Background
The US OptIPuter project is "A Powerful Distributed Cyberinfrastructure to Support Data-Intensive Scientific Research and Collaboration". It is NSF’s largest single computer science project and is led by the world’s premier ICT institute: The California Institute for Telecommunications and Information Technology, CAL(IT)-22. While it explores next generation conceptions of a computer, a primary practical benefit is that the project has driven high-end network deployment and performance tuning.

Aims
This project aims to install a new computer display at UQ-VisLab (co-located with the HPC Unit) that will have the highest capacity in Australia: 36 Mega-pixels (compared to regular desktops with 1-2 Mpixel). This display would be the first OptIPuter portal in Australia and would cement for QCIF a growing partnership with CalIT2 and EVL.
Methodology
The primary objective of this project is to more tightly couple computing, data and display devices by computer networks. This is required because, increasingly, research now involves larger data sets and more intensive computations. A significant obstacle for end users is whole-of-systems performance, which is demonstrably poor. And users need better visualisation interfaces to these larger outputs. Additionally, deployed networks can have high theoretical performance (1 Gbps [Giga bits per sec.]), but routinely are not well tuned so that measured data transfer rates are more typically ~5 Mbps. That would be equivalent to driving on a freeway at 0.5 km/h.
The physical system to be built is a tiled LCD display, with associated PC graphics systems and network manager; a simple mechanical frame supports and aligns the array of LCD displays.
![]() |
![]() |
Participants
Prof. B Pailthorpe, Dr N Bordes,Dr David Green, C Willing, D Kosovic, J White - UQ
Dr J Young - QUT
J Bell - CQU
International Participants
Prof. Larry Smarr - CALIT2, UCSDDr J Leigh - EVL, UIC
Reports
Progress Report (February 2008) (73 KB PDF)Project Proposal - (341 KB PDF)
Construction progress - UQ Vislab


