No Seminar, Thanksgiving
December 3	Adam Weirman, Caltech	Data centers and energy: Did we get it backwards?	[LINK]
December 10	Tudor Marian, Google	Dataflow: A Unified Model for Batch and Streaming Data Processing	[LINK]

AMPLab Seminar Schedule (Spring 2015)

Date	Speaker	Title	Abstract	Slides

January 28	Eugene Wu, UC Berkeley and Columbia	Closing the loop on data analysis	[LINK]
February 4	Matei Zaharia, Databricks and MIT	Spark in the Real World	[LINK]
February 18	Michael Cafarella, University of Michigan	Tools for the Data-Intensive Engineer	[LINK]
February 19	Alex Smola, CMU	The Parameter Server - Concepts, Optimization, and Partitioning	[LINK]
March 4	Yucheng Low, Dato	Systems For Real World Machine Learning on One Machine (and a cluster too)	[LINK]
March 11	Ankur Goyal and Anders Papitto, MemSQL	MemSQL, a Distributed In-Memory SQL Database	[LINK]
March 18	John Wilkes, Google	Cluster management at Google	[LINK]
March 25	No seminar -- Spring Recess
April 8	Brendan Burns, Google	Enabling distributed system development with Kubernetes	[LINK]
April 15	Evan Jones, Twitter	Startup Engineering versus Big Company Engineering	[LINK]
April 23	Ippokratis Pandis, Cloudera	Eliminating unscalable communication in transaction processing, toward bionic databases	[LINK]
April 29	Daisy Wang, University of Florida	Archimedes: A Probabilistic Master Knowledge Base System	[LINK]

AMPLab Seminar Schedule (Fall 2014)

Date	Speaker	Title	Abstract	Slides
September 3	George Porter, UCSD	Going beyond scalability to build resource-efficient data-intensive applications	[LINK]
September 10	Chris Re, Stanford	Simple Algorithms for Sophisticated Analytics	[LINK]
September 17	Ivan Filho, Google	The Challenge of Fairly Comparing Cloud Providers and What We're Doing About It	[LINK]
September 24	Gary Kazantsev, Bloomberg	Teaching Machines to Read for Fun and Profit: Machine Learning at Bloomberg	[LINK]
October 1	Max Heimel, TU Berlin	A Database Engine Design for Heterogeneous Hardware Environments	[LINK]
October 8	Pat Helland, Salesforce.com	Keystone: Between a ROC and a SOFT Place	[LINK]
October 15	Wenting Zheng, UC Berkeley	Fast Databases with Fast Durability and Recovery Through Multicore Parallelism	[LINK]
October 22	Daniel Abadi, Yale	The power of determinism in database systems	[LINK]
October 29	Mark Callaghan, Facebook	Still doing it wrong ["Small data (OLTP) at scale"]	[LINK]
November 5	Adam Coates, Baidu	Scaling up AI	[LINK]
December 3	Ashish Gupta, Google	Mesa: Geo-Replicated, Near Real-Time, Scalable Data Warehousing	[LINK]
December 10	James Cowling, Dropbox	Highly-reliable geo-distributed storage at Dropbox	[LINK]

AMPLab Seminar Schedule (Spring 2014)

Date	Speaker	Title	Abstract	Slides
January 15	Derek Murray, Microsoft Research SVC	Naiad: a timely dataflow system	[LINK]
January 22	Andy Pavlo, CMU	Cache Rules Everything Around Me	[LINK]
January 29	Christoph Koch, EPFL	Abstraction without regret in systems building: The case of databases	[LINK]
February 5	John Cieslewicz, Google	Google's F1 Distributed Relational Database	[LINK]
February 12	Diego Ongaro, Stanford University	The Raft Consensus Algorithm	[LINK]
February 19	Hakim Weatherspoon, Cornell University	From the Cloud to SoNIC: Precise Realtime Software Access and Control of Wired Networks	[LINK]
February 26	Nathan Bronson, Facebook	TAO: Facebook's Online Graph Data Store	[LINK]
March 5	Sam McVeety and Reuven Lax, Google	MillWheel: Fault-Tolerant Stream Processing at Internet Scale	[LINK]
March 26	spring recess
April 2	Shel Finkelstein & Anil Goel, SAP HANA	SAP HANA: Delivering A Data Platform for Enterprise Applications on Modern Hardware	[LINK]
April 9	Ben Zhao, UC Santa Barbara	Fun with Big Graphs	[LINK]
April 16	Marvin Theimer	Data in Motion at Scale	[LINK]
April 30	Hal Varian, Google/UC Berkeley	What Machine Learning Can Learn from Econometrics and Vice Versa	[LINK]
May 7	Yuan Yu, Microsoft Research SVC	From DryadLINQ to Dandelion: Compilers and Runtimes for Scalable Heterogeneous Systems	[LINK]
May 20	Miguel Castro, MSR Cambridge	FARM: Fast Available Remote Memory	[LINK]

