N e t w o r k   A p p l i c a t i o n s   a n d
  P e r f o r m a n c e   P r o j e c t
CROP Subproject
CROP has developed a data processing and delivery system for agribusiness
data products such as wheat crop forecasts derived from satellite
imagery. The demonstrator is implemented using Java sockets and embedded
Unix processes running data processing jobs. It was developed primarily
for Agrecon Pty Ltd, but with application to environmental and
agribusiness management.
Achievements & Milestones 1997/98
Progress toward the CROP milestones proceeded as planned during 1997/98.
The innovative CROP prototype architecture was completed on schedule by
June 1998 and gateways put in place for the processing of Agrecon time
series data. The CROP architecture is fully exploiting the OLDA computing
facility.
The CROP project has produced a prototype system for combining
distributed heterogeneous computing resources and data stores to create
high-performance processing pipelines. Features of the system are:
a written almost entirely in Java and supports most modern UNIX platforms;
a provides a transparent shell-like wrapper around non-Java components,
even though the processing stages themselves may be legacy code;
a handles issues such as efficient high-speed transfer, process
management and processor load balancing;
a supports a range of data sources and a data catalogue search facility
based on the ANZLIC Metadata recommendations.
Research Planned
- Although the CROP Project is completed, relevant work is planned under
the DGIS project. There is a proposal to extend and refine the CROP
architecture with Agrecon Pty Ltd, to process Agrecon historical (eight
year) archive of AVHRR data and establish procedures for ongoing
processing of this data.
- Incorporate the CSIRO CAPS processing software for AVHRR datasets, to
give the ability for Tend-to-endU AVHRR processing from raw data to a
variety of information products.
- Implement a distributed processing system between Adelaide/Canberra
that links the CROP architecture with the DHPC-I and DGIS prototypes.
- Incorporate data formats other than the AVHRR, such as GMS and vector
data, to allow the fast production of multi-format data products for
fields such as fire response and agricultural monitoring.
Milestones for 1998/99
CROP architecture review | August,
1998 |
CROP architecture update | November,
1998 |
Process Agrecon AVHRR data | October,
1998 |
Incorporate ACSys NT server | November, 1998 |
Incorporate CAPS software | December,
1998 |
Incorporate StorageTek system | February, 1999 |
Demonstrate graphical system monitor | February, 1999 |
Demonstrate multi-format pipeline
processing | March, 1999 |
|