Immersive Teaching and Research in Data Sciences via Cloud Computing Cloud Era Ltd 13 June 2013 Karim Chine

22 Outline Introduction The Rise of Data Science What is missing on the Cloud for Research and Education? Elastic-R: The Next Generation Data Science Platform Elastic-R: Design and Technologies Overview Rethinking Virtual Research and Teaching Demo Conclusion

33 Introduction Science and the 4th Paradigm The promises of Grid Computing for e-Research The e-Science program (UK) The NSF-funded cyber infrastructure (USA) e-Infrastructure and ICT Calls (EC) Experimental Science Theoretical Science Computational Science e-Science / Data-intensive Science

4 Introduction The Dancing Bear The townspeople gather to see the wondrous sight as the massive, lumbering beast shambles and shuffles from paw to paw. The bear is really a terrible dancer, and the wonder isn't that the bear dances well but that the bear dances at all The tragic failure of the Grid paradigm No democratic access for all scientists, a tool for Particle Physicists and Geeky scientists No sustainable infrastructures No SLA, best effort support No flexibility, no isolation, only admins can install software, X.509 certificates, restrictive policies No interaction design, no user-centric approach

55 Introduction The Data Deluge and the challenges of Big Data *

6 The Rise of Data Science * Data Scientists Led by NASA Star Most Sought for Century: Jobs By Aki Ito - Jun 11, :01 AM ET (Bloomberg.com) Harvard Business Review last year called this profession “the sexiest job of the 21st century” One measure of demand: Hours billed for work in statistical analysis grew by 522 percent in the first quarter compared with the same period in 2011

77 What is missing on the Cloud for Research and Education? Science-as-a-Service: (SCEs) Scientific Computing Environments-as-a-service, Models-as-a-service… Generalized Real-time Collaboration, including interaction with SCEs and with Data One consistent platform for homogenous access to Science Clouds and new abstractions for stateful access to remote infrastructure/Data Reproducibility and traceability of data analysis and computational research Science Gateways building and publishing for everyone

8 Elastic-R: The Next Generation Data Science Platform Arduino / Raspberry pi Democratizing Electronics Elastic-R Democratizing Data Science

99 Elastic-R: The Next Generation Data Science Platform Elastic-R is:  A link between the data scientist/the educator/the student and the Infrastructure-as-a-Service  A consistent set of concepts, architectures, frameworks and interaction design models to wrap and complete the existing public and private clouds with the missing capabilities to radically improve the data scientist productivity  « Super Glue » between the virtual IT capabilities, the software capabilities, the mathematical/statistical capabilities and the man- machine interaction components  A Data Operating System enabling infinite combinations for analysing data and building rapidly Data Science-centric applications and services

10 Elastic-R: The Next Generation Data Science Platform Elastic-R is:  A hyper-dropbox allowing to run a large spectrum of capabilities next to the centrally stored and ubiquitously accessible data.  A platform using the cloud to introduce traceability and reproducibility to the data science arena.  A federation paradigm for scientific clouds  A framework for building on-the-fly user-defined software-as-a-service

11 Elastic-R: Design and Technologies Overview

12 Elastic-R: Design and Technologies Overview Robot submarine dives to the deepest part of the ocean controlled by a 7-mile cable as thin as single human hair

13 Elastic-R: Design and Technologies Overview Remote Java/R Processes  Events-driven Remote Objects/Engines  R, Python, Mathematica, Matlab, Scilab,..  Collaborative Spreadsheets  Collaborative Scientific Graphics Canvas  Collaborative Dashboard with collaborative widgets

14 Elastic-R: Design and Technologies Overview

15 Elastic-R: Design and Technologies Overview Elastic-R AMI 1 R 2.10 BioC 2.5 Elastic-R AMI 2 R 2.9 BioC 2.3 Elastic-R AMI 3 R 2.8 BioC 2.0 Elastic-R Amazon Machine Images Elastic-R EBS 1 Data Set XXX Elastic-R EBS 2 Data Set YYY Elastic-R EBS 3 Data Set ZZZ Elastic-R EBS 4 Data Set VVV Elastic-R AMI 2 R 2.9 BioC 2.3 Elastic-R EBS 4 Data Set VVV Amazon Elastic Block Stores Elastic-R AMI 2 R 2.9 BioC 2.3 Elastic-R.org Elastic-R EBS 4 Data Set VVV

16 Modes of access Individuals with AWS accounts  Standard AMIs (Amazon Machine Images): paid-per-use to Amazon. For Academic use and trial purposes  Paid AMI: Paid-per-use software model. For business users Individuals without AWS accounts  Trial tokens, purchased tokens, tokens granted by other users.  Resources (data science engines) shared by other users  Individual subscribtions Companies/Educational & Research Institutions  Dedicated Platform and AMIs on an Amazon VPC (Virtual Private Cloud). Paid via subscribtion

17 Rethinking Virtual Research and Teaching Free researchers from the IT service dictatorship. Give researchers self-service access to the IT resources they need Allow Researchers to share without restrictions Make real-time collaboration a free and ubuiquitous service Allow researchers to produce and publish to the web advanced applications/services without recourse to developers/admins

18 Rethinking Virtual Research and Teaching Make the cloud an ecosystem for Open Science where all research artifacts can be produced, discovered and reused Allow Researchers to « sell » the software/models/algorithms/techniques they invent seamlessly Bridge the gap between the different computational research tools: interconnect SCEs, workflow workbenches, Documents editors…

19 Rethinking Virtual Research and Teaching Provide affordable and reliable tools for remote education  High-quality voice and video chat for a large number of users  Self-service collaboration tools: Editors, White boards, IDEs, etc.  Modules for Traceability/reproducibilty Extend existing on-line courses platforms to include capabilities such as:  Companion software environments in SaaS mode  Collaborative problem solving tools  Interactive courses  Tokens for Ready-to-run e-Learning applications  E-Learning environments’ visual designers

20 Demo Register to Elastic-R academic and trial portal ( ) Create Data Science Engines using trial tokens Work with R, Python and Scientific Spreadsheets in the browser Share Data Science Engine and Collaborate Use The Visual and Collaborative Scientific Applications designer to create and publish to the web an interactive dashboard Connect to the remote Data Science Engine from withing a local R session, push and pull data, execute commands and show impact on the dashboard

21 Conclusion Elastic -R unlocks the potential of the cloud for Data Scientists and Educators With Elastic-R, the cloud becomes a cyberspace for collaborative research and sharing and an eco-system suited for open Science, open innovation and open education Elastic-R improves dramatically the productivity of the Data Scientists: The entire Data Science Factory chain, from resources acquisition to services and applications publishing, becomes under their direct control Elastic-R provides Analytics-as-a-Service platform that can extend any existing portal or application

22 What to do Next Register to Elastic-R and try the HTML 5 Workbench and the collaboration Download the R package elasticR and use it to access the cloud from local R sessions Download the Java SDK and try to create your first Analytical application using AWS and the most advanced tools for programming with data. Get in touch with me to explore potential partnerships and collaborations

23 Contact details Karim Chine