JPH11219313A - Content prefetching method - Google Patents

Abstract

(57) [Summary] [PROBLEMS] In a method in which a client browses a hypertext provided from a server, the idle time of the network is effectively used to efficiently browse the hypertext. SOLUTION: A client 20 analyzes a content 28a of a page displayed according to a user's instruction, and inspects a page linked from the page.
For each of the pages to which the link is provided, page requests 26b and 26c are made by link prefetching. Then, the content of each page is acquired, and the content of the pre-read page is stored in the cache memory 24 in advance. When the browsing of the user is completed and there is an instruction from the user to newly browse the linked page, the content of the content read in advance from the cacher memory 24 is taken out and displayed. Accordingly, a content prefetching method that can quickly respond to an instruction from a user while effectively using the idle time of the network is realized.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method for a client to browse contents from a server on a network. In particular, WWW (World Wide W)
eb) A method in which a client browses content provided by a server or the like.

[0002]

2. Description of the Related Art HTML (Hyper Text Ma)
WWW technology for distributing contents using rk up language) is widely used on the Internet and the like. When reading out the content, generally, a browser program is installed in the client computer, and the content of the page desired by the user is read out from the server using the browser.

FIG. 13 is a sequence diagram showing this read operation.

In a client 10 having a conventional browser program (hereinafter simply referred to as a browser),
First, when the user inputs a desired page, the browser issues a page request 14a to the server.
The server 12 receives the page request 14a transmitted via the network, and transmits the content 16a of the requested page to the client 10 also via the network. In this way, every time a page request 14 is received from the user, the browser of the client 10 acquires the relevant page via the network.
The client 10 displays the content 16a of the desired page acquired in this manner, so that the user can see the content of the page (see FIG. 13).

As shown in FIG. 13, when the contents of a page are transmitted to the client 10 and the user is browsing the contents, the network is kept in an idle state. When the user finishes browsing the page and browses a new page, the client 10 transmits a page request 14b to the server 12 again to acquire the page specified by the user. The server 12 responds to the page request 14b by
The content 16b of the corresponding page is returned to the client 10 via the network. Client 10
By displaying the content 16b of the transmitted page, the user can browse the content of the desired page.

As shown in FIG. 13, the client 10 issues a page request 14 when the user wants to browse the contents of a new page. Therefore, while the user is browsing the contents of the predetermined page, the network is not used at all and is in an idle state.

[0007] Various types of devices for browsing contents provided by a server through a network have been proposed. For example, JP-A-8-875
No. 26 discloses a hypermedia document communication device. According to the description of this document, it is described that a pre-read can be effectively performed, and a hypermedia document communication device with good responsiveness can be realized.

[0008]

As described above, FIG.
The client 10 using the conventional browser as shown in FIG. 3 issues a page request 14 every time the user designates a page to be browsed, so that the user browses the contents of the page. During this time, no data flows in the network and the network is in an idle state, causing a problem that network efficiency is deteriorated.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and it is an object of the present invention to provide a content prefetching method which makes effective use of a network idle time and has good responsiveness to a user's page instruction.

[0010]

According to the present invention, there is provided a page instruction receiving step of receiving from a user an instruction of a page to be viewed by the user, and a page indicated by the instruction received in the page instruction receiving step is stored in a cache memory. A checking step of checking whether the specified page exists in the cache memory, and in the checking step, if the specified page does not exist in the cache memory, requesting the specified page to a server; In a server reading step of displaying a page transmitted from a server to a user, and in the checking step, when the specified page exists in the cache memory, the content of the specified page is read from the cache memory. Read the cache memory and display the read page to the user Inspect the steps, and the content of the page displayed on the user at the server reading step or the cache memory read step,
A link destination page that requests the server for content of a link destination page linked from the displayed page, and stores the content of the link destination page transmitted from the server as a result of the request in the cache memory. And a look-ahead step.

Further, in the present invention, in the cache memory reading step, the expiration date of the content of the designated page in the cache memory is checked, and only when the expiration date has not passed, the cache memory is read from the cache memory. The content of the designated page is read, and the read page is displayed to a user.

