|
ABSTRACTS
Network direct attached storage (NDAS) is a network storage architecture that allows direct attachment of existing ATA/ATAPI devices to Ethernet without a separate server. Unlike other architectures such as NAS, SAN, and USB mass storage, no server computer intervenes between the storage and the client hosts.
We describe an NDAS disk controller (NDC) amenable to lowcost single-chip implementation that processes a
simplified L3/L4 protocol and converts commands between ATA/ATAPI and Ethernet, while the remaining
complex tasks are performed by remote hosts. Unlike NAS architectures that use TCP/IP, NDAS uses a
TCP-like lean protocol that lends itself well to high-performance hardware realization. Thanks to the simple NDAS architecture and protocol, an NDC implemented on a single 4 mm X 4 mm chip in 0.18 μm CMOS technology achieves a maximum throughput of 55 Mbytes/s on Gigabit Ethernet, which is comparable to that of a highperformance disk locally attached to a host computer.
INTRODUCTION
Today, digital multimedia devices such as set-top boxes, DVD players, and HDTVs have become very popular; thus, large storage for high-capacity multimedia data has played an important role even in the home.
The exploration of a suitable efficient architecture for high-capacity storage targeted at home applications is underway. Although many high-capacity storage devices such as hard disk drives and DVD drives are widely used in a PC environment, mere high capacity is not enough in today's home environments. Advances in network technology have expanded the convenience and efficiency of the network from the PC environment into the home; the era of the home network is coming. To keep pace with this trend, storage devices also should adapt to the home network. True advantages of the home network are realized when storage devices dedicated solely to one application are offered for sharing among other systems for many different applications through the network. However, it is difficult to share storage efficiently and conveniently because many existing storage devices are targeted for use solely inside application devices.
For example, to play an HDTV program on a computer with no HDTV tuner card, the program is first demodulated in a set-top box and saved in the hard disk drive inside the set-top box. Then it must be copied to the disk in the
computer to be played on the computer display.
Copying is very inconvenient since it is time consuming and wastes storage capacity with duplicated contents. The network storage addresses this problem by allowing access to its contents symmetrically by participating application devices. The set-top box would directly save the recorded program in the network storage, and it would be directly accessed and played on the computer for viewing. Therefore, the network interface to such storage and application devices needs to be adequately designed for efficient use by multimedia devices at home. For successful home network storage devices, two requirements, cost and performance, should be satisfied. Consumer electronics devices are sold at very low prices. Furthermore, to cope
with rapid growth of available contents, additional devices for expanding capacity must be supplemented frequently. Therefore, low cost is a primary requirement for home network storage. Also, it should offer enough performance to support multimedia applications requiring high bandwidth. In our opinion, the performance of network storage must be limited only by the performance of the storage itself. That is, the network must perform like an I/O, and the storage should be able to yield performance comparable to that of locally connected storage. Although there have been many network storage devices, they are not well suited to applications for home networks. While in some research [1, 2] network file servers have been....
Full Text PDF
 |