INDIANAUNIVERSITYINDIANAUNIVERSITY 1 APAN – An historical perspective and a view toward the future Michael McRobbie PhD Vice President for Research Vice.

1

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

APAN – An historical perspective and a view toward the future

Michael McRobbie PhD

Vice President for Research

Vice President for Information Technology and CIO

Professor of Computer Science

Indiana University

[email protected]

2

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Foundation of APAN

• March 1996 – APEC Symposium for Realizing the Information Society held at Tsukuba, Japan– Kilnam Chon and Michael McRobbie gave keynotes

separately proposing the establishment of an Asia Pacific high speed network

– Initial meetings held where first draft of what were to become the APAN charter was prepared (Chon, McRobbie, Goto, Konishi, Goldstein (!), others?)

– From the first draft:• “proposed to interconnect [the regional high-speed

testbeds] … at the rate of the fastest testbeds to form the AP Testbed [Asia/Pacific High Speed Network testbed]”

• “additional opportunity to establish a bridge from USA to AP Testbed through NSF …”

3

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Foundation of APAN

• Further organizational meetings held in Tokyo in May, 1996

• June 1996 - APEC/APII Forum held in Seoul, Korea– Chon and McRobbie present joint keynote “Towards an

Asia-Pacific Advanced Network” that outlines a framework for APAN

– Additional meetings in Seoul led to formal establishment of APAN

– Considerable debate as to direction of APAN - regional coordinating organization or builder of infrastructure. Eventually a little of both

4

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Asia-Pacific Advanced Network

• “Establishment of an Asia-Pacific Network (APAN) broadband network that could:– develop Asia-Pacific hubs– interconnect national broadband test beds in the region– interconnect national research networks in the region– support international collaboration between groups in the

region that requires connectivity of this speed– allow coordinated involvement of the region in GIBN– contribute to the development of an Asia-Pacific

Information Infrastructure as part of a Global Information Infrastructure

– support connectivity at lower speeds for developing countries”

Given this state of networking in 1996, what were the plans for APAN? (from Chon/McRobbie 1996)

5

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Structure of a Global Information Infrastructure

1996

6

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Application Technologies(Chon/McRobbie 1996)

• Remote virtual reality (telepresence)• Tele-collaborative environments (colaboratories)• Remote access to specialized equipment and facilities• Multimedia• Data access and data fusion”

“There are various application technologies that could form the basis of regional collaboration that would require a broadband network, e.g.:

7

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Application Areas(Chon/McRobbie 1996)

• Medicine (telemedicine)• Distance education (virtual university/institute)• Remote sensing data• Environment• Weather• Mining• Infrastructure development• Agriculture• Fishing”

“These in turn could be applied to problems in the following application areas:

8

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

A Flagship Project (Chon/McRobbie 1996)

• “Many countries in the region have a major requirement for access to remotely sensed earth observation data for economic, scientific, environmental, and other reasons

• This data will:– be stored in different places– be from different kinds of remote satellite sensors (e.g.

photometric, radar infrared, etc.)– require terabytes and eventually petabytes of massive

storage using advanced database techniques– be subject to various levels of pre-processing– require high performance computing systems,

sophisticated visualization techniques (e.g. 3D, VR, etc.) and constantly changing analytical tools for its processing

– comprise files >1GB in size– comprise long time-series of data”

9

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

State of networking at this time

• No GEANT, only Ten-34 in Europe• No Abilene, only NSFnet in the US• Asia-Pacific connectivity heavily US-centric• Global interconnects rudimentary

10

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Ten-34 research network

11

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

NSFnet

12

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Asia-Pacific Connectivity

13

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Global Research

• So at a global level, what happened over the next 8 years in research?

14

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

E-Research

• Research become almost totally digital in all disciplines - e-research

• The data of research is now being generated, collected, processed, analyzed, vizualized & stored in digital form

• Simulations & modelling are being carried out completely digitally

• The historical & contemporary archives of science are being converted into digital form

15

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Global e-Research

• Network-enabled world-wide collaborative e-research communities (grids) have rapidly formed in nearly every discipline from anthropology to zoology – each can number 1000s or more

• These communities are based around e.g large-scale possibly distributed data holdings, a few large sometimes unique instruments or distributed complexes of sensors

• global e-research communities carry out research based on this data using computation, storage and visualization facilities distributed world-wide

• All of this is global cyberinfrastructure• The digital data of e-research can be shared with collaborators

anywhere on earth• e-research is completely international – it knows no boundaries• Global high-performance networks are the critical glue

that makes e-research possible

16

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

The Global e-Research Community

• How big is the global e-research community?• Based on NSF statistics approx 1 million in

higher education alone in USA• Maybe 1 million for higher education in the

rest of world based on bibliometric measures• This does not include researchers in industry &

government labs

17

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Enabling Global e-Research

• NRENs (e.g.Internet2/Abilene, AARNet) connect e-researchers nationally

• Regional networks (e.g. GEANT, CLARA) connect e-researchers regionally

• International connections (e.g. TransPAC) interconnect these regional networks

• This whole delicate fabric is essential for the 2 million global e-researchers in higher education & many more

• But the development of this fabric cannot be left to chance - good public policy demands that governments, agencies & multi-lateral organizations ensure its continued development & functioning

• APAN has taken on the vital role of interconnecting e-researchers in the Asia Pacific region

18

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

APAN Now

• Enormous progress has been made• Great credit due to Professor Chon, Goto-san, Konishi-

san and scores of others from all APAN countries in the region!

• Interconnections within Asia have increased• Connections between Asia and the world have

increased– As an example, in 1998 TransPAC was 34Mbps.

Today, TransPAC is 2 x 2.5Gbps. Soon, it will be 2 x 10Gbps

– NSF HPIIS Program vital here

19

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

APAN Now

20

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

APAN remains a work in progress

• Old issues continue – Connections to US remain strong, but a true

Asia-Pacific backbone (like GEANT) has yet to be built• more north/south linkages within Asia are

needed and better connectivity to Europe, possibly via west Asia

– Applications are developing but stronger intra-Asia applications, grid applications in particular, need support• APAN Working Groups play a vital role here

21

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

New issues demand attention

Application measurement, particularly end-to-end network performance measurement is increasingly critical (deterministic networking)

Security must now be a consideration for every application and every network.

22

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Central Issues for APAN this decade

• Stronger linkages between applications and infrastructure - neither can exist independently

• Stronger application and infrastructure linkages among APAN members.

• Continuing focus on APAN as an organization that represents infrastructure interests in Asia

• Closer connection between APAN the infrastructure & applications organization and regional political organizations (e.g. APEC, ASEAN)

23

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

A view into the future

• Importance of global interconnections will increase in importance even more due to the international and collaborative nature of e-research

• Critical nature of security must be reflected in every aspect of cyberinfrastructure

• Importance of measurement and deterministic networking will increase

• Applications (particularly gird applications) will continue to press network technology at the campus level, nationally and internationally

24

I

N

D

I

A

N

A

U

N

I

V

E

R

S

I

T

Y

Conclusion

• APAN has completed a very successful establishment phase

• Now it must move forward as an active mature network, supporting the broad, diverse interests of e-research in Asia, with all the organizational and political complexities that entails