Further, the present invention is characterized in that the expiration date is determined separately and independently for each page.

Further, the present invention is characterized in that the expiration date is determined independently for each component of each page.

Further, the present invention is characterized in that the linked page prefetching step executes reading of contents of a plurality of linked pages in parallel.

[0015] In the present invention, the link destination page prefetching step includes the step of: when the user issues an instruction to view any one of the plurality of link destination pages from which the content is read, The reading of the linked page to which the instruction is issued is continued, and the reading of the linked pages other than the page to which the instruction is issued is interrupted.

Further, in the present invention, in the step of prefetching linked pages, when reading the contents of the plurality of linked pages in parallel, the text data in the contents of the plurality of linked pages is replaced with data other than the text data. It is characterized by reading out earlier.

Further, the present invention is characterized in that, in the step of prefetching linked pages, the contents of a plurality of linked pages are read out sequentially for each linked page.

[0018] In the present invention, the link destination page prefetching step includes the step of: when the user gives an instruction of a page to be viewed before reading of the content of the link destination page is completed, Is interrupted.

[0019] In the present invention, the link destination page prefetching step includes the step of: when a user instructs browsing of a link source page linked to the link destination page, The content is displayed to the user, and the reading of the linked page whose reading has been interrupted is restarted.

In the present invention, the step of pre-reading the linked page may include, when resuming reading of the linked page whose reading has been interrupted, setting an expiration date of a read result in the middle of the linked page whose reading has been interrupted. Inspect,
If the expiration date has passed, read the linked page again, and if the expiration date has not passed, continue reading from the part where reading of the linked page was interrupted. Is what you do.

Further, the present invention is characterized in that in the link destination page prefetching step, the reading order of the contents of a plurality of link destination pages is determined by the priority of each link destination page.

In the present invention, the step of prefetching linked pages may include dividing the plurality of linked pages into a predetermined number of groups, and reading out the contents of the plurality of linked pages sequentially for each of the groups. The reading of the contents of the linked pages included in each group is performed in parallel for a plurality of linked pages included in the group.

[0023] In the present invention, the linked page prefetching step may include the step of grouping the plurality of linked pages by a predetermined number of pages in the order of priority assigned to the linked pages. It is characterized by forming a group.

Further, in the present invention, in the step of prefetching a linked page, the priority is determined based on a frequency of access to a page in the past.

Further, the present invention is characterized in that the access frequency is the number of accesses.

Further, the present invention is characterized in that, in the link destination page reading step, the content of the link destination page is prefetched only when the prefetch operation is permitted.

[0027]

Preferred embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1 FIG. 1 is a sequence diagram illustrating the operation of the content prefetching method according to the preferred embodiment of the present invention. In this figure,
Client 20 as in FIG. 11 showing the conventional method,
The transmission and reception of data with the server 22 are shown.

The client 20 according to the present embodiment has a cache memory 24 therein. The cache memory 24 is a memory for appropriately storing the contents of the page provided by the server 22 as described later.

Further, similarly to the conventional client 10,
The client 20 according to the present embodiment also issues a page request to the server 22 in principle based on an instruction of a page to be viewed from a user. Here, a feature of the present embodiment is that the client 20 determines whether or not the content of the corresponding page is stored in the cache memory 24 before actually transmitting the page request to the server 22. It is to inspect. For example, the client 20 first requests the server 22 for a page request 26.
a, the cache memory 24
Does not contain any page content. Here, the operation when the contents of the page are stored in the cache memory 24 will be described later.

As described above, when the corresponding page does not exist in the cache memory 24, the client 20 requests the page request 26 similarly to the conventional client 10.
a to the server 22. The server 22 has exactly the same configuration as the conventional server 12, and the operation is exactly the same. Accordingly, the server 22 returns the content 28a of the corresponding page to the client 20 via the network in response to the page request 26a. The client 20 displays the transmitted content to the user similarly to the conventional client 10.

A feature of this embodiment is that the client 20 analyzes the page content 28a sent from the server 22. This analysis is represented by “page analysis” in FIG. Specifically, this page analysis is to inspect a linked page from a transmitted page. The inspection of the linked page is performed on the currently displayed page, and there may be a plurality of linked pages.

