View
2
Download
0
Category
Preview:
Citation preview
Alex ReidAdvisor, eResearch & Middleware, AARNet
&Honorary Professorial Fellow, University of Western Australia
Networks DisruptUniversities?
CAUDIT Enterprise Architecture Symposium21 November 2011
Perth
Alex Reid: Networks Disrupt - 21-Nov-11
©
Thesis
Alex Reid: Networks Disrupt - 21-Nov-11
• Networking is a key piece of data centre, enterprise, national and international e-infrastructure, and has been for many years.
• Indeed, it is perhaps so taken for granted nowadays that it has largely become invisible to end users.
• Enterprise e-infrastructure architects need to ensure that they don't also overlook its critical role not just in supporting the enterprise mission, but also in shaping it.
• Universities are essentially information enterprises, and ICT therefore plays a vital role - both in teaching/learning and in research, as well as in supporting corporate functions.
• In this context, networking has the potential to be a highly disruptive technology, which could dramatically change the nature of both learning and research, as well as the very shape of the enterprise itself.
2
©
Contents
• Network Developments– National/International R&E– Community– Enterprise– Data Centre
• Disruptive Technologies• Impact on Higher Education
– Teaching– Research– Corporate Systems
• Conclusions
Alex Reid: Networks Disrupt - 21-Nov-113
©
Maintain World-Class Standing
• Upgrade backbone to 100Gbps• Customer connections: 1-> 10-> 40Gbps• Extend to unreached locations (eg Darwin, Hobart)• Dedicated wavelengths, circuits• 24x7 NOC• VERY high availability, reliability• Resilient international links (eg 10Gbps from Perth)• etc
Alex Reid: Networks Disrupt - 21-Nov-116
©
AARNet’s Strategic Roadmap for 10+ Years
Alex Reid: Networks Disrupt - 21-Nov-11
Source: AARNet
10G
1G Access
100G
10G Access
AARNet3
32x 10G 40 x 100G
Customer Access
Internet Backbone
DWDM Backbone
AARNet20AARNet4
1.5P
320 x 1P
100G Access
2006 2010/11 2020?
7
©
SKA will require solving Petabit networking
Alex Reid: Networks Disrupt - 21-Nov-11
1989 1991 1996 2010
34 Mbps
2 Mbps
48 Kbps
10 Gbps
BA
ND
WID
TH
TIME
2 Pbps
2029
Source: AARNet
9
©
Broadband Penetration
Alex Reid: Networks Disrupt - 21-Nov-11
http://www.broadbandcommission.org/Documents/A_Platform_for_Progress_Report_Overview.pdf
10
©
Data Communications Developments
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Sood & Tellis, 201111
Chart of Data Communications technology performance for improvements for copper, fibre, wireless 1962-2006.
©
NBN Impact
Alex Reid: Networks Disrupt - 21-Nov-11
http://www.nbn.gov.au/media-centre/videos/the-nbn-explained/
12
Part of animated video promoting the NBN.
©
NBN Impact
Alex Reid: Networks Disrupt - 21-Nov-11
Education & Learning:
Ubiquitous high-speed broadband provides a way to improve the educational capabilities of students, teachers and the wider community. IBES is fostering innovative research to enrich learning and demonstrate how connecting Australians via broadband opens up new opportunities.
Institute for a Broadband-Enabled Society (IBES):http://broadband.unimelb.edu.au/
13
©
NBN Impact - IBES
Alex Reid: Networks Disrupt - 21-Nov-11
Health & Wellbeing:
IBES is fostering multidisciplinary research that investigates the use of next generation broadband technologies in health care. New broadband technologies can improve the quality and safety of care, access to services, and innovations in clinical care and health maintenance.
14
©
NBN Impact - IBES
Alex Reid: Networks Disrupt - 21-Nov-11
Services & Business Transformation:
High-speed broadband has the power to dramatically impact the quality, innovativeness and reliability of services delivered to businesses, governments and the not-for-profit sector. This research will assist the development and implementation of new tools and technologies.
15
©
NBN Impact - IBES
Alex Reid: Networks Disrupt - 21-Nov-11
Social Infrastructure & Communities:
Broadband availability drives changes in patterns of social interaction across a wide spectrum of activities. This theme is fostering innovative interdisciplinary research that will investigate the emerging opportunities and challenges of a broadband-enabled society.
16
©
NBN Impact - IBES
Alex Reid: Networks Disrupt - 21-Nov-11
Network Deployment & Economics:
The technology base for advanced broadband network infrastructure is well developed. Research in this theme is focusing on advanced technologies for future enhancement of fibre and wireless networks and how to measure and model the benefits from broadband.