AMPLab Seminar Schedule (Fall 2013)

Date	Speaker	Title	Abstract	Slides
October 9	Marcos K. Aguilera	Transaction Chains: Achieving Serializability with Low Latency in Geo-Distributed Storage Systems	[LINK]
October 16	Vijay Raman, IBM Almaden Research Center	Query Processing in DB2 BLU	[LINK]
October 23	James Demmel, UC Berkeley	Communication Avoiding Algorithms for Linear Algebra and Beyond	[LINK]
October 30	Dan Olteanu, Oxford	Factorized Relational Databases	[LINK]
November 6	Donald Kossmann, ETHZ	Sharing Up and Down and Left and Right	[LINK]
November 13	Dahlia Malkhi, MSR SVC	Programming Transactional Data Services over a Shared Log	[LINK]
November 20	Jeffrey Dean	Scaling Deep Learning	[LINK]
December 4	Dhruba Borthakur, Facebook	Rocksdb: A high performance key-value store for SSD	[LINK]
December 11	Douglas B. Terry, MSR	Consistency-Based Service Level Agreements for Cloud Storage	[LINK]

AMPLab Seminar Schedule (Fall 2012)

Date	Speaker	Title	Abstract	Slides
September 19	Joseph Gonzalez, AMPLab	Algorithms and Systems for Graph-Parallel Machine Learning	[LINK]	[pptx]
September 26	George Varghese (Microsoft Research, on leave from UCSD)	Querying a Million Genomes in a few millseconds?	[LINK]	[pptx]
October 3	Joseph Gonzalez, UC Berkeley	PowerGraph: Distributed Graph-Parallel Computation on Natural Graphs	[LINK]	[pptx]
October 5	Changhoon Kim, Microsoft	Network Virtualization for Large Data Centers and Enterprises	[LINK]
October 10	Neil Conway, UC Berkeley	Logic and Lattices for Distributed Programming	[LINK]	[pptx]
October 17	Hiroshi Wada, NICTA	Automatic Undo for Cloud Management via AI Planning	[LINK]	[pdf]
October 17	Allen Clement, Max Planck Institute for Software Systems	Making Geo-Replicated Systems Fast as Possible, Consistent when Necessary	[LINK]
October 24	Pat Helland, Salesforce.com	Immutability Changes Everything!	[LINK]	[pptx]
October 31	Alan Fekete, University of Sydney	Weak Isolation: Theory and its Impact	[LINK]	[ppt]
November 7	Manpreet Singh, Google	Photon: Fault-tolerant and scalable joining of continuous data streams	[LINK]
November 14	Dhruba Borthakur, Facebook	A Survey of Petabyte Scale Databases and Storage Systems Deployed at Facebook	[LINK]	[pdf]
November 28	Manju Rajashekhar, Twitter.com	Caching at Twitter and moving towards a persistent, in-memory key value store	[LINK]	[pdf]
December 4	Mike Freedman, Princeton University	Performance Isolation and Fairness for Multi-Tenant Cloud Storage	[LINK]
December 5	Michael Walfish, University of Texas at Austin	Making proof-based verified computation almost practical	[LINK]	[pptx]
December 12	Ed Nightingale, Microsoft Research	Flat Datacenter Storage	[LINK]	[pptx]
December 19	Kapil Surlaker, LinkedIn	All Aboard the Databus. LinkedIn's Scalable Consistent Change Data Capture Platform	[LINK]

Schedule (Spring 2012)

