JPH10222411A - Proxy cache server control method and proxy cache server - Google Patents

Abstract

(57) [Summary] [Problem] Only data that may be inquired remains in the upper proxy cache server, and the cacheable data capacity can be effectively used, and as a result, the cache data hit rate is increased. It is an object of the present invention to provide a proxy cache server control method and a proxy cache server capable of performing the above. According to the present invention, it is confirmed that, among the multi-stage proxy cache servers, all the lower-stage proxy cache servers located immediately below the upper-stage proxy cache server have cached the same data in the upper-stage proxy cache server. At that point, the data in the upper proxy cache server is deleted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a proxy cache server control method and a proxy cache server, and more particularly to an efficient cache data stored in a proxy cache server in a network for storing data transfer such as WWW. The present invention relates to a method of controlling a proxy cache server connected in multiple stages for accumulating well and a proxy cache server.

[0002]

2. Description of the Related Art Each user accesses, via a proxy cache server, a data storage server provided with data such as WWW pages having target data.
When a user accesses a data storage server and transfers data from the data storage server to the user,
The proxy cache server caches the data. When another user requests the data, the proxy cache server transfers the cached data to the user instead of the data in the data storage server. If the proxy cache servers are connected in multiple stages as shown in FIG. 6, the upper proxy cache server can cache all data cached by a plurality of lower proxy cache servers. When a user tries to get some data,
Even if the lower proxy cache server does not cache the data, if the upper proxy cache server caches the data, the user can acquire the data from the upper proxy cache server.
As a result, the hit rate of cache data can be increased.

In a conventional proxy cache server connected in multiple stages on a tree, each proxy cache server caches and deletes data independently, and the upper and lower proxy cache servers in the configuration of FIG. If not, when a certain user obtains data from the data storage server, both the upper and lower proxy cache servers cache the data. When erasing cache data, each proxy cache server compares "current time-data cache time" with a threshold value, and if it exceeds the threshold value, deletes it. Alternatively, when new data is cached, if it exceeds the cacheable data capacity of the proxy cache server, it is erased according to the cache data replacement algorithm. As the replacement algorithm, for example, an algorithm having an old cache time or an algorithm having a large data amount is deleted.

The processing of the proxy cache server connected in multiple stages on the tree is summarized as follows.
In a conventional proxy cache server, the same processing is performed by both upper and lower proxy cache servers. <When data is stored>-If the data is not cached in both the lower and upper proxy cache servers: Cache with the upper and lower proxy cache servers,
Transfer the data to the user.

When only the upper proxy cache server caches the data: The lower proxy cache server caches the data and simultaneously transfers the data to the user. When the lower row has cached the data: the lower proxy cache server transfers the data to the user.

<Erase of data> Elapsed time since data is cached> When there is cache data that becomes a threshold: The data is erased. -When trying to cache new data exceeding the cacheable data capacity: Erase according to the cache data replacement algorithm.

[0007]

However, in the above-described conventional proxy cache server connected in multiple stages on the tree, when the data of the upper proxy cache server is erased, especially when the cacheable data capacity of the proxy cache server is exceeded. It is not efficient to delete old cache time or large data amount. This is because, in a multi-stage connected proxy cache server, the data cached by all the lower-level proxy cache servers is not cached by the upper-level proxy cache server even though there is no longer any inquiry to the upper-level proxy cache server. By keeping it as.

The present invention has been made in view of the above points, and eliminates useless cache data retention in a conventional upper-stage proxy cache server so that only data that may have an inquiry to the upper-stage proxy cache server can be obtained. It is another object of the present invention to provide a proxy cache server control method and a proxy cache server that can effectively use the cacheable data capacity and consequently increase the cache data hit rate.

[0009]

FIG. 1 is a diagram for explaining the principle of the present invention. The present invention relates to a proxy cache server control method for controlling a proxy cache server in a multi-tier proxy cache server network of a network for storing-type data transfer such as WWW. When it is confirmed that all the lower-level proxy cache servers located in the first row have cached the same data, the data of the upper-level proxy cache server is deleted.

Further, according to the present invention, among the multi-stage proxy cache servers, the lower-stage proxy cache server connected to the user host accesses the upper-stage proxy cache server (step 1), and the upper-stage proxy cache server sends a request to the lower-stage proxy cache server. When the cache data is transferred (step 2), and the upper proxy cache server confirms that all the lower proxy cache servers have cached the same data (step 3), the cache data of the upper proxy cache server is transmitted. Is erased (step 4).

