JPH0784855A - Selection system for file server - Google Patents

Abstract

PURPOSE:To provide the selection system for file servers which can select a file server most suitably for access time, cost, etc., from respective computers. CONSTITUTION:Each computer is provided with an access frequency measuring part 13 which specifies opposite-side computers which have accessed an individual file and measures the logical distance to them and the access frequencies, an access frequency management table 12 which finds the access frequencies by multiplying the logical distances to itself and the access frequencies as to the individual opposite-side computers and the access frequency index obtained by adding them on the basis of information measured by itself and other computers, a file transfer part 15 which compares the access frequency indexes found by itself and other computers according to specific algorithm and determines one computer as a new file server, and a file transfer part 15 which transfers a file substance to the computer determined as the file server.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention selects which computer stores a file system of a computer system, particularly a network file system accessed via a network, that is, which computer serves as a file server. More specifically, the present invention relates to a system that selects a more appropriate file server according to the usage status of each file.

[0002]

2. Description of the Related Art As a network file system for workstations, for example, Sun Micro Systems
NFS or AFS of Transarc, Inc. are widely used. NF
S stores the file entity (the data that constitutes the file itself) in hardware called a file server, that is, in a large-capacity magnetic disk device of a computer, and stores the data portion required by each of a plurality of client workstations. It is a mechanism to share file data by making a copy accessible via LAN.

In such a method, the file entity is always located in the file server, and the file entity may not move to other hardware in the network although the file size may increase or decrease. However, even if files are located on a file server from a workstation via an inter-network connection device called a router or a bridge even within the same premises, or even remotely via a high-speed network (WAN) such as AFS. When accessing from the ground, there is a risk that the access time will significantly increase and it will not be within the time limit required for the same LAN.

[0004]

Conventionally, particularly in AFS, the whole or part of a file is transferred from a file server to a hard disk (local disk) as a storage medium of a client workstation in advance when necessary, and the local disk is used. Is used as a cache memory and data is used. According to this method, it takes a certain amount of time to cache the file on the local disk, so that the user is forced to wait, and some kind of abandonment occurs. In any case, however, file access via the network has the disadvantage that the file stays on the file server where it was originally generated, considering the time and economic cost of accessing the network. Is very large.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a file server selection system capable of optimally selecting a file server in terms of access time from each computer, cost, and the like. And

[0006]

A file server selection system according to the present invention is a system for selecting any one of a plurality of computers connected via a network as a file server where a file entity should be located. When each of the plurality of computers is a file server, each computer identifies the partner computer that has accessed each file, and the logical distance from the identified partner computer. Based on the information measured by the access frequency measuring means for measuring the number of accesses and the access frequency measuring means or the information measured by the access frequency measuring means of other computers The access frequency multiplied by the logical distance and the number of accesses Therefore, by comparing the access frequency recording means for obtaining the access frequency index obtained by adding them with the access frequency index obtained by the access frequency recording means of itself and other computers according to a predetermined algorithm, The file entity position determining means for determining any one of the above as a file server where the target file should be located, and the file transfer means for transferring the file entity to the computer determined as a new file server.

[0007]

In the file server selection system of the present invention,
The computer that is the file server measures the access frequency to the file in which the file entity is located by its own access frequency measuring means and obtains the access frequency index by its own access frequency recording means. Receiving the information measured by the access frequency measuring means of the computer, and obtaining the access frequency by multiplying the logical distance between itself and the access frequency by the access frequency recording means of itself, and further the access frequency recording means of itself. Then, the access frequency index obtained by adding them is calculated, and the computer that should be the new file server is determined by the file entity position determining means of each computer, and the computer that was the file server until then is newly determined as the file server. To transfer the file entity by the own file transfer means New to the computer.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the drawings showing the embodiments thereof. FIG. 5 is a schematic diagram showing an example of a hardware environment to which the file server selection system according to the present invention is applied. In FIG. 5, reference numerals L1 to L5 are LAN (Lo
cal Area Network: local area network)
FS1 and FS2 are computers that can be file servers, H1 to H6 are host computers, R1 to R5 are routers, and W1 is a WAN (Wide Area Network).

