The ongoing massive growth of data presents enormous opportunities and challenges to government, business and science. Taking advantage of the data, making sense of it, and extracting value requires a broad set of tools, technologies and skills – and remains quite challenging. There is a constant churn of software and hardware technologies aimed at enabling ever-greater performance (Productivity) of systems, analytics and people. But are these ‘advances’ actually enabling analysts to be more productive?
At this fourth CLSAC workshop, we will discuss the requirements of large-scale data analytic workloads and productivity from the various perspectives of analysts, mathematicians, data scientists, developers, and computer architects. How do we characterize and measure productivity? What math ‘works’ at what scale and accuracy? What types of systems are optimal for which tasks? At what scale do they ‘break’? What are the advantages/disadvantages of various languages in use today? How do analysts work productively across multiple security or public/private domains? Are there existing/planned analytic frameworks that scale, are platform independent, and useful to the novice user as well as the expert?
The workshop’s goals are to bring together thought leaders across Government, Industry and Academia to address these questions.
At this fourth CLSAC workshop, we will discuss the requirements of large-scale data analytic workloads and productivity from the various perspectives of analysts, mathematicians, data scientists, developers, and computer architects. How do we characterize and measure productivity? What math ‘works’ at what scale and accuracy? What types of systems are optimal for which tasks? At what scale do they ‘break’? What are the advantages/disadvantages of various languages in use today? How do analysts work productively across multiple security or public/private domains? Are there existing/planned analytic frameworks that scale, are platform independent, and useful to the novice user as well as the expert?
The workshop’s goals are to bring together thought leaders across Government, Industry and Academia to address these questions.
Organizing Committee:
|
2015 Sponsors
|
Program

clsac-final_agenda.pdf | |
File Size: | 658 kb |
File Type: |
Tuesday, October 13
8:45 – 12:00 p.m.
|
Session 1: System Productivity - Hardware
|
Converging Hadoop and HPC
|
Mac Dougherty, Cognitive Electronics
|
Designing network to scale high-performance analytics
|
Craig Stunkel, IBM
|
From Cell Phones to Super Computers and the pivotal role of memory
|
Steve Pawlowski, Micron
|
1:00 – 2:00 p.m.
|
Keynote
|
Big Analytics, DBMSs and HPC (PDF)
|
Michael Stonebraker, MIT
|
2:00 – 5:15 p.m.
|
Session 2: System Productivity – Software Stack
|
Scalable Automated Linking Technology (SALT) (PDF)
|
Edin Maharemagic, LexisNexis
|
Buried Alive! Massive Graph Analytics in 20 Lines of Code or Less
|
Jesse Shaw, LexisNexis
|
The Other HPC: High-Productivity Computing in Federated Big Dave Environments
|
Bill Howe, University of Washington
|
Wednesday, October 14
8:30 – 12:00 p.m.
|
Session 3: Human Productivity - Programming Models
|
Polystore, Julia, and Productivity in the Big Data World (PDF)
|
Tim Mattson, Intel
|
Enabling Efficient Analysis
|
Eric Dull, Deloitte and Touche, LLP
|
Why Predictive Analysis is Slow, and How to Fix It (PDF)
|
Art Munson, Context Relevant
|
Thursday, October 15
8:30 – 9:30 a.m.
|
Keynote
|
The Moore/Sloan Data Science Environments: Advancing Data-Intensive Discovery (PDF)
|
Ed Lazowska, University of Washington
|
9:30 – 12:30 p.m.
|
Session 4: Human Productivity – Math and SW Methods
|
Enhancing Human Performance at the Airport Security Checkpoint: Human Factors Research at the Transportation Security Administration
|
Bonnie Kudrick, TSA and Ann Speed, SNL
|
Building Tools for Urban Data Science
|
Claudio Silva, NYU
|
Finding Needles in Airport Haystacks
|
Andy Wilson, SNL
|
1:30 – 4:45 p.m.
|
Session 5: Applications and Productive User Scenarios
|
Integrative Multi-scale Analysis in Biomedical Data Science: Tools, Methods and Challenges
|
Joel Saltz, Stony Brook
|
Working Session: Leading and Managing Data Science Teams – What's 'Production'?
|
Clayton Chandler, Credit Suisse
|
Exploratory Data Analysis at Scale
|
J.T. Halbert, Tetra Concepts
|