A scalable system can be built when the storage devices such as HDDs are provided to the system via network and the parallel data paths between the hosts and the storage devices are provided independent of others. We built a parallel network direct attached storage architecture that offers two requirements for scalable storage system; (1) unlimited storage expansion via network, and (2) parallel direct data paths between the hosts and the HDDs. Our prototype was verified to have practically unlimited storage capacity up to hundreds of HDDs with its total aggregated I/O bandwidth proportional to the number of HDDs attached to the network. From the experiments, actual performance of individual HDDs in our prototype was verified to be about 70...
The ideal storage system is always available and incrementally expandable. Existing storage systems ...
PIOUS is a parallel file system architecture that provides cost-effective, scalable bandwidth in a n...
Ever more powerful computers and rapidly enlarging data sets require unprecedented levels of data st...
We have designed, built, and analyzed a distributed parallel storage system that will supply image s...
Emerging applications such as data warehousing, multimedia content distribution, electronic commerce...
Abstract: Streaming media is pervasive on the Internet now and continues to grow rapidly. The dramat...
Incorporating scalability into the storage of multimedia should be a fundamental design issue that w...
Network attached storage systems must provide highly available access to data while maintaining high...
Large data stores are pushing the limits of modern technology. Parallel file systems provide high I/...
We present a new approach to create large-scale cost-effective mass storage systems using high-capac...
Distributed storage systems running on clusters of commodity hardware are challenged by the ever-gro...
The storage and bandwidth requirements of digital video and audio exceed those that can be supported...
We present an I/O architecture, called Swift, that addresses the problem of data-rate mismatches bet...
We describe the architecture, implementation, use, and potential use of a scale, high-performance, d...
The current demand on high-performance I/O is higher than ever and cannot afford to allow any unnece...
The ideal storage system is always available and incrementally expandable. Existing storage systems ...
PIOUS is a parallel file system architecture that provides cost-effective, scalable bandwidth in a n...
Ever more powerful computers and rapidly enlarging data sets require unprecedented levels of data st...
We have designed, built, and analyzed a distributed parallel storage system that will supply image s...
Emerging applications such as data warehousing, multimedia content distribution, electronic commerce...
Abstract: Streaming media is pervasive on the Internet now and continues to grow rapidly. The dramat...
Incorporating scalability into the storage of multimedia should be a fundamental design issue that w...
Network attached storage systems must provide highly available access to data while maintaining high...
Large data stores are pushing the limits of modern technology. Parallel file systems provide high I/...
We present a new approach to create large-scale cost-effective mass storage systems using high-capac...
Distributed storage systems running on clusters of commodity hardware are challenged by the ever-gro...
The storage and bandwidth requirements of digital video and audio exceed those that can be supported...
We present an I/O architecture, called Swift, that addresses the problem of data-rate mismatches bet...
We describe the architecture, implementation, use, and potential use of a scale, high-performance, d...
The current demand on high-performance I/O is higher than ever and cannot afford to allow any unnece...
The ideal storage system is always available and incrementally expandable. Existing storage systems ...
PIOUS is a parallel file system architecture that provides cost-effective, scalable bandwidth in a n...
Ever more powerful computers and rapidly enlarging data sets require unprecedented levels of data st...