By Radha Basu
Sept 10, 2003

Store and retrieve data from remote servers across the world - in seconds. Possible? Yes, researchers in Singapore and the United States are working on it.

Researchers at Nanyang Technological University (NTU) are collaborating with their counterparts in the United States to develop software tools that will allow users to store data on remote computers anywhere in the world and retrieve it at will.

The project may go a long way in solving storage problems arising from the global data deluge, sparked in part by recent advancements in life sciences research.

Known as 'logistical networking', the technology does not involve storing data in conventional databases. Instead, the IBP server software can transform any computer on the Internet into a 'depot' for storing data.

The data is stored in the hard disks of individual computers from users who wish to install the IBP tool and make their computer a depot.

However, at present, data contained in IBP depots is difficult to retrieve.

The NTU team, led by Associate Professor Francis Lee Bu Sung of the School of Computer Engineering (SCE), is developing and testing data retrieval and management tools that will ease the process.

With Prof Lee overseas, his colleague, Associate Professor Yeo Chai Kiat, who is also a collaborator on the project, elaborated on its significance.

Storage-related research and development, Prof Yeo said, has gained considerable importance in view of the recent deluge in digital data worldwide.

'According to an article in Bio.IT World, the flood of complex genomic data alone is expected to grow to about 25 terabytes (TB) in the next 12 months and to 100TB the following year,' she said.

Twenty-five TB is the equivalent of about 6.5 million four-megabyte MP3 files.

Building and maintaining databases to store so much information would be hugely expensive, said Prof Yeo. 'Networked storage allows the aggregation of resources of individual computers, making it cheaper as well as more open and accessible.'

The IBP network is already up and running. According to Prof Yeo, there are 234 public IBP servers in 19 countries.

When a user stores a file on the shared network, the network returns a software 'key', known as the ExNode, which must be used to retrieve the data.

At present, users wishing to retrieve the data have to manually key in complex commands to do so, making the data inaccessible to the general public who are unfamiliar with the commands.

The NTU researchers have developed a software tool called the ExNode File Warehouse (EFW), which simplifies the data access process by providing pointers to the physical location of the data.

The researchers have also integrated the IBP software with the Globus Replica Catalog (GRC), an existing software tool that helps locate copies of a particular file on a distributed storage network.