Electrical computers and digital data processing systems: input/ – Input/output data processing – Input/output access regulation
Reexamination Certificate
2001-08-10
2004-12-14
Gaffin, Jeffrey (Department: 2182)
Electrical computers and digital data processing systems: input/
Input/output data processing
Input/output access regulation
C710S038000, C710S008000
Reexamination Certificate
active
06832272
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATION
This application is related to and claims priority from Japanese Application No. 2001-177180, filed Jun. 12, 2001.
BACKGROUND OF THE INVENTION
The present invention relates to a storage system and more particularly to a clustering storage system capable of operating a plurality of storage system unit as a single storage system.
As a prior art of a storage system, a technique disclosed in, for example, JP-A-11-167521 is known. In this prior art, interfaces (host adapters CHA) to high-ranking apparatuses (host computers) which use the storage system, interfaces (disk adapters DKA) to storage devices such as magnetic disk devices within the storage system, cache memories (CACHE), and shared memories (SM) are connected with one another in a common path manner.
FIG. 15
is a block diagram illustrating a configuration example of such a storage system of the prior art. Referring now to
FIG. 15
, the prior art is described. In
FIG. 15
, numeral
100
denotes the storage system,
110
host adapters (CHA),
120
disk adapters (DKA),
130
cache memories (CACHE),
140
shared memories (SM),
150
storage devices (HDD),
160
common paths, and
170
connection lines.
The storage system
100
of the prior art includes, as shown in
FIG. 15
, the host adapters
110
, the disk adapters
120
, the cache memories
130
, the shared memories
140
, the storage devices
150
, the common paths
160
and the connection lines
170
. The host adapters
110
, the disk adapters
120
, the cache memories
130
and the shared memories
140
are connected with one another by means of the common paths. The common paths
160
are duplicated in case a failure of the common path
160
occurs. Each of the storage devices
150
is connected to two disk adapters
120
by means of different connection lines
170
so that each storage device
150
can be used even upon failure of one disk adapter
120
or one connection line
170
.
The host adapters
110
control data transfer between host computers not shown and the cache memories
130
. The disk adapters
120
control data transfer between the cache memories
130
and the storage devices
150
. The cache memories
130
are memories for temporarily storing data received from the host computers and data read out from the storage devices
150
. The shared memories
140
are memories shared by all of the host adapters
110
and disk adapters
120
. Further, although not shown, a maintenance terminal (SVP) for performing setting, monitoring, maintenance and the like of the storage system
100
is connected to all of the host adapters
110
and disk adapters
120
by means of a hot line.
The case where the system configuration of the storage system
100
having the above configuration is expanded is now considered. In such a case, heretofore, constituent elements such as the host adapters
110
, the disk adapters
120
, the cache memories
130
, the shared memories
140
and the storage devices
150
are newly added within the storage system
100
. For example, when the number of host computers is increased, a host adapter
110
is newly connected to the common paths
160
. Further, when the memory capacity of the storage system
100
is increased, a storage device
150
is added within the storage system
100
or a disk adapter
120
is newly connected to the common paths
160
and a storage device
150
is added.
SUMMARY OF THE INVENTION
In the above prior art, in order to expand the storage system, the host adapter, the disk adapter, the cache memory, the shared memory, the storage device and the like are increased as constituent elements of the storage system. Accordingly, the expansibility of the storage system is limited to the maximum mountable number of constituent elements in the storage system. Consequently, in the prior art, in order to satisfy a request for a large-scale storage system, the storage system having the increased maximum mountable number of constituent elements must be adopted from the introduction time of the storage system. As a result, when a small-scale storage system is not expanded as it is, there occurs a problem relative to increased cost and installation area and the like.
As a method of solving the above problem, a clustering storage system is considered. The clustering storage system includes a plurality of storage system units connected with one another and each unit constituted by the aforementioned storage system and can operate as a single storage system as viewed from the host computers. A unit of system for constituting the clustering storage system is hereinafter named a storage system unit. When a small-scale storage system is required, a clustering storage system can be constituted by a small number of storage system units, while when the scale of the storage system is expanded, the expansion can be solved by increasing the storage system units in the clustering storage system. In this manner, the clustering storage system can applicable to from a small-scale system to a large-scale system and can operate as a single storage system as viewed from the host computers. Accordingly, it is advantageous that management thereof is easy.
In the clustering storage system, however, there occurs a case where a storage system unit for receiving an access command from a host computer and a storage system unit for storing data to be accessed are different. Consequently, there is a problem that data transfer between the storage system units is required and the access performance is degraded.
It is an object of the present invention to solve the above problem in the conventional clustering storage system by providing a clustering storage system capable of improving the access performance.
According to the present invention, the above object can be achieved by a clustering storage system capable of operating a plurality of storage system units as a single storage system, comprising means for collecting access information within the storage system units and between the storage system units, means for showing the collected access information to a maintenance person, means for receiving at least one of a remapping command of port and a remapping command of data from the maintenance person, and at least one of means for remapping ports in the clustering storage system and means for remapping data between the storage system units in accordance with the command.
The access information contains, for example, a data transfer amount or time necessary for data transfer in a unit time within the storage system units and between the storage system units. The access information is collected for a unit of combination of a logical port receiving an access command and a logical volume constituting a unit for management of data. By collecting the access information, it is understood that an access by a host computer from which logical port to which logical volume requires a lot of data transfer between the storage system units. The maintenance person can refer to the access information by means of the maintenance terminal. Accordingly, the maintenance person can recognize a combination of the logical port and the logical volume having a high frequency of data transfer between the storage system units and study remapping of the logical port and the logical volume. Further, the clustering storage system can perform remapping of the ports and the logical volumes between the storage system units in response to a command from the maintenance person to make lower the frequency of data transfer between the storage system units. Thus, the access performance of the clustering storage system can be improved.
Further, according to the present invention, the clustering storage system or the maintenance terminal in the clustering storage system can be provided with the function of studying remapping of the ports and the logical volumes on the basis of the access information to thereby reduce burden on the maintenance person.
In addition, according to the present invention, the clustering storage system may be provided with the f
Arai Kouji
Arakawa Hiroshi
Hirakawa Yuusuke
Oeda Takashi
Gaffin Jeffrey
Hitachi , Ltd.
Martinez David
Townsend and Townsend / and Crew LLP
LandOfFree
Clustering storage system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Clustering storage system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Clustering storage system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3300510