LAN L1 has host computers H1, H2 and
H3, LAN L3 is connected to host computer H5, LAN L4 is connected to host computer H4 and computer FS2 which can be a file server, LAN L5 is connected to host computer H6, WAN W1 is connected to a file server. Possible computer FS2 is connected. LAN L1 goes through router R1, LAN L3 goes through router R2, LAN L4
Is connected to LAN L2 via router R3, WAN W1 is connected to LAN L2 via router R4, and LAN L5 is connected via router R5.
It is connected to WAN W1.

With the file server selection system according to the present invention, the file to be operated is the file A, the computer in which the substance of this file A is currently located is the file server FS2, and it can be located. Let the computer be the file server FS2. The file server FS2 where the files are currently located is also called the home site. In FIG. 5, it is assumed that the file A is located in the file server FS2. That is, the file server FS2 is the current home site.

Computers that need and can access the file A are called host computers. In FIG. 5, six computers H1 to H6 are arranged as described above. However, although some computers do not access the file A, they are omitted in FIG.

FIG. 1 is a block diagram showing the configuration of the main parts of the file servers FS1 and FS2. Both file servers FS
1, FS2 is basically a computer system equipped with a large-scale storage device such as a magnetic disk. Each file server FS1, FS2 is provided with a file system management unit 11, an access frequency management table 12, an access frequency measurement unit 13, a file home site determination unit 14, a file transfer unit 15, a communication unit 16, and the like. Although not shown in FIG. 1, a large-capacity storage device such as a hard disk is provided for its original function as a file server.

The file system management unit 11 controls the entire file servers FS1 and FS2. The access frequency measuring unit 13 functions as an access frequency measuring means, and for each target file, the host computer that accessed it.
Information identifying H1-H6 and its host computer H1-
Logical distance to H6 (via LAN, W
The value obtained by multiplying the number of ANs by a predetermined coefficient) and the access frequency (number of times) per unit time are measured. However, the logical distance referred to here is, for example, a packet hop count used in the TPC / IP communication protocol. The packet hop count is a mechanism for adding a count number each time a packet passes through a router, which is an inter-network relay device, and erasing packets passing the network unnecessarily.

Further, the access frequency measuring unit 13 expresses the cost when the file A is accessed from the host computer via the WAN (high-frequency network) whose transmission speed is slower than that of the LAN. Therefore, a high coefficient different from the logical distance in the case of access via LAN is used. In the present embodiment, the coefficient of the logical distance when access is made via one LAN is set to "1", and one WAN is set.
The coefficient of the logical distance is "5" when the access is performed via.

The access frequency management table 12 functions as an access frequency recording means, and records the result of measurement within the unit time by the access frequency measuring unit 13 as described above. Figure 2
FIG. 6 is a schematic diagram showing an example of recorded contents of the access frequency management table 12. As shown in Fig.2, in each file server FS1 and FS2, for all the host computers that have been accessed, information that identifies the host computer, logical distance from each host computer, from each host computer The number of times of access is recorded in each access frequency management table 12. After that, the logical distance of the access frequency management table 12 and the access frequency are multiplied to calculate the index of the access frequency for each host computer, and finally the overall access frequency index is obtained.

The file home site determining unit 14 functions as a file entity position determining means and determines whether or not to change the file home site, which will be described in detail later. The file transfer unit 15 functions as a file transfer unit, and when the home site is changed to another file server, the file transfer from the file server provided with the file to another home site newly becomes the home site. Control transfer. The communication unit 16 controls various types of communication with other file servers.

The file home site determining unit 14 determines whether or not to change the home site as follows. When the access frequency index is obtained in the access frequency management table 12 of the file server FS2, which is the current home site, as described above, the information identifying the other host computer among them, the logical distance between each host computer, The number of times of access from the host computer is transferred by the communication unit 16 to the other file server (file server FS1 in the example of FIG. 5) of the access frequency management table 12 via LAN and WAN.

