KAREN Tools of the trade Today s Plan

KAREN Tools of the trade!

Today’s Plan o o o Introduction Collaboration – now and in the future Lunch Tools Capability Development Wrap up 2

e. Research infrastructure Middleware Applications Identifiers Authentication Authorization Resource Management Distributed Science and Learning Digital Libraries Collaborative Science and Education Access Policies Legal and Institutional Privacy Credit: NSF Network MIddleware Initiative, 2004. Resources Computation Information Instrumentation Network

Middleware o Software programs that provide essential access, communication, accounting, security, trust, and coordination services between the (computational and data) resources of the grid and the higherlevel services that use them. Credit: Distributed Systems Technology Centre (DSTC). E-Research Middleware: The Missing Link in Australia’s e-Research Agenda. http: //www. itee. uq. edu. au/~eresearch/papers/e. Research. Middleware. pdf 4

Some types of middleware o Collaborative research environments n n o Grid services n n n o Video conferencing Support for collaboration Data storage, transfer and synchronisation Resource scheduling, monitoring and management Workflow management Federated identity & access management n n Authentication, authorisation, access control, certification Goal is single ‘sign-on’ for distributed resources 5

Videoconferencing o o Widely used in CRIs and universities Can be ISDN or IP-based videoconferencing can be routed over KAREN Different tools for different situations n n n Access Grid – groups EVO – desktop High definition 6

Access Grid o o Group-to-group collaboration between potentially large numbers of sites Room-based advanced videoconferencing suite with shared applications n n n Visualisation Remote control of instruments Data-sharing Shared whiteboard Shared browser 7

8

Access Grid New Zealand nodes Credit: Access Grid New Zealand. http: //accessgrid. karen. net. nz 9

NEES at Auckland 10 Credit: Be. STGRID. http: //www. bestgrid. org

Personal Access Grid 11

Desktop Collaboration - EVO o o o o User operated Web cam and head set on desktop Collaborators go to a web link Cross platform High quality video and sound, multiple cameras Shared whiteboard and desktop Recorded session No associated costs Credit: Be. STGRID. http: //www. bestgrid. org 12

EVO (Enabling Virtual Organisations) Credit: Be. STGRID. http: //www. bestgrid. org 13

14

15

16

17

18

19

Want to try EVO? o o All the information you need is at http: //www. bestgrid. org/index. php/Getting_ Started_with_EVO_in_New_Zealand You need a webcam and headset (Be. STGRID offers recommendations) EVO is at http: //evo. caltech. edu Email us if you would like to meet in EVO for a quick demo 20

Videoconferencing interoperability o Different systems in use n n o o o Different requirements Historical Collaboration requires interoperability Multipoint Control Units (MCUs) and ‘bridging services’ enable different systems to talk to each other Few of these services available in NZ now, but should develop in next couple of years 21

Virtual Research Environments (VREs), or ‘Collaboratories’ o Researchers want spaces where they can easily and securely n Communicate o o n n In real time, e. g. instant messaging Asynchronously, e. g. email, bulletin boards Share documents and other resources Manage a "virtual organisation" e. g. meeting calendar 22

What happens now? o o Researchers use project websites, wikis and email Ad hoc workspaces are created for individual projects Organisational tools can’t be used for collaboration with external partners Technical expertise to set up and maintain 23

Example VRE: Sakai o o o o o Web based and researcher driven Announcements Chat rooms Discussion boards Email lists and archive Shared resources (documents and files) Calendar Wiki And more… 24

Sakai o o http: //sakaiproject. org/ International open source development Software for e-learning and e-research 149 production R&E organisations n o 115 pilots Auckland University’s Be. STGRID hosts a Sakai Collaboration Server for KAREN at http: //sakai. bestgrid. org/portal 25

Sakai Foundation Members Credit: http: //sakaiproject. org 26

My Workspace Tools Sakai worksite setup 27

Shared Project Workspace 28

Shared Resources 29

VREs: Future vision o Support day-to-day work by integrating access to n n n Communication, including audio and video Hardware and scientific equipment Software applications, e. g. analysis tools Automated workflows Data repositories and library resources Common desktop applications, e. g. email. 30

