August, 2003
Version 2.3 of the open source SABUL library is released. Version 1.0 of the open source UDT library is released.
April, 2003
The DSTP team introduces an algorithm for integrating distributed chemical libraries using a universal chemical key (UCK), which can be attached to chemical compounds. The work will be presented at CSB 2003.
March, 2003
Version 3.0 of the open source DSTP server is released.
November, 2002
At the SC 02 conference in Baltimore, the second phase of the Terra Wide Data Mining Testbed (TWDM) is launched. Phase 2 of the TWDM uses optical links and the StarLight optical interchange to connect three distributed clusters: one at StarLight in Chicago, one Surfnet in Amsterdam, and one at CANARIE in Ottawa.
At the SC 02 conference, Photonic Data Services enable the TWDM to mine global data at Gb/s speeds, with application level signaling for optical paths.
Data web applications with the US Census data are developed.
October, 2002
At the iGrid 2002 conference in Amsterdam, SABUL is used to set a conference record of 2.8 Gb/s for moving data across the Altantic. The data moved is between a three node cluster in Amsterdam to a three node cluster in Chicago at the StarLight facility. SABUL is able to transport data at about 940 Mb/s per node for an aggregate bandwidth of 2.8 G/bs.
Version 2.2 of the reference implementations for major DataSpace components are released:
- DSTP Server 2.2, an open source DSTP server, is released.
- HP-DSTP Server 2.2, an open source DSTP server for high performance networks, is released.
- SABUL Version 2.1 is released, a protocol for moving data over high performance networks is released.
September, 2002
Data web applications involving chemical compound data are developed.
The University of Illinois at Chicago's NCDM and Northwestern University's iCAIR announce a collaborative project to develop Photonic Data Services, which combine iCAIR's ODIN to set up optical paths, NCDM's SABUL for high performance data transport, and DSTP for data services.
NCSA integrates DSTP into its D2K software for data mining.
August, 2002
Open source development for Project DataSpace is moved to Source Forge.
InforSense integrates DSTP into its product Kensington.
May, 2002
DSTP is integrated with the W3C's Web Service Desciption Language (WSDL) and Universal Discription, Discovery, and Integration standards.
April, 2002
Version 2.1 of the DataSpace software is released.
November, 2001
The Terra Wide Data Mining (TWDM) Testbed is launched at the SC 2001 Conference in Denver. In Phase 1 of the TWDM, the University of Illinois, Dalhousie University in Halifax, and SARA (Stichting Academisch Rekencentrum) at the University of Amsterdam are linked by OC-12 (622 Mb/s) networks provided by Illinois' I-Wire, Canada's CA*net 3 network, and the Netherland's SURFnet4 network. The interchange point is provided by StarTap.
October, 2001
Version 2.0 of the DataSpace software is released.
September, 2001
The Global Discovery Network (GDN) is formed, a collaboration between Project DataSpace and Discovery Net, a data grid project started by Imperial College. Discovery Net is one of six e-science projects recently announced by the British Government.
August, 2001
The Data Mining Group (DMG) announces Version 2.0 of the Predictive Model Markup Language at the the Second Annual PMML Workshop held at the KDD-2001 Conference in San Francisco. DataSpace uses PMML to provide application and platform independent for the statistical and data mining models used in data space.
November, 2000
The 2000 Terabyte Challenge links five sites using OC-3 and OC-12 links at the SC 2000 Conference in Dallas during November 4-7, 2000.
September, 2000
Science Online reports on DataSpace.
August, 2000
The Chronicle of Higher Education reports about DataSpace.
On August 23, 2000, at the KDD-2000 Conference, the Data Mining Group (DMG) announced the release of Version 1.1 of the Predictive Model Markup Language (PMML). PMML is a mark up language for statistical models supported by products from IBM, Oracle, SPSS, NCR, Magnify, Angoss and others, and is used in the DataSpace Infrastructure.
Project DataSpace released version 1.0 of its DataSpace Transfer Protocol (DSTP) client and server software on August 14, 2000. The software can be downloaded from the DataSpace web site www.dataspace.net. It is written in Java and is open source.
During August 14-16, 2000, Project DataSpace set a new record for bandwidth utilization for data intensive computing at the Federal Interagency Planning Meeting in Mountain View, California. The DataSpace infrastructure moved data at over 250 Mb/s over several days over the Terabyte Challenge Testbed.
November, 1999
The 1999 Terabyte Challenge links 12 sites around the world using high performance networks for a demonstration of distributed and high performance data mining over DataSpace at the SC 1999 Conference in Portland during November 13-19, 1999.
September, 1999
Project DataSpace begins with a $2M four year grant from the National Science Foundation.