Network Research Problems and Challenges for DOE Scientists Workshop
BETHESDA NORTH MARRIOTT
HOTEL & CONVENTION CENTER
BETHESDA, MD
FEBRUARY 1-2, 2016
Panel and Breakout Questions
Adobe Reader is necessary to view PDF files. If you don't have the latest version of Reader, you can download a free copy at the Adobe download site.
Panel Questions
The panel questions are available as a PDF file (13 KB).
- From the list of break-out session questions, what should be removed from the list?
- From the list of break-out session questions, what should be added to the list?
- What are the cross-cutting issues that keep you up at night?
- What does the future network look like to your science community?
- What changes are necessary and what stays the same?
- What must remain constant and what would be nice to leave alone?
- What are the foundational research issues you want to see addressed?
Breakout Sessions Questions
The breakout sessions questions are available as a PDF file (19 KB).
Overarching Questions for All 3 Break-out Sessions
- What is the mix of traffic and how will it change over time?
- What are the application drivers for your science?
- What are the basic services/functions you need/expect from the network?
- How do you expect the network to be used, and how does that change or stay the same?
Break-out Session 1: Short: 1-3 years (Terabyte/hour bulk transfer rates)
- What is the most pressing problem for scientists?
- What is the most pressing problem for network operators?
- Rank order and Discuss these issues
- status of Analysis Tools
- status of simulations and models
- status of management tools
- multi-domain troubleshooting
- protocol performance and implementations
- traffic growth and projections
- testing and deploying new tools and services
Break-out Session 2: Medium: 4-6 years (Petabyte/hour transfer rates)
- Assume that the network infrastructure consists of highly parallel and redundant physical links.
- Assume that new protocols can be deployed over the global Research and Education Network infrastructure.
- Rank order and discuss the following problems
- Effectively support for bulk data and more interactive flows
- Routing over this infrastructure
- Active queue management in routers/switches
- monitoring and managing the network
- debugging and troubleshooting
- predictable transport performance
- When must work begin on solving each specific problem (now, or delay 3 years)
Break-out Session 3: Long: 10-12 years (Exabyte/hour transfer rates)
- Assume that the network infrastructure consists of massively parallel links with multiple DWDM channels per link.
- What revolutionary changes are required (e.g.; parallel bit streams)?
- What work must start now to be ready in time?
- What work can be delayed 3-6 years?