FIG. 2 is a diagram showing the principle of the present invention. The present invention provides a multi-stage proxy cache server 100, 20 on a network for storing-type data transfer such as WWW.
0, the proxy cache server 100,
200 is a proxy cache server that controls 200
The data transfer unit 10 that transfers cache data based on an access from the proxy cache server 200 below the own proxy cache server 100 and all the proxy cache servers 200 below the own proxy cache server 100 transfer the same cache data. Cache determination means 2 for determining whether or not cached
0, and when the cache determining unit 20 determines that the data is cached, the cache data deleting unit 3 deletes the data of the own proxy cache server 100.
0.

As described above, according to the present invention, the cache data is transferred to the accessed lower-level proxy cache server, and it is checked whether the cache data has been cached. By deleting the cache data of the cache server, only the data that may be accessed remains in the upper proxy cache server.

[0013]

FIG. 3 shows the configuration of a proxy cache server according to the present invention. The network diagram shown in FIG. 1 represents a network for the purpose of transfer of stored data such as WWW pages. The multi-stage proxy cache server shown in FIG. 1 includes an upper-stage proxy cache server 1, lower-stage proxy cache servers 2, 3, and a host 6 of a user who requests acquisition of a WWW page.

The upper proxy cache server 1 has a cache data management unit 4 and a lower proxy cache server information management unit 5. The cache data management unit 4
The data of the WWW page of the WWW server is stored and managed. The lower-stage proxy cache server information management unit 5 manages information on which cache data is held in which lower-level proxy cache server.

When there is a request for cache data from the lower proxy cache server A2 to the upper proxy cache server 1, the cache data management unit 4 of the upper proxy cache server 1 stores the stored cache data in the lower proxy cache server A2. And notifies the lower proxy cache server information management unit 5 that the cache data has been transferred to the lower proxy cache server A2. Thus, the lower proxy cache server information management unit 5 manages the transfer destination of the cache data.

The upper proxy cache server 1
Both the lower proxy cache server A2 and the lower proxy cache server B3 delete the cache data of the upper proxy cache server 1 according to the present invention, in addition to the conventional cache control method. FIG. 4 is a sequence chart of a cache data erasing operation of the upper proxy cache server of the present invention.

Step 101) The user host 6 accesses the lower proxy cache server 2. Step 102) There is no desired WWW page data in the lower proxy cache server 2, and the lower proxy cache server 2 accesses the upper proxy cache server 1.

Step 103) The upper proxy cache server 1 transfers the cache data to the lower proxy cache server 2. Step 104) The lower-level proxy cache server information management unit in the upper-level proxy cache server 1 has already notified that the lower-level proxy cache servers 2 and 3 (the lower-level proxy cache servers A and B in the example of FIG. 3) cache the same data. When 5 is detected, the cache data in the cache data management unit 4 in the upper proxy cache server 1 is deleted.

[0019]

An embodiment of the present invention will now be described with reference to FIGS.
It will be described based on. When the above series of operations is applied to the example of FIG. 3, only the lower proxy cache server A2 and the lower proxy cache server B3 located at the lower level of the proxy cache server 1 are connected to the upper proxy cache server 1.

Here, when a data request is issued from the lower proxy cache server A2 requested by the user host 6 to the upper proxy cache server 1, the cache data management unit 4 of the upper proxy cache server 1 transmits the data request to the lower proxy cache server 1. The cache data is transferred to the server A2, and the transferred information (= A) and the cache data identifier (ID = a) of the lower proxy cache server A2 are notified to the lower proxy cache server information management unit 5. As a result, the lower proxy cache server information management unit 5 transmits the transferred server name “A” and the cache data identifier (ID =
a) and the time.

Next, the lower proxy cache server B3
Issue a data request similar to the data requested from the lower proxy cache server A2 to the upper proxy cache server 1, the cache data management unit 4 of the upper proxy cache server 1 stores the cache data corresponding to the data request. The data is transferred to the lower proxy cache server B3. Then, the lower proxy cache server information management unit 5 is notified of the information (= B) and the cache data identifier (ID = a) of the lower proxy cache server B3 that has transferred the data. As a result, the lower proxy cache server information management unit 5 records the transferred server name “B”, the cache data identifier (ID = a), the time, and the like.

