An experimental platform to study massively multi-core processor support for active data storage services

点击翻译按钮获取中文摘要

This application seeks funding for an experimental platform to support the exploration of the following three hypotheses. First, we believe that active storage techniques can be efficiently and transparently integrated with massively multi-core compute resources to accelerate the execution of data-intensive applications in fields as diverse as medical image analysis and bioinformatics. The second hypothesis is that the computational power of these massively multi-core processors can be harnessed not only to accelerate specialized applications, but also to accelerate, enhance, or reduce the cost of generic storage systems. Inexpensive computation can be used to support core techniques that impact storage systems' reliability, security, or performance by using, for example, erasure coding, encryption, or on-the-fly data compression. Finally, the third hypothesis is that a specific class of commodity multi-core processors, namely graphical processing units (GPUs), can be effectively used in the above context.