Next, the client 20 checks the priority of the linked page. To set the priority for this landing page, refer to the past history, etc.
This is done so as to set a higher priority for the one that is most likely to be viewed next among the linked pages. The specific operation of setting the priority will be described later in detail.

Now, the client 2 according to the present embodiment
0 prefetches the content of the linked page linked from the currently displayed page. The prefetching of the link destination page is represented as "link prefetching" in FIG.

A feature of this embodiment is that a predetermined page is displayed, and while the user is browsing the contents, the client 20 is linked to the linked page from the displayed page. Is requested to the server 22. As mentioned above,
Since there may be a plurality of the linked pages for one page, the page request 26b for the linked ahead reading,
26c is generally performed for a plurality of pages.

When the client 20 acquires the content of the linked page from the currently displayed page, the client 20 stores the content in the cache memory 24.

The user terminates the browsing of a predetermined page, and instructs the client 20 to browse the next page by clicking a predetermined link button in the page. Upon receiving the instruction for the new page, the client 20 first checks whether or not the contents of the page corresponding to the cache memory 24 are stored. As described above, while the user is browsing the current page, the content of the linked page provided with a link from the currently displayed page is read by the client 20 and stored in the cache memory 24. Have been. Therefore, the client 20 finds that the corresponding page is stored in the cache memory 24.

For this reason, the client 20 does not transmit the page request of the link destination to the server 22 but retrieves the content from the cache memory 24 inside the client 20 and displays it. This operation is represented by "process locally" in FIG. In this way, the page request 26 of the linked page is sent to the server 22
The processing corresponding to the user's instruction on the browse page is performed only inside the client 20. Therefore, according to the present embodiment, it is possible to promptly respond to a user's instruction on a linked page, and to realize a display having a quick response.

It should be noted that the prefetching of the linked pages may be performed in order for each page, or it may be preferably performed in parallel for a plurality of linked pages.

When the prefetching of the linked page is performed in order for each page, it is preferable to read the content first from the linked page having the higher priority. By storing the content in the cache memory 24 prior to the link destination page having the higher priority, the page which is likely to be browsed is preferentially prefetched, and the prefetched content is displayed according to the instruction of the user. Is expected to be improved, and the effect of prefetching can be further enhanced.

Performing prefetching of linked pages in parallel for a plurality of linked pages is particularly preferable when priority is not yet set and all linked pages are equal. . In this case, the network can be more effectively used by simultaneously performing page reading on a plurality of linked pages.

Now, it is conceivable that the user gives an instruction of a new browse page while reading the link destination page. In this case, in this embodiment, the reading of the linked page is interrupted. When the prefetching of the browse page specified by the user has been completed and the content is stored in the cache memory 24, the client 20 reads out the content and displays it. On the other hand, when the pre-reading is not completed yet, the designated browse page is read and the read content is displayed.

When the link destination page is read out in parallel, if the user gives an instruction for a new browse page during the prefetching, the prefetching of all the pages is in the middle and the prefetching is performed. No pages have been completed. Therefore, only the prefetching of the browse page supported by the user is continued, and the prefetching of the other pages is interrupted.

A feature of this embodiment is that when a new instruction for a browse page is issued by a user during a prefetch operation, prefetching of pages other than the designated page is interrupted, and The reading of the page that has been completed is to be continued. By operating in this way, even if an instruction for a new browse page is issued during the pre-reading operation, the previous pre-reading operation can be effectively used.

Therefore, according to the present embodiment, a content prefetching method is realized which can effectively use the idle time of the network and can quickly respond to a user's instruction.

Further, in this embodiment, when the user finishes browsing the linked page and returns to the linked page to browse the linked page which is the link source of the linked page, The prefetching of the interrupted linked page is resumed. In this way, even if the prefetching of the content is interrupted, if the conditions for prefetching the content are satisfied again, the prefetching is continued, thereby effectively reducing unnecessary content reading and unnecessary network idle time. Can be reduced.