17
©
NBN Business
19891995
2000
NOW
2030
10 Mbps
1 Pbps
BA
ND
WID
TH
U p l o a d
D o w n l o a d
1 Gbps
NBN Home
ADSL 2+
ADSL 1
ISDN PRI
Dialup
Media rich interactive cloud services(new & future web-applications)
1 Pbps
1 Gbps
1x Low-def videoconference
1x HD Videoconference
1 Tbps
1 Tbps
100 Mbps
Education & Research institutions
1 Pbps= 1,000 Tbps
= 1,000,000 Gbps
What’s next ??
Dial-up/ISDNADSL
NBN
AARNet and Public Internet speeds
Alex Reid: Networks Disrupt - 21-Nov-1118
©
Networking: Direct Consequences for EA
• Match campus capabilities with national/ international• Ubiquitous connectivity• Accommodate high-end users’ needs (eg wavelengths)• Prevent Firewall congestion• High availability, quick response• Upgrade paths/ strategy/ flexibility• Data Centre networking (eg QFabric/ BrocadeOne, Cisco
Universal Computing System for simplicity, flexibility, efficiency, …)
• Network virtualisation• …
Alex Reid: Networks Disrupt - 21-Nov-1119
©
Disruptive Technology: Innovator’s Dilemma
• Incremental improvements (sustaining technological innovation) vs Disruptive technological innovation.
• Start by creating large growth opportunities away from the core of the established market.
• Disruptive Innovation is a dynamic form of industry change that unlocks tremendous gains in economic & social welfare:
– Eliminates inefficiency by creative destruction– Drives improvements via creative construction– Allows broader group of people to do things only an expert or wealthy
group could do– Convenience increases, consumption rises, prices slump
• Established providers usually cannot respond (eg out of loyalty to existing customer base), or do not see this as a threat.
• They listen to their customers and make gradual improvements or enhancements, but this outstrips the needs of the core market.
Alex Reid: Networks Disrupt - 21-Nov-1120
©
Disruptive Technologies
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Christensen, 2003, Figure 1
21
Chart of Definitions of Sustaining and Disruptive Innovations: Performance vs Time.
©
Disruptive Technologies
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Christensen, 2010, Figure 2
22
Chart of the 2 Types of Disruptive Innovation: Low-end disruption & Non-consumers.
©
Disrupted Industries/Markets• ModelT Ford vs other cars• Home power tools ($20) vs professional tools ($150)• Canon countertop copiers vs Xerox business copiers• Department stores vs Speciality shops• Discount stores vs Department stores• Catalogue companies vs Department stores• Amazon vs high street book stores• Mobile phones vs landlines• Bell telephone vs telegraph• Desktop publishing vs Offset Print shops• Email vs Fax• Email vs Postal services• Minicomputers vs Mainframes• Microcomputers vs Minicomputers & Mainframes• Cheap airlines vs Standard carriers• Online travel vs Travel agents• Community colleges vs State universities• Online universities vs Brick & Mortar universities
Alex Reid: Networks Disrupt - 21-Nov-1123
©
Disruptive Technologies
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Christensen, 2003, Figure 124
Chart of Centralization Followed by Decentralization in Computing.
©
Disruptive Technologies
Alex Reid: Networks Disrupt - 21-Nov-11
http://radar.oreilly.com/2011/10/post-pc-revolution.html/
25
Chart of Each Computing Wave Dwarfing & Disrupting its Predecessor.
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Christensen, 2008, Figure 5.226
Chart of Characteristics of the Two Stages of Disruption in Learning Content Delivery.
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Christensen, 2008, Figure 4.2
27
Chart of The Pace of Substitution of Computer-Based Learning for Monolithic Learning.
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-1128
Chart of The Pace of Substitution of Computer-Based Learning for Monolithic Learning, when converted to S-Curve.
Source: Christensen, 2008, Figure 4.2
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2012154
29
Chart of Percentage of Undergraduates Enrolled in Distance Education Courses or Degree Programs (USA,) by year 1999/200, 2003/2004, 2007/2008.
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Hofstetter, F T: Using the PLATO System: Features, Research and the Future, ADCIS Proceedings, 1981
30
Chart showing growth in cost of conventional teaching (public schools, state colleges, private colleges) vs decline in cost of Plato on-line learning, 1972-1986.
©
Learning: Conversational Framework
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Laurillard, 200231
Chart of Diana Laurillard’s Conversational Framework for Learning, showing interactions between teacher & student.
©
Learning: Conversational Framework
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Laurillard, 2002
For On-Line Learning:
32
Chart of Diana Laurillard’s Conversational Framework for Learning, showing interactions between teacher & student – for the case of on-line learning.
