Electrical computers and digital data processing systems: input/ – Input/output data processing
Reexamination Certificate
2000-02-22
2004-08-31
Gaffin, Jeffrey (Department: 2182)
Electrical computers and digital data processing systems: input/
Input/output data processing
C710S033000, C710S036000, C710S074000
Reexamination Certificate
active
06785742
ABSTRACT:
BACKGROUND
1. Technical Field
The present invention relates generally to the field of computer networks, and more particularly, to the management of storage networks.
2. Background of the Invention
As a result of continuous advances in technology, particularly in the area of computer networking, the last decade has seen an explosion in the volume of data that is being captured, processed, stored and manipulated in business environments. This data explosion has fueled an increase in demand for data storage capacity. The challenges presented by this increased demand for data storage further are amplified by the fact that it is not just the existence of data alone that provides value, but rather the manner in which the data is stored, accessed, and managed, that creates a competitive advantage.
Increased reliance on applications ranging from business intelligence and decision support, data warehousing and data mining of large databases, disaster tolerance and recovery, enterprise software, and imaging and graphics have all contributed to this trend. In addition, the development of Internet-based business operations and electronic-commerce specifically has intensified the demand placed on data centers. Customer interactions over the Internet have increased operational focus on the performance, scalability, management and flexibility of systems that use business-critical data. This dependence on data for fundamental business processes by employees, customers and suppliers has greatly increased the number of input and output (“I/O”) transactions required of computer storage systems and servers. Thus, it is important that data storage be viewed as a centralized, managed resource that is available and capable of expansion without sacrificing access or performance.
Despite the increased attention and resources devoted to data storage requirements, the technical capabilities of data storage systems have not kept pace with increasing data management demands and with the advancements in other networking technologies. In the 1980s, the near ubiquity of personal computers (“PCs”), workstations and servers required broader connectivity, resulting in the development of local and wide area networks to support messaging between computer systems. The data used by computers and servers connected to local and wide area networks typically are located on computer storage systems and servers, which store, process and manipulate data. The adoption of high speed messaging technologies such as gigabit Ethernet and asynchronous transfer mode (“ATM”), increased local and wide area network transmission speeds by more than 1,000 times during the 1990s. However, storage-to-server data transmission speeds increased by less than ten times during this period, creating a bottleneck between the local or wide area network and business-critical storage systems and servers.
One conventional solution with regard to data storage and retrieval is Network Attached Storage (“NAS”), which commonly utilizes a Network File System (“NFS”). As illustrated in
FIG. 1
, such a system
100
includes a host device
105
, a file server
110
, a Small Computer Systems Interface (“SCSI”) bus
116
and a plurality of storage devices
122
,
124
,
126
and
130
. Device
130
further includes a logical controller unit
131
and logical units
132
and
134
. A logical unit is a target-resident entity that implements a device model and executes SCSI commands originated by an initiator. An initiator is a device that initiates communication with, and transmits commands to other devices.
Host device
105
and file server
110
can be conventional personal computers from, for instance, IBM Corporation of Armonk, N.Y., or high-end computer workstations from, for instance, Sun Microsystems, Inc. of Palo Alto, Calif. Storage devices
122
,
124
,
126
can include, for instance, IBM's Ultrastar 18LZX. Device
130
can be a Redundant Array of Inexpensive Disks (“RAID”) system, such as a GigaRAID/AA from nStor Corporation of San Diego, Calif. with the logical units
132
and
134
representing individual hard disks, such as Barracuda 18LP-18.2GB-ST318275LW/LC of Seagate of Scotts Valley, Calif.
File server
110
is coupled via SCSI bus
116
to storage devices
122
,
124
,
126
and
130
. The SCSI standard was adopted as the open I/O interface standard for storage-to-server connections in the 1980s. When host device
105
attempts to access any of these storage devices
122
,
124
,
126
, or
130
, host device
105
must access them via file server
110
. In particular, host device
105
must establish a connection
103
with file server
110
, and then retrieve information stored within storage devices
130
,
122
,
124
, and
126
by communicating with file server
110
. Such an NAS system
100
, however, is inherently inefficient because host device
105
must access each device via file server
10
. By requiring access to storage devices
130
,
122
,
124
and
126
to be routed through file server
110
, bandwidth constraints directly related to file server
110
will exist. In addition, with such a system
100
dedicated to a specific application, the system
100
is not scalable. There also is a single point of failure in that if file server
110
is not available, the host device
105
will not be able to access any of the storage devices
122
,
124
,
126
and
130
.
An alternative Storage Area Network (“SAN”) architecture, which attempts to solve the limitations associated with NAS systems described above, is a fabric-based SAN. A fabric typically is constructed with one or more routing devices, such as switches, and each storage device (or group of storage devices, for example, in the case of a loop-based architecture) is coupled to the fabric. Generally, devices coupled to the fabric are capable of communicating with every other device coupled to the fabric. This eliminates the bottleneck created in NAS, where every device coupled to a file server
110
has to be accessed via that single file server
110
. Further, the scalability limitations of the NAS system also are avoided by fabric-based SANs. In addition, since every device does not have to be accessed via a single file server, there is no single point of failure in the fabric-based SANs.
One implementation of a fabric-based SAN is Fibre Channel, which is an American National Standards Institute (“ANSI”) high-speed, high-performance storage-to-server and server-to-server interconnect protocol. This application relates to “Fibre Channel Protocol for SCSI (FCP), Rev 012, May 30, 1995” and “Fibre Channel Physical and Signalling Interface-3, Rev 9.4, Nov. 5 1997,” both published by American National Standard for Information Technology, which are each incorporated herein by reference in their entirety.
Since Fibre Channel can support large data block transfers at gigabit speeds, Fibre Channel is well suited for data transfers between storage systems and servers. Fibre Channel also supports multiple protocols such as SCSI and Internet Protocol (“IP”). Furthermore, Fibre Channel provides transmission reliability with guaranteed delivery and transmission distances of up to 10 kilometers. Fibre Channel complements and supports advancements in local and wide area network technologies, such as gigabit Ethernet and ATM, which directly cannot effectively transfer large blocks of data.
Fibre Channel-based SANs deliver centrally managed storage through high availability architectures, managing systems, and exploiting effective storage methodologies like RAID and mirroring. The Fibre Channel Protocol for SCSI (“FCP”) relies upon the Small Computer System Interface (“SCSI”) communication scheme to communicate between devices, which are coupled to the fabric. One version of SCSI on which FCP can rely is SCSI-3. By incorporating SCSI, a protocol already native to many devices, into the fabric-based scheme, Fibre Channel is able to provide high-performance (e.g. gigabit per second data delivery and gigabit per second backup and recovery), a highly available storage network, and conti
Ramsay Paul
Teow Kha Sin
Brocade Communications Systems, Inc.
Farooq Mohammad O.
Gaffin Jeffrey
Wong, Cabello, Lutsch, Rutherford & Brucculeri L.L.P.
LandOfFree
SCSI enclosure services does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with SCSI enclosure services, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and SCSI enclosure services will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3272187