The file server FS1 that is not currently the home site is the file server FS that is the current home site.
When this data is received from 2, the access frequency index and the overall access frequency index for each host computer are calculated. And both file server FS
The communication section 16 transmits the access frequency index between the FS1 and the FS2, and the file home site determining section 14 compares the two and determines the one having the lowest access cost as the next home site.

When the file home site determining unit 14 determines to change the home site as described above, the file transfer unit 15 transfers the file entity. The current home site during the transfer of this file entity
Continue to service files from FS2 to each host computer H1 to H6, without interruption. After the transfer of the file entity is completed, each host computer H1
~ H6 file server FS2 of home site of file A
From the host computer H1 to
In H6, when the file A is accessed in the application, the internal information is changed to access the file server FS1 which is the new home site.

It should be noted that all are described in advance by logical position algorithm names or the like so as not to be affected by the change of the physical position of the file, and the physical /
It can be easily realized by rewriting the logical correspondence.

Further, as a premise of file use, it is premised that it is possible to predict the near future use frequency from the access frequency if the applications to be used are the same.

As described above, by changing the home site of the file, the time for accessing the file is shortened and the pay communication network (high-range network etc)
In that case, it is possible to reduce communication charges.

The actual procedure will be specifically described below. In FIG. 5, file A is stored in the file server FS2.
Is located. That is, the file server FS2 is the current home site. The number of accesses from each host computer H1 to H6 to the file A located in the file server FS2 was measured over, for example, one hour, and the frequency was recorded in the file server FS2. However, when accessing the file A from each host computer H1 to H6, the coefficient "1" of the logical distance is counted each time it goes through any one of LAN L1 to L5, and it is slow and charged when passing through WAN W1. Therefore, the coefficient "5" of the logical distance is counted each time the vehicle passes once.

The coefficient "1" for the above LAN and
The coefficient "5" for WAN can be set appropriately by weighting it based on the home site decision strategy of the file.

Here, it is assumed that the measurement result found in the storage mechanism of the file server FS2 is as shown in FIG. In other words, from the host computer H1 via LAN L1, L2 and WAN W1, two LAN
If the logical distance coefficient for L1 and L2 is “2” and the logical distance coefficient for WAN W1 is “5”, the logical distance becomes “7” and the access frequency is “10”, then the index becomes “70”. Become.

Since the host computer H2 passes through LANs L1, L2 and WAN W1, the logical distance is "7" and the access frequency is "20", so the index is "140". Since the host computer H3 goes through LAN L1, L2 and WAN W1, the logical distance is "7" and the access frequency is "1".
Since it is 0 ", the index is" 70. "Since the host computer H4 goes through LAN L4, L2 and WAN W1, the logical distance is" 7 "and the access frequency is" 30 ", so the index is" 210 ”. LAN from host computer H5
Since it goes through L3, L2 and WAN W1, the logical distance is "7" and the access frequency is "30", so the index is "210". The logical distance is "3" from the host computer H6 via LAN L5 and WAN W1, and the index is "30" because the access frequency is "10". From the above, the final access frequency index is calculated as "730".

Next, the file server FS2 is a computer that can be a home site for the file A, that is, FIG.
In the example shown in, the contents of the access frequency management table (information identifying the access source host computer, access frequency information) are sent to the file server FS1. The file server FS1 obtains the access frequency index based on the information sent from the file server FS2 by the same processing as that performed by the file FS2. Specifically, it is as follows.

