With the growing scale of the computer storage systems, the likelihood of multi-disk failures happening in the storage systems has increased dramatically. Based on a thorough analysis on the fault-tolerance capability on various existing storage systems, we propose a new hierarchical, highly reliable, multi-disk fault-tolerant storage system architecture: High Availability Object Storage System (HAOSS). In the HAOSS, each object has an attribute field for reliability level, which can be set by the user according to the importance of data. Higher reliability level corresponds to better data survivability in case of multi-device failure. The HAOSS is composed of two layers: the upper-layer and the lower-layer. The upper-layer achieves the high availability by storing multiple replicas for each storage object in a set of storage devices. The individual replicas can service the I/O requests in parallel so as to obtain high performance. The lower-layer deploys RAID5, RAID6 or RAID_Blaum coding schemes to tolerate multi-disk failures. In addition, the disk utilization rate of RAID_Blaum is higher than that of multiple replicas, and it can be further improved by growing the RAID group size. These advantages come at the price of more complicated fault-tolerant coding schemes, which involve a large amount of calculation for encoding and cause an adverse impact on the I/O performance, especially on the write performance. Results from both our internal experiments and third-party independent tests have shown that HAOSS servers have better multi-disk- failure tolerance than existing similar products. In a 1000Mb Ethernet interconnection environment, with a request block size of 1024KB, the sequential read performance for a HAOSS server reaches 104MB/s, which is very close to the theoretical maximum effective bandwidth of Ethernet networks. The HAOSS offers a complete storage solution for high availability applications without the compromises that today's storage systems require in either performance or fault-tolerance.
In this paper, we present the architecture and implementation of a virtual network computers' (VNC) optical storage
virtualization scheme called VOSV. Its task is to manage the mapping of virtual optical storage to physical optical
storage, a technique known as optical storage virtualization. The design of VOSV aims at the optical storage resources of
different clients and servers that have high read-sharing patterns. VOSV uses several schemes such as a two-level Cache
mechanism, a VNC server embedded module and the iSCSI protocols to improve the performance. The results measured
on the prototype are encouraging, and indicating that VOSV provides the high I/O performance.
Storage demands are growing rapidly with the increasing usages of multimedia, stream casting, and large scale database. High-performance RAID storage is a critical component for many large-scale data-intensive applica¬tions. The goal of our project is to implement high performance external RAID controller based on Intel IOP321. The software design of the RAID controller is introduced, which consists of six functional modules: SCSI Target, Cache Management, RAID Kernel, I/O Schedule, SCSI Initiator and Global Configuration. A PC architecture RAID controller with Fibre Channel interface to host has been implemented to test algorithms and evaluate performances. We also present hardware scheme of the controller based on Intel IOP321 processor.
CD mirror system can separate the reading from storing of CD information, enabling user to share CD resource with higher speed. Existing remote CD mirror system is generally based on NAS mode. NAS-based device accesses the data at file level and the data must pass through the file system layer, increasing the system burden; iSCSI system encapsulates SCSI commands over IP, which just corresponds with the transport mechanism of remote CD mirror system. Therefore, in this paper we have implemented a new remote CD mirror system by introducing iSCSI protocol and third-party transferring mode. With third-party transferring mode data information and control information are transferred independently, the data information is transferred by iSCSI protocol, and control information is transferred by TCP/IP protocol, so as to increase the network transfer rate. The result of experiment has showed that the CD mirror system can achieve higher data transfer rate and support more concurrent sessions relative to existing systems, and have better scalability using third-party transferring mode.
With the increasing of CD data in internet, CD mirror server has become the new technology. Considering the performance requirement of the traditional CD mirror server, we present a novel high performance VCL (Virtual CD Library) server. What makes VCL server superior is the two patented technologies: a new caching architecture and an efficient network protocol specifically tailored to VCL applications. VCL server is built based on an innovative caching technology. It employs a two-level cache structure on both a client side and the server side. Instead of using existing network and file protocols such as SMB/CIFS etc that are generally used by existing CD server, we have developed a set of new protocols specifically suitable to VCL environment. The new protocol is a native VCL protocol built directly on TCP/IP protocol. VCL protocol optimizes data transfer performance for block level data as opposed to file system level data. The advantage of using block level native protocol is reduced network-bandwidth requirement to transfer same amount of data as compared to file system level protocol. Our experiment and independent testing have shown that VCL servers allow much more number of concurrent users than existing products. For very high resolution DVD videos, VCL with 100Mbps NIC supports over 10 concurrent users viewing the same or different videos simultaneously. For VCD videos, the same VCL can support over 65 concurrent users viewing videos simultaneously. For data CDs, the VCL can support over 500 concurrent data stream users.