Here, conditions for prefetching the content of the linked page in the present embodiment will be described.

This condition is that the link source page of the link destination page is currently displayed and is being browsed by the user. That is, the client 20 according to the present embodiment analyzes the currently displayed / browsed page,
It always tries to prefetch the contents of the linked page to which a link is provided from that page.
At this time, if the prefetching of the linked page has already been performed halfway, that is, if the prefetching is currently interrupted, the prefetching of the content is restarted.
As a result, more efficient prefetching can be performed.

As described above, in the present embodiment, when there are a plurality of linked pages, prefetching is performed in parallel for a plurality of linked pages unless a priority is set. Is executed. At this time, the text portion of the linked page, specifically, for example, HTML
It is also preferable to pre-read a hypertext language or the like before image data or the like. That is, for a plurality of link destination pages, only the text portion is read ahead first. Then, after all the text portions for the plurality of linked pages are read ahead, the content such as image data is read out for the plurality of linked pages. in this way,
Since the data of the text portion is read out before the image data and the audio data, the outline of the page can be displayed first, and the rough contents of the page can be grasped more quickly.

As described above, according to the present embodiment, while the client 20 displays the content of the page to the user, the content of the linked page to which the link is provided is read in advance from the page. It is possible to provide a client 20 that can quickly respond to a user's instruction on a browse page.

More specifically, it is better to say that a content prefetching method which is a prefetching method executed by the client and which can quickly respond to a user's instruction on a browse page is realized. There will be. This is because the prefetching operation in the present embodiment is all executed by the browser program, and the content prefetching method of the present invention is substantially realized in the present embodiment by the browser program.

Although the operation in which the client 20 prefetches the linked page has been described in the present embodiment, it is also preferable that a so-called proxy server performs the same operation. The proxy server is a proxy server provided between the client 20 and the server 22. The proxy server includes a cache memory therein and may perform the operation of the client 20 as shown in FIG. preferable.

In this case, the operation of the proxy server is almost the same as that of the client 20 in FIG. The only difference is that when the client 20 performs display to the user, the proxy server transmits the contents of the page to the client 20.

Further, in the embodiment shown in FIG. 1, it has been described that the priority is set by the access frequency or the like for each linked page in order to determine the priority of the prefetched page. It is also preferable to simply use the access count as it is. Further, the example shown in FIG. 1 shows a method of sequentially reading out the linked pages one by one according to the priority.
As described above, it is preferable to read out all the plurality of linked pages at the same time (in parallel) when the priority is not particularly set.

However, a situation in which the number of linked pages becomes extremely large is also assumed. For example, when the link destination page is about 100 pages, if the content of 100 pages is prefetched at the same time, it may take too much time to return, and the effect of prefetching may be reduced. In such a case, it is preferable to arrange the link destination pages in the order of priority, and to group, for example, 10 pages in descending order of priority. Then, the content of the page is prefetched in parallel for each group in descending order of priority.

That is, first, pre-reading is performed in parallel with respect to the link destination page having a high priority. After the prefetching of the ten linked pages is completed, the prefetching is performed in parallel on the ten linked pages with the next highest priority. As described above, if the linked pages are grouped by a predetermined number (for example, 10) and the content of the page is prefetched in parallel for each group,
Effective prefetching can be performed while preventing the network load from becoming excessive.

The operation between the client 20 and the server 22 has been described based on the sequence diagram of FIG.
Next, the basic operation of only the client 20 will be described with reference to the flowchart of FIG.

First, in a page instruction receiving step 100, the client 20 receives an instruction of a page to be browsed from a user. The user may directly input the URL of the page in some cases, or the instruction may be made by clicking a link button from a predetermined page.

In the checking step 102, the client 20 checks whether or not the contents of the specified page exist in the cache memory 24. As a result of this check, if the content exists in the cache memory 24, the process proceeds to the cache memory reading step 104, and if not, the process proceeds to the server reading step 106.

Cache memory read step 104
In, the content of the page specified by the user is read from the cache memory 24 and displayed to the user. The user can browse the displayed page.