Since the lower proxy cache servers connected to the upper proxy cache server 1 are only the lower proxy cache server A2 and the lower proxy cache server B3, the lower proxy cache server information management unit 5 executes all lower proxy cache servers. It is confirmed that the same cache data (ID = a) has been transferred to the server A2 and the lower proxy cache server B3. Thereby, the cache data identifier (ID = a) stored in the cache data management unit 4
Clear the cache data of

The present invention is not limited to the above embodiment, but can be variously modified and applied within the scope of the claims.

[0024]

The effects of the present invention will be described below. First, a description will be given using the model of FIG. In the figure, it is assumed that 100 users access each lower-level proxy cache server with one upper-level proxy cache server, two lower-level proxy cache servers, and a total number of users of 200. The data access from each user is assumed to arrive at Poisson, and the average access time interval is set to 100. Here, data i
Is generated in accordance with λ (i) = q (1−q) ⁱ (1). Here, q is a parameter for determining the average of the index i of the access destination data.
The average is given by 1 / q. Here, q =
0.99. As a result, the smaller i is, the higher the access rate is. It is assumed that the user in FIG. 6 does not have a data storage server. The timeout occurs in 500 unit times, and the cache data that has timed out is erased. The number of cache data in the proxy cache server is limited to the number of entries in the proxy cache server, and there is no upper limit due to the data amount. Also, with the cache data for the number of entries already,
When new data is cached, the conventional method uses a replacement algorithm LRU (Least Recently Used).
, The old data is deleted. In the replacement algorithm LRU, the last access time is erased from the cache data having the oldest possible time. Also, the upper limit of the number of entries of the proxy cache server is the same in the upper and lower rows.

FIG. 5 is an evaluation result showing the effect of the present invention. The hit ratio per entry when the number of entries that can be cached by each proxy cache server changes was determined. As can be seen from the evaluation results shown in the figure, the hit ratio of the upper proxy cache server is more effective in the proposed method than in the conventional method.

The cache data erasing procedure of the upper proxy cache server according to the method of the present invention can be used together with a conventional cache control method. The reason is that the erase execution time in the method of the present invention is different from the conventional erase execution time. Further, according to the present invention, the function can be realized only by the upper proxy cache server. Since the upper proxy cache server can always know the data cached by the lower proxy cache server, the present invention does not require a special protocol between the upper proxy cache server and the lower proxy cache server.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the principle of the present invention.

FIG. 2 is a principle configuration diagram of the present invention.

FIG. 3 is a configuration diagram of a proxy cache server of the present invention.

FIG. 4 is a sequence chart of a cache data erasing operation of the upper-stage proxy cache server of the present invention.

FIG. 5 is an evaluation result showing the effect of the present invention.

FIG. 6 is an example of a proxy cache server connected in multiple stages.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Upper proxy cache server 2 Lower proxy cache server A 3 Lower proxy cache server B 4 Cache data management part 5 Lower proxy cache server information management part 6 User host 10 Data transfer means 20 Cache determination means 30 Cache data deletion means 100 Upper proxy cache Server 200 Lower proxy cache server

Claims (3)

[Claims] 1. A multi-stage proxy cache server control method for controlling a proxy cache server in a multi-stage proxy cache server network of a network for storing data transfer such as WWW. A method of controlling a proxy cache server, characterized in that when it is confirmed that all lower-level proxy cache servers located immediately below have cached the same data, the data of the upper-level proxy cache server is deleted. 2. The multi-tier proxy cache server, wherein the lower proxy cache server connected to the user host accesses the upper proxy cache server, and the upper proxy cache server transmits cache data to the lower proxy cache server. 2. The proxy cache server according to claim 1, wherein the upper proxy cache server erases cached data of the upper proxy cache server when confirming that all lower proxy cache servers have cached the same data. 3. Control method. 3. A proxy cache server comprising a multi-stage proxy cache server on a network for storing data transfer such as WWW and controlling the proxy cache server, wherein the proxy cache server is lower than the own proxy cache server. A data transfer unit that transfers cache data based on access from a server; and a cache determination that determines whether all proxy cache servers below the own proxy cache server have cached the same cache data as the cache data. Means, and a cache data deleting means for deleting data of the upper proxy cache server when the cache judging means judges that the data is cached. .