Date	Speaker	Title	Abstract	Slides
January 23	Michael Isard, Microsoft Research (SVC)	Naiad: Iterative and Incremental Data-Parallelism using Differential Dataflow	[LINK]	[pptx]
January 27	Yatin Chawathe, Google	Systems challenges in global scale map rendering	[LINK]
January 30	Nathan Marz, Twitter.com	Storm: distributed and fault-tolerant realtime computation	[LINK]	[pdf]
February 6	Lavanya Ramakrishnan, LBNL	Science in the Clouds and Beyond	[LINK]	[pdf]
February 13	Emre Kiciman (MSR ISRC)	Learning about the world through social media	[LINK]	[pptx]
February 27	Christopher Olston, Google	Programming and Debugging Large-Scale Data Processing Workflows	[LINK]	[pdf]
March 5	Dahlia Malkhi, Microsoft Research (SVC)	CORFU: A Shared-Log Design for Network-Attached Flash (NSDI 12)	[LINK]	[ppsx]
March 12	Mohammad Alizadeh, Stanford University	HULL: A High-bandwidth Ultra-Low Latency Data Center Fabric Architecture (NSDI 12)	[LINK]	[pptx]
March 19	Hairong Kuang, Facebook	Running the largest HDFS cluster	[LINK]	[pdf]
March 26	Jeffrey Dean, Google	Achieving Rapid Response Times in Large Online Services	[LINK]	[pdf]
April 2	Meg Walraed-Sullivan, University of California, San Diego	Scalable Label Assignment in Data Center Networks	[LINK]	[pptx]
April 9	Wyatt Lloyd, Princeton University	Scalable Causal Consistency for Wide-Area Storage with COPS (SOSP 11)	[LINK]	[pptx]
April 30	Theophilus A. Benson (UW-Madison)	Demystifying and controlling the performance of data center networks	[LINK]	[pptx]

Schedule (Fall 2011)

Date	Speaker	Title	Venue	Slides
September 7	Joseph Gonzalez, CMU	GraphLab: A New Framework for Parallel Machine Learning	UAI'10	pptx
September 14	Internal AMPLab Seminar
September 21	Internal AMPLab Seminar
September 28	Internal AMPLab Seminar
October 5	Marcos K. Aguilera, MSR-SVC	Transactional storage for geo-replicated systems	SOSP'11	pptx
October 19	Nisha Talagala, Fusion-IO	Implications of Non Volatile Memory on Software Architectures		pdf
November 2	Sriram Rao, Yahoo! Research	I-files: Handling Intermediate Data In Parallel Dataflow Graphs (Sailfish)		pptx
November 9	Indrajit Roy, HP Labs	Complex and Continuous Analytics with Distributed Arrays
November 16	Scott Lowe, EMC
November 23	Reto Kramer, Amazon.com
November 30	Walfredo Cirne, Google Inc.	Managing Descheduling Risk in the Google Cloud		pdf

Schedule (Spring 2011)

Date	Speaker	Title	Venue	Slides
February 7	Ganesh Ananthanarayanan, UC Berkeley	Disk-locality Considered Irrelevant in Datacenter Computing	HotOS'11	pptx
February 14	Matei Zaharia, UC Berkeley	The Datacenter Needs an Operating System	HotCloud'11	pptx
February 28	Sam Madden, MIT	RelationalCloud	CIDR'11	pptx
March 7	Haryadi Gunawadi, UC Berkeley	FATE & DESTINI	NSDI'11	pptx
March 28	John Ousterhout, Stanford University	RAMCloud: Scalable High-Performance Storage Entirely in DRAM		pptx
April 4	Dhruba Borthakur, Facebook Inc.	The Architecture of HDFS and its usage at Facebook		pdf
April 11	Peter Alvaro, UC Berkeley	Bloom: disorderly programming. CALM analysis. (CIDR'11)	CIDR'11	pptx
April 18	Raluca Popa, MIT	CryptDB: Confidentiality for Database Applications with Encrypted Query Processing		pptx
April 25	Matt Tolton, Google Inc.	Dremel: Interactive Analysis of Web-Scale Datasets	VLDB'10	pptx
May 2	Yingyi Bu, UC Irvine	HaLoop: Efficient Iterative Data Processing on Large Clusters (VLDB'10)	VLDB'10	pdf
May 9	JJ Furman, James Larson, Adi Ofer, Google Inc.	Megastore: Providing Scalable, Highly Available Storage for Interactive Services (CIDR'11)	CIDR'11
May 12 (4pm)	Derek Murray, University of Cambridge	CIEL: a universal execution engine	NSDI'11

General Information

Time	Thursdays 12:00pm-01:00pm
Location	430, Soda Hall, Berkeley [Directions]
Mailing list	amp-cloud@lists [UCB internal]
Info for speakers	25-30 minute talks are ideal due to interactivity and Q & A. Audience consists of 30-50 students and faculty with backgrounds in databases, networking, and systems.

To give a talk in the seminar please contact Joseph Gonzales: jegonzal AT eecs DOT berkeley DOT edu

Design by Minimalistic Design