SAN FRANCISCO, CA -- Oracle Corp. (Nasdaq: ORCL) today announced that unique capabilities in Oracle9i Database are helping CERN, the European Organization for Nuclear Research, to push back the boundaries of human knowledge. CERN has embarked upon a project, the Large Hadron Collider (LHC) project, which will generate petabytes of data -- an amount well beyond the capability of any relational database technology today. CERN is trailblazing a new route in data management and analysis using Oracle9i Real Application Cluster technology. The 27-km CERN (LHC) is a particle accelerator project to analyze the nature of matter through observation of very high-speed collisions between subatomic particles. However, the process of observing the collisions will produce unimaginably huge quantities of raw data -- a collision will happen every 25 nanoseconds, but only one in a billion of these will be of any interest at all, and just one in a trillion is likely to be of significant interest. Searching for a needle in a haystack is a legendarily difficult job; in the analysis of the LHC output, CERN is facing the prospect of finding a needle in 20 million haystacks. By 2006, the LHC will accumulate over ten million gigabytes (ten petabytes) of data, equivalent to the contents of a 40-kilometer high stack of CD-ROMs, each year of its operation. A thousand times more computing power will be needed than is available to CERN today. CERN's approach to the problem of storing petabytes of information is to divide and conquer -- split the data categories and results from different experiments. CERN sees Oracle's unique Real Application Cluster technology as a key to enabling them to meet their LHC data storage requirements on commodity hardware, thus relying on mainstream technologies and low risk solutions. CERN can envisage a path from today's databases of 1 - 3 terabyte (TB) based on a dual-processor Intel/Linux machine to 100TB based upon a cluster of ten machines. CERN's approach is: first, to demonstrate that 100TB is viable, using Oracle9i Real Application Clusters on an Intel/Linux platform; and secondly to show that such components can be used as building blocks to solve a multi-pedabyte problem. This demands near-linear scalability from any clustering technology. CERN is encouraged by its experiences so far with Oracle's Real Application Clusters. ``CERN saw the need for clustering capability on commodity hardware a long time ago,'' said Jamie Shiers, Database Group Manager at CERN. ``With Oracle9i Real Application Clusters, Oracle provides that capability in its mainstream technology, thus giving CERN a realistic, low risk route to coping with the staggering storage requirements of the LHC.'' ``Oracle has a very close relationship with CERN which has enabled us to work towards supporting the comprehensive information handling capabilities needed for the success of the LHC project,'' said Sergio Giacoletto, executive vice president, Oracle Europe, Middle East and Africa (EMEA). ``CERN is one of the world's most pre-eminent scientific organizations, and Oracle is delighted to be part of the world's most exciting database project.'' Oracle began work with CERN in 1999, when it demonstrated a 37TB proof-of-concept database. This demonstration encouraged CERN to team up with Oracle to explore the paths towards being able to handle the phenomenal data requirements involved in its LHC project.