FTRAID: A Fat-tree Based Parallel Storage Architecture for Very Large Disk Array

Traditional disk arrays have a centralized architecture, with a single controller through which all requests flow. Such a controller is a single point of failure, and its performance limits the maximum number of disks to which the array can scale. Fat-trees are well-adopted as the topologies of interconnection networks because of many nice properties they have. We propose a novel fat-tree based parallel RAID architecture - FTRAID. The unit of deployment in FTRAID is a storage node, a small rack-mounted storage appliance built from commodity components including CPU, RAM, NIC and disks. We also establish stochastic Petri net (SPN) model of storage node to analyze its utilization. In particular we have designed and implemented a seven node FTRAID prototype to demonstrate its feasibility. Numerical results using popular benchmark have shown good performance gain.