In the server reading step, similarly to the conventional client 10, the content of the designated page is requested to the server 22, and the obtained content is displayed to the user.

In the link destination page prefetching step, the client 20 analyzes the newly displayed page and extracts a link destination page provided with a link from the page. Specifically, the URL of the linked page is obtained. Further, based on the URL, the content of the linked page is read from the server (prefetch), and the read content is stored in the cache memory 24.

The above operation is the basic operation of the client 20. As described with reference to FIG. 1, with respect to the detailed part of the operation, the prefetching of the page in consideration of the priority, the reading operation of all the pages is performed in parallel, or the prefetching of the page is performed for each predetermined group. Can be appropriately selected depending on the nature of the page to be read (the difference between the update frequency and the page access frequency).

As described above, according to the present embodiment, the prefetching of the linked page from the currently displayed page is performed and stored in the cache memory 24, so that the idle time of the network can be effectively used, and It is possible to realize a content prefetching method that can quickly respond to a user's instruction.

Embodiment 2 In the present embodiment,
The operation flow of the client 20 will be described in detail. Hereinafter, the operation flow of the client 20 will be described in order.

(1) First, the client 20 waits for a browsing page instruction from the user.

(2) Next, the client 20 receives an instruction for a browse page from the user.

(3) When the client 20 receives a page instruction and performs a prefetching operation, the prefetching process is interrupted. Here, the interruption is performed because the prefetching of the page may be resumed as described in the first embodiment. It should be noted that the page instruction from the user may be any one of the linked pages currently prefetched. In this case, the prefetch for the page specified by the user is continued. Then, prefetching is interrupted for pages other than the page for which the page instruction has been given. In the present embodiment, prefetching of pages other than the page for which browsing is instructed is interrupted, so that the page for which instructing has been read can be quickly read.

(4) Next, the client 20 holds a prefetch text list inside. This means that the cache memory 24 has been read from the server 22 until now.
Is a list showing the state of pre-reading for the text part of the page stored in. FIG. 3A shows an example of specific contents of the pre-read text list 30. As shown in this figure, the look-ahead text list 30 is:
The text portion, that is, the URL where the HTML file is stored, the frequency of access to the page (represented by the URL), the priority for the page, and a completion mark indicating whether or not reading is completed. ,
It has. Further, a next page mark indicating the expiration date of the data read for each page and whether or not the page is currently linked to the currently displayed page is provided for each page. The next page mark is a mark that is constantly updated according to the currently displayed page, and thus is updated each time the displayed page changes.

In the present embodiment, the number of accesses is used as it is as the access frequency. The contents of the pre-read text list 30 are stored in the client 20
Changes as needed as the content of the page is read. By the way, when an instruction of a page to be browsed is received from the user, the look-ahead text list 30 includes
An inspection is performed to determine whether a page specified as a page to be viewed exists. If there is, the access frequency in the row of the page is incremented (1 is added).

(5) Whether or not the pre-reading of the text portion (HTML) of the page designated as the page to be browsed by the user has been completed is determined by whether or not the completion mark of the pre-reading text list 30 has been added. Inspection is performed by That is, the completion mark indicates whether or not the pre-reading for the text portion (HTML) of each page has been completed. In the example shown in FIG. 3A, “complete” is added to all URLs (pages) as a completion mark, indicating that the prefetching of all pages has been completed. If the pre-reading is not completed, “not yet” is added as the completion mark.

If the prefetching of the text portion of the page to which the user has instructed browsing is completed, the text portion is displayed. On the other hand, if the prefetching of the text part is not completed, the client 20
Make a request for the text part. Then, the pre-read processing that has not been completed is completed. Of course, when the completion is completed in this way, the completion mark of the corresponding predetermined column in FIG. 3A is changed from "not yet" to "complete".

(6) Next, the client 20 analyzes the text of the currently displayed text portion and extracts the components included in the page. Here, the component means content other than the text portion, that is, image data, audio data, and the like. Now, a look-ahead component list 32 as shown in FIG. 3B is provided for such components in order to represent the state of prefetching. The client 20 stores not only the prefetch text list 30 but also such a prefetch component list 32 therein.

