Download to read offline
Uploaded byChris Dwan
GP-Write computing group
This document provides a charter and roadmap for a Computing, Data, and Informatics Working Group. It discusses their vision of enabling data, computing, and identity services at unlimited scale. It highlights how information technology has been critical but also a source of tension in large projects like the Human Genome Project. The document outlines current enabling technologies like machine learning, blockchain, and DevOps practices. It identifies key challenge areas the working group will focus on, including identity and authorization, information security and privacy, and issues around data storage in multi-cloud environments. The working group members are then listed.
More Related Content
Similar to GP-Write computing group
![ALKALIPHILES.ppt].pptx introduction and important](https://cdn.slidesharecdn.com/ss_thumbnails/oumtaoxdr1o9zqxl0yqy-alkaliphiles-251104125907-0f4755cb-thumbnail.jpg?width=640&height=640&fit=bounds)
![MUSHROOM_PRODUCTION_UNITS and mechanism 1][1].pptx](https://cdn.slidesharecdn.com/ss_thumbnails/mushroomproductionunits11-251106185651-32f95ca8-thumbnail.jpg?width=640&height=640&fit=bounds)
GP-Write computing group
- 1. Computing, Data, andInformatics Working Group Charter and Roadmap May 1, 2018
- 2. Our Vision: Data,computing, and identity services at unlimited scale
- 3. Historical Perspective In similarlyaudacious projects (Human Genome Project, HapMap, ENCODE, TCGA, …) … Information technology been a critical enabler … … It has also been a regular source of tension, frustration, and unexpected expense… A “Cray 2” supercomputer on display at the national cryptographic museum. This model of system were the fastest computers in the world until 1990. By one measure of raw performance, the Cray 2 is comparable to a single iPhone
- 4. 2018 Enablers Machine learning/ AI • Many real-world examples • Domain expertise still matters • No free lunch Blockchain • Enables trust without a central authority • Most data should never be “on-chain” • Try to see past the cryptocurrency hype • Does not change “everything” Devops / SRE / Agile Design • Progress in how we build robust, scalable digital technologies within scientific R&D • Still challenging to get right • Doubly so in academic environments A 100Gb/sec uplink connects Manhattan institutions to a high performance research and engineering network.
- 5. 2018 Challenge /Focus Areas Identity and Authorization • Need flexibility / simplicity • Without sacrificing trust and accountability Information security, privacy, and appropriate usage • Effective governance for GDPR / HIPAA / other compliance • Practical InfoSec • Clear chain of custody / ownership for commercialization Data gravity, cloud skew • There is no single best provider for data storage and computing. • Cloud providers impose both technical and financial lock-in. • We must plan for a multi-cloud environment. Even in 2018, some labs still use FedEx as a high throughput, high latency data transfer protocol
- 6. Working Group: Jason BobePersonal Genomes Brian Bot Sage Bionetworks Jack Collins Frederick National Laboratory Chris Dwan* Bridgeplate Nancy Kelley NJK Associates Amy Schwartz NJK Associates Nam Pho NYU Langone Health Bruce Wilson Oak Ridge National Laboratory *: Chairperson