Since the host computer H1 passes through LANs L1, L2 and L4, the logical distance is "3" and the access frequency is "10", so the index is "30". The logical distance is "3" from the host computer H2 via LAN L1, L2 and L4, and the index is "60" because the access frequency is "20". LA from the host computer H3
Since it passes through N L1, L2, and L4, the logical distance is “3”, and the access frequency is “10”, so the index is “30”. The logical distance is "1" from the host computer H4 via LAN L4, and the index is "30" because the access frequency is "30". From the host computer H5
Since it passes through LAN L3, L2 and L4, the logical distance is "3" and the access frequency is "30", so the index is "90". LAN L6, WAN W1, LAN from the host computer H6
Since it goes through L2 and LAN L4, the logical distance becomes "8",
Since the access frequency is "10", the index is "80".
From the above, the final access frequency index when the file A is located in the file server FS1 is calculated as "320".

The two file servers FS1 and FS2 bring in an index for comparison and determine the file server FS1 having a lower cost as a new home site for the file A. In this case, only the access frequency index obtained as described above may be used as a guide, or weighting for the logical distance between the file servers FS1 and FS2 and the host computers H1 to H6, or comparison logic, for example, the index difference is " Since there can be various determination methods such as only for 500 "or more,
The user may select it appropriately.

After determining the new home site for the file A as the file server FS1 as described above, the file server FS2 transfers the file A to the file server FS1.
During the transfer of this file A, each host computer H1 ~
H6 continues to access file server FS2.

File server FS2 to FS1 of file A
When the transfer is completed, the file server FS1 notifies all the host computers H1 to H6 that it is the home site of the file A. In response to this, each host computer
H1 to H6 change the access of the file A in its own system to the file server FS1.

In the above embodiment, an example in which two file servers are connected to the network has been described, but it goes without saying that any number of file servers may be used.

[0033]

As described above in detail, according to the conventional method, the access cost from each host to the file server is not evaluated and cannot be corrected if it is used continuously. It can be omitted by rearranging. In recent years, network configurations have become complicated, and optimization can be achieved by consciously measuring access costs that are difficult to grasp intuitively. Considering that the real-time property (simultaneity) of file access is deteriorated every time it passes through the router in the LAN and not only in the case of WAN but also the charged communication fee is charged, the present invention The effectiveness is clear.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a main part of a file server.

FIG. 2 is a schematic diagram showing an example of recorded contents of an access frequency management table provided in a file server.

FIG. 3 is a schematic diagram showing a specific example of recorded contents of an access frequency management table provided in a file server.

FIG. 4 is a schematic diagram showing a specific example of recorded contents of an access frequency management table provided in a file server.

FIG. 5 is a schematic diagram showing an example of a hardware environment to which a file server selection system according to the present invention is applied.

[Explanation of symbols]

 12 Access frequency management table 13 Access frequency measurement unit 14 File home site determination unit 15 File transfer unit

Claims (1)

[Claims] 1. A file server selection system for selecting one of a plurality of computers connected via a network as a file server in which a file entity should be located, wherein each of the plurality of computers has its own file server. In the case of each of the individual files, an access frequency measuring unit that specifies the partner computer that has accessed each file, and measures the logical distance and the number of accesses to the specified partner computer, Based on the information measured by the access frequency measuring means or the information measured by the access frequency measuring means of another computer, access obtained by multiplying the logical distance between itself and the other partner computer and the number of accesses. The frequency is calculated and the Access frequency recording means for obtaining the access frequency index and the access frequency index obtained by the access frequency recording means of itself and another computer are compared according to a predetermined algorithm, so that any one of the plurality of computers is targeted. Is provided with a file entity position determining means for determining a file server to be located, and a file transferring means for transferring the file entity to the computer determined as a new file server, and the file server computer measures its own access frequency. The access frequency to the file in which the file entity is located is measured by the means, and the access frequency index is obtained by the access frequency recording means of the means. The access obtained by receiving the information measured by the measuring means, obtaining the access frequency by multiplying the logical distance between itself and the number of times of access by the access frequency recording means of its own, and further adding them by the access frequency recording means of its own The frequency index is calculated, and the computer that should be the new file server is determined by the file entity position determining means of each computer. The computer that was the file server until then transfers its own file to the computer that is newly determined as the file server. A file server selection system characterized by transferring a file entity by means.