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
http://en.wikipedia.org/wiki/Medieval_university
33
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
http://en.wikipedia.org/wiki/Lecture_hall
34
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-11
http://socrates58.blogspot.com/2009/02/luther-meets-his-match-part-vii-erasmus.htmlhttp://en.wikipedia.org/wiki/Desiderius_Erasmus
35
Same picture of Erasmus doctored to show him instead using a Mac.
©
Disruption in Learning
Alex Reid: Networks Disrupt - 21-Nov-1136
Images from University of Phoenix:http://www.phoenix.edu/
Images from Think Education Group:http://www.think.edu.au/
©
Disruption in Research
Alex Reid: Networks Disrupt - 21-Nov-11Source: Hey, 2009
37
Chart of the Four Science Paradigms:- Empirical- Theoretical- Computational- Data Exploration (eScience)
©
Disruption in Research
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Hey, 200938
Diagram of a High-level view of a research infrastructure that brings together knowledge bases and computational services.Savas Parastatidis, Figure 1.
©
Disruption in Research
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Hey, 2009
39
Chart of explosive growth in volume of Genome Sequencing Projects on Genomics OnLineDatabase 1997-2009, reaching 4,370 projects by Jan-09.Christopher Southan, Figure 3.
©
Disruption in Research
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Hey, 2009
40
Chart of exponential decline in rate of dissemination of medical discoveries to clinical practice over the past 2,500 years, and over the past 150 years, projected to reach ~0 by 2025.Michael Gillam, Figure 2.
©
Disruption in Research
• Networked world impacts Research publications in 3 ways:
– digital space allows authors to interact with their peers and readers, placing a new emphasis on the process of writing rather than the final product, with the potential of a text that is forever open to comments and rewriting;
– digital texts are interlinked in a way that printed books are not;– internet offers new multimedia tools, including the use of videos
and images, audios and blogs as a way of expressing thoughts instead of, or in conjunction with, words.
• Humanities as well as Science⇒ dynamic, interconnected, multi-media publications⇒ Open Access, Innovative Publishing models.
Alex Reid: Networks Disrupt - 21-Nov-1141
©
Research CloudNectar's Research CloudBuilding a national Research Cloud for Australian researchers in partnership with Australian institutions and research organisations:
– Hosting research applications in the cloud;– A computational resource which complements existing and new
supercomputing facilities;– Fostering innovation in eResearch services in the cloud.
Benefits:• supports research collaboration by making it easier to share access
to innovative research applications on a national scale.• cost-effective computational infrastructure which complements
existing and new supercomputing facilities.• fosters innovation in eResearch and research software services by
allowing innovative new services for researchers to be rapidly deployed at low cost.
Alex Reid: Networks Disrupt - 21-Nov-1142
Source: NeCTAR
©
Cloud Computing• Jeff Weiner, LinkedIn CEO: Cloud makes it easier and cheaper than ever for anyone
anywhere to be an entrepreneur and to have access to all the best infrastructure of innovation.
• Bill St Arnaud: It is not only the private sector that is creating new opportunities in this field, but funding councils and R&E networks are also playing a critical role in the transformation of the future computing. The recent announcements by Internet2, SURFnet, JISC/JANET, etc to broker commercial cloud services for researchers are great examples of this trend.
• Juniper: QFabric enables exponential improvements in data centre speed, scale and efficiency, by removing legacy barriers and improving business agility.
• Sinclair Schuller, CIO Apporenda: As the sea-change that is cloud computing washes over the industry, certain aspects of cloud that make their way into the enterprise as private cloud will move enterprise architecture from fairy-tale to reality. The most important of the cloud form factors for enterprise IT will be Platform as a Service (PaaS).
• St Arnaud: cloud computing is fueling the next startup boom.• St Arnaud: NRENs are transforming the global Internet through disruptive innovation.
Alex Reid: Networks Disrupt - 21-Nov-1143
©
Cloud Computing• Michael Hugos: telecom providers around the world need to actively promote
awareness of cloud computing and explain the benefits of cloud computing so as to create demand for products that they are potentially well suited to make and deliver over their networks – eg PaaS and SaaS.
• Scott Jamieson, a2b Fiber Inc: With its high bandwidth capability and low latency, local access fibre will be a catalyst in the growth of cloud services.
• Juniper: Enabling a virtualized, cloud-ready data center means re-thinking the very fundamentals of networks.
• St Arnaud: Researchers are voting with their feet (and wallets) to use commercial cloud applications because of their ease of use, simplicity, ability to scale quickly, no need for upfront capital costs for computing, and no bureaucratic hassles to access university computing resources. A good example is Galaxy (http://wiki.g2.bx.psu.edu/Admin/Cloud), which is taking the genomics and bio-informatics research community by storm.