Sakai’s Future Vision i. Phone Sakai 31

What is Grid Computing? o o o The next generation computing infrastructure to support the growing need for computational based science Analogy with power grid User has no knowledge of resource (data or computational) location 32

A Grid in Action site B authentication site C site A user 1 request data analysis 2 contact resource broker 5 results site G site D 3 initiate tasks 4 access parameter values site F site E 33

Driving a Grid o Grid tools n o o Globus, Condor, Sun. Grid Engine Scripts Workflows Portal Web service Submit Command condor_submit Job. cmd universe = vanilla environment = path=c: WINDOWSSYSTEM 32 executable = hanoi. exe Transfer. Input. Files = cygwin 1. dll arguments = 10 28 $(Process) output = hanoi. $(Cluster). $(Process). out error = hanoi. $(Cluster). $(Process). err log = hanoi. $(Cluster). $(Process). log Requirements = (Op. Sys == "WINNT 51") && (Arch == "INTEL") Rank = Mips Should. Transfer. Files = YES When. To. Transfer. Output = ON_EXIT queue 512 34

Kepler Workflow Tool o Example of the 'R' system in a Kepler workflow 35

my. Experiment. org 36

The NZ Grid Scene o Be. STGRID n n n Auckland, Massey and Canterbury with funding from TEC Data grid 100 Tbytes Computational Grid o o o Canterbury Massey cluster Other n n o Ag. Research VUW “desktop” grid n n o 250 Unix w/s managed by SGE 900 Windows w/s managed by Condor Gravitational Lensing n n 2000 cpu hours/day OGLE-390 event o Earth sized planet o 23, 000 light years away 37

Identity Management Defined o What is Identity Management? “Identity management is the set of business processes, and a supporting infrastructure, for the creation, maintenance, and use of digital identities. ” The Burton Group (a research firm specializing in IT infrastructure for the enterprise) o o Identity Management in this sense is often called “Identity and Access Management” (IAM) What problems do Identity and Access Management address? Credit: Keith Hazelton, UW-Madison, via Ken Klingenstein, Director, Internet 2 Middleware and Security 38

Identity and Access Management o o “Hi! I’m Lisa. ” (Identity) “…and here’s my Net. ID / password to prove it. ” (Authentication) “I want to do some E-Reserves reading. ” (Authorization : Allowing Lisa to use the services for which she’s authorized) “And I want to change my grade in last semester’s Physics course. ” (Authorization : Preventing her from doing things she’s not supposed to do) Credit: Keith Hazelton, UW-Madison, via Ken Klingenstein, Director, Internet 2 Middleware and Security 39

Identity Management Today o Tedious registration o Inconsistent user data o Different login processes Literature DB o Different passwords Research DB o Unprotected resources because of difficulty o University A Expensive interinstitutional access Student Admin Web mail e-Learning Library B e-Journals University C e-Learning Authentication Authorization Resource 40

Tomorrow’s Vision No user registration or maintenance at resource o Single login process o Many new resources available to users Literature DB o Enlarged user community for resources Research DB o Authorization independent of location o University A o Efficient implementation of inter-institutional access Student Admin Web mail e-Learning Library B e-Journals University C e-Learning Authentication Authorization Resource 41

Shibboleth o o Authenticate at home organisation Authorise at resource provider without knowing user’s identity Credit: Ken Klingenstein, Director, Internet 2 Middleware and Security 42

Trust Federations o o Technologies like Shibboleth need the resource provider and home organisation to trust each other. Australia n n o Functioning trial federation Funding to develop national trust federation (AAF) New Zealand n n Universities looking to join Australian federation Project starting to assess organisational IAM capability within the KAREN community 43

Closing Observations o o Key tools and technologies are available NZ organisations must start aligning with standards to achieve interoperability Not just technical issues – sharing requires policies and processes, and these can be even more difficult and time-consuming ‘Trust federations’ are emerging - NZ must be part of these if we want to collaborate internationally 44