NSF Workshop on Basic Exploration in Systems administration: Future Headings and Dreams for Systems administration Resea.


77 views
Uploaded on:
Description
NSF Workshop on Fundamental Research in Networking: Future Directions and Visions for Networking Research. Ness B. Shroff School of Electrical and Computer Engineering Network Engineering and Wireless Systems (NEWS) Group Purdue University, West Lafayette, IN 47907 shroff@purdue.edu.
Transcripts
Slide 1

NSF Workshop on Fundamental Research in Networking: Future Directions and Visions for Networking Research Ness B. Shroff School of Electrical and Computer Engineering Network Engineering and Wireless Systems (NEWS) Group Purdue University, West Lafayette, IN 47907 shroff@purdue.edu

Slide 2

NSF Workshop The NSF Workshop on Fundamental Research in Networking was held in Arlie House, Virginia, April 24-25. In participation: 27 Experts traversing an expansive range of systems administration zones and teaches. Mission: Prepare a report on the ebb and flow condition of systems administration and to express a dream for the eventual fate of principal exploration in systems administration

Slide 3

NSF Workshop: Logistics and Ground Rules Two day workshop that was mission concentrated Only 2 board sessions A considerable measure of time in breakout bunches Writing started amid the workshop Not to just have a scholarly discourse Not to compose a report comprising of a clothing rundown of individuals\' most loved subjects Represent the group everywhere, Hope that report will animate expansive future examination activities and joint efforts.

Slide 4

Fundamental Research Fundamental Research was characterized as having one or more attributes of being expansive, strong, long haul, worldview evolving, hazard taking, problematic, high-affect scrutinize that could be supported by a system, for example, the Network Research program at NSF. Traverses the whole range of systems administration research, from hypothesis to frameworks arranged exploration

Slide 5

Executive Summary: Findings and Recommendations Models of Success Case investigations of systems administration research victories Network Innovator\'s problem Need for reproducibility and benchmarking Importance and Impact of Basic Research in Networking New Applications and Paradigms Grand Challenges in Networking Quick Look at Workshop Report Focus on chose things from report here … Full report is accessible at this gathering furthermore on http://www.cs.virginia.edu/~jorg/workshop1

Slide 6

Network Innovator\'s Dilemma Looking to models of progress for future examination, we should maintain a strategic distance from the allurement of falling into the trap of Innovator\'s Dilemma (Clayton Christenson) For instance: Success of the Internet conveys the danger of smugness about the requirement for out-of-the-container considering. Neglect to watch problematic advancements that take radical new shape Past: Needed new group to think fundamentally (e.g., chime heads were abandoned in the Internet blast) Future: Do it ourselves!

Slide 7

Importance of Basic Research Basic Research in Networking addresses more foundational and hypothetical parts of systems administration. Unites procedures from an assortment of various fields to address particular issues of pertinence to systems administration. Points: create bits of knowledge and standards comprehend execution confinements propose "great" arrangements and approaches guide the advancement of conventions and control components

Slide 8

Importance of Basic Research Basic examination: key part in the comprehension and improvement of Internet and remote systems Distributed steering conventions (Bellman-Ford and Dijkstra). Multiaccess instruments from ALOHA to CSMA/CD TCP and more broad AIMD blockage control 3G booking components in remote frameworks Queueing and Large Deviations lead to comprehension of system clog and asset designation LRD properties in system activity… Also gave the force to headways in related fields . Required now like never before to plan and control systems on a much more excellent scale than some time recently! More prominent effect if fundamental exploration is sought after in conjunction with center territories and genuine issues.

Slide 9

New Applications and Paradigms Applications are and will keep on being the raison d\'etre of systems! 3 illustrative applications were picked as prone to assume an imperative part driving future systems administration. Omnipresent Tele-nearness (capacity to collaborate with others and environment without shackles of space) Universal Sensing (expansive scale sensor systems coordinated into all parts of life). Crisis Response and Disaster Recovery Networks Applications will actuate client engaged and also arrange driven difficulties Question: Are there an arrangement of excellent difficulties for systems administration?

Slide 10

Grand Challenges Classical definition "Grand Challenges are central issues in science and building with conceivably wide social, political, financial, and exploratory effect that can be progressed by applying elite registering assets." Definition utilized as a part of Workshop " As a basic issue with wide monetary and investigative effect that is as of now viewed as obstinate, and whose arrangement can be progressed by key examination ."

Slide 11

Grand Challenges Meta Grand Challenge To grow new system hypotheses, structures, and approachs that will encourage the improvement of the up and coming era of administrations and applications Indvidual Grand Challenges Network Information Theory for Wireless Networks Providing Market Incentives (The Economics of Networking) Development of flexible Networks Sensorized Universe Design of Overlay Networks Virtual Networks Discuss just initial three shots

Slide 12

Network Information Theory for Wireless Networks Classical Information hypothesis: premise for significant advancements in interchanges, mistake control, and flag handling in the course of the most recent 50 years. Fundamental confinement : portrayal of deferral is truant from customary data hypothesis. Numerous essential hypothetical improvements (booking, queueing, control, directing), there is no such brought together data hypothesis for systems.

Slide 13

Network Information Theory Challenges: Characterize as far as possible execution limites of versatile and static remote systems Capacity Delay Power Connectivity and Coverage Can we create dispersed calculations to accomplish these breaking points? System Coding? Can we accomplish requests of extent increments in system peformance by abusing fundamental advancements (physical layer, VLSI outline, and so forth)?

Slide 14

The Economics of Networking Lack of Marketing Incentives are one of the primary hindrances to the organization of new administrations on the Internet. Late history has shown that the bottleneck to enhancing the worldwide system might be financial aspects and not innovation. There has been colossal enthusiasm for valuing to oversee systems. In any case, most works don\'t consider an aggressive business sector environment. Be that as it may, the Internet is not a solitary administration supplier!

Slide 15

The Economics of Networking Major Challenge: In creating monetary instruments for the making of effective markets: Allow everybody to be a player (not just social boost) Designed for multi-supplier and multi-business environment (heterogeneous) Incentive-perfect (Don\'t expect selflessness) Scalable (12,000+ AS!) Flexible: Support distinctive administration sorts (e.g., Dependable administration, variable cost; Dependable cost, variable administration) Increases Revenue Quality Improves

Slide 16

Resilient Networking As the Internet develops in scale it confronts an inexorably unfavorable environment Due to the sheer scale, deficiencies turn out to be entirely ordinary, as opposed to the special cases Component disappointment Human operational blunders Spread of programming infections Malicious assaults.

Slide 17

Resilient Networking Challenge: To give the Internet a flexible foundation. Need to build up a fundamental comprehension of the guidelines and standards in system convention design that encourage the acknowledgment of resiliency in expansive scale systems. Can we create particular improvements that can promptly be connected to the present Internet to viably fortify its guard capacity? Could we truly plan against the sudden?

Slide 18

Conclusion Networking has seen extraordinary achievement and development Future is splendid and holds significantly more noteworthy and more fabulous difficulties Our main goal: Innovate, lead brilliant research, and instruct industry and government organizations of the huge proceeded with significance of the field. Much obliged to You!

Recommended
View more...