• St Arnaud: Locate data centers in locations where there is cheap/green power, and connect with high-capacity fibre links (eg Google & Amazon @ Oregon; Pawsey).
Alex Reid: Networks Disrupt - 21-Nov-1144
©
Cloud Computing
Alex Reid: Networks Disrupt - 21-Nov-1145
• Cloud turns so many things on their heads:– Locate the data centre where there is cheap, clean power:
• Google, Amazon, Microsoft, Yahoo at Columbia River, Oregon (14 hydroelectric dams)
– Low cost of entry to expensive infrastructure, eg HPC– Scalable– Locate services where they are best provided– Does this apply to universities too?
Bonneville Dam: http://en.wikipedia.org/wiki/Columbia_River
©
Education Hype Cycle
Alex Reid: Networks Disrupt - 21-Nov-11
Source: Gartner, 201046
Gartner Hype Cycle for Education, 2010, showing a plethora of relevant technologies “in the pipeline”.
©
Education Hype Cycle
Alex Reid: Networks Disrupt - 21-Nov-11Source: Gartner, 2004
47
Gartner Hype Cycle for Higher Education, 2004, showing Course Management Systems, etc already on the “plateau of productivity”.
©
Conclusions1. How is IT changing the global education ecosystem and impacting the future
workforce in society? There are blurring boundaries and increased interdependencies. It's no longer possible to treat a higher education institution like an ivory tower. [Chistensen]
2. How should higher education institutions strategize and govern to make the most out of IT?This gains importance with increasing turmoil. The simple fact that IT now touches virtually all stakeholders in the education ecosystem puts special pressure on the CIO not only to choose well, but also to communicate well. [Chistensen]
3. How should higher education institutions invest in applications, systems and infrastructure?The higher education CIO has a key role as both inspirer and supplier of IT services. In the role of inspirer, CIOs must identify disruptive and enabling technologies that can transform the institution and inject it early into the strategic planning process. In the role of supplier, CIOs must be diligent in choosing the right solutions and vendors to meet often varied demands for cost-effectiveness, security and availability, for example. [Chistensen]
Alex Reid: Networks Disrupt - 21-Nov-1148
©
Conclusions4. Management of IT: (of the design, architecture, operation of IT infrastructure) this is
essentially a “complex” process, not amenable to formulaic solutions; therefore management should be pragmatic, not idealised; agility is the key to success, not SOEs. Complexity thinking vs Systems thinking. [Snowden]
5. Relish Hardship: it is in times of recession and financial deprivation that innovation and new ideas thrive – look out for low-cost disruptive technological innovations.[Snowden]
6. Measurement Obsession: beware of the sort of measurement obsession that leads to neglect of the primary goal; this has beset parts of the NHS in UK to the point where Staffordshire nursing staff spent so much time monitoring that they ignored their patients & army medics had to be drafted in to keep the hospital running. [Snowden]
Alex Reid: Networks Disrupt - 21-Nov-1149
©
Principal References• Christensen, Clayton & Raynor, Michael: The Innovator’s Solution, HBR Press, 2003.• Christensen, Clayton et al: Disrupting Class, McGraw-Hill, 2008.• Laurillard, Diana: Rethinking University Teaching, RoutledgeFalmer, 2002.• Hey, Tony et al (Eds): The Fourth Paradigm, Microsoft Research, 2009.• Christensen, Clayton: How to Manage the Disruption of Higher Education, EDUCAUSE HE
Futures Forum, 2010.• Christensen, Clayton et al: Disruption in Education, EDUCAUSE Review, Jan/Feb 2003 p44.• Lowendahl, Jan-Martin et al: Key Issues for Higher Education, 2011, Gartner, 14-Mar-11.• Snowden, David & Boone, Mary: A Leader’s Framework for Decision Making, HBR, Nov
2007, p69.• Sood, Ashish & Tellis, Gerard: Demystifying Disruption: A New Model for Understanding &
Predicting Disruptive Technologies, Marketing Science, Mar 2011, p339.• Fitzpatrick, Kathleen: Planned Obsolescence: Publishing, Technology, and the Future of
the Academy, 2011.• Lowendahl, Jan-Martin: Hype Cycle for Education, 2010, Gartner, 26-Jul-10.• Yanosky, R et al: Hype Cycle for Higher Education, 2004, Gartner, 14-Jul-04.• St Arnaud, Bill: Future Internet, R&E Networks, Green Internet, Green IT
http://billstarnaud.blogspot.com/
Alex Reid: Networks Disrupt - 21-Nov-1150
Recommended