Further, the client 20 adds a link destination page linked from the currently displayed page to the current page look-ahead list. FIG. 4 is an explanatory diagram showing the stored contents of the current page prefetch list 34. As shown in FIG. 4, the current page prefetch list 34 is a list in which URLs of linked pages are arranged. In FIG. 4, two pages are currently stored in the page prefetch list 34.

(7) By the way, in the above (5), the prefetching of the text portion of the linked page is completed.
Then, the components (images and sounds) extracted in the above (6) are prefetched. Specifically, the following processing is repeated by the number of components extracted in the above (6).

(7-1) It is checked whether or not each component has been prefetched by referring to the prefetch component list 32. In the prefetching component list 32, those having "complete" as the completion mark have been prefetched. In the example shown in FIG. 3B, the prefetching of “f.gif” and “h.gif” is not completed.

(7-2) As a result of this inspection, the components for which prefetching has been completed are displayed to the user. On the other hand, for the components for which the prefetching has not been completed, the server 22 is requested to transmit the content. Then, the content is transmitted from the server 22 as a result of this request, and when the prefetching is completed, the completion mark of the prefetching component list 32 is changed to “complete”.

In this embodiment, the above (7-
The processes shown in 1) and (7-2) are repeated by the number of components. It is of course preferable to read the content and display it to the user in parallel with respect to all the components, instead of repeating the above processing for each component in order.

(8) Next, of the pages in the pre-read text list 30 shown in FIG. 3A, only the page with the next page mark is extracted. This next page mark means that the page is a link destination page linked from the currently displayed page. FIG.
In the example shown in (a), “B.html”
The next page mark is attached to “C. html”.

Next, a priority is set for the extracted link destination page based on the access frequency provided in the current page prefetch list of the currently displayed page. The priorities are assigned in descending order of access frequency. In the example shown in FIG. 3A, “0” is assigned to the priority “A.html”,
“2.” is added to “B. html”. The page “C.html” is given a priority “1”. Here, in the present embodiment, “0” in the priority column means the current page, that is, the currently displayed page. Also, a page that is not shown in FIG. 3A but does not have a link from the currently displayed page is not given a priority, and “-” is given in the priority column. Will be. FIG.
In the example shown in (a), the page "A.htm
"1" indicates that the page is currently displayed. Also, two pages “B.html” and “C.htm” linked from the currently displayed page.
ml ”, the page“ Ch.
tml "has a higher priority. In the present embodiment, the priority is a positive integer of 1 or more, and 1 means the highest priority, and the larger the number, the lower the priority.

(9) Next, the pre-read text list 30
Priority (priority) shown in FIG. 3 (a)
, A series of processes shown in the following (9-1) and (9-2) are executed for a linked page linked from the current page. Note that these processes may be sequentially performed for each page by, for example, a loop process, but it is also preferable to proceed in parallel for a plurality of pages.

(9-1) For each line (each record) of the look-ahead text list 30, it is checked whether the completion mark is “complete”.

(9-2) As a result of this inspection, if the completion mark is not “completed” but “not yet”, a request for the text of the next page is transmitted to the server 22. In this case, the text of the next page means a text portion of the linked page, that is, text data such as HTML. When receiving the text of the next page from the server 22, the client 20 changes the completion mark of the prefetch text list 30 from "not yet" to "complete".

(10) Next, the following processes (10-1), (10-2), and (10-3) are executed according to the order of priority for the number of next pages. The next page means a page linked from the current page, that is, a linked page. In addition, it is preferable to execute a series of processing described below for each page according to the priority order,
It is also preferable to execute the processing on all pages in parallel. Hereinafter, the processing will be described in order.

(10-1) The text of the next page is analyzed, and the components included in the page and the link to the next page are extracted. Then, the extracted component is added to the prefetch component list 32, and the link to the extracted next page is added to the prefetch text list 30.

(10-2) Request the server 22 for the data of the components included in the page. This request is preferably processed in order for each component (loop processing), but is preferably performed in parallel (simultaneously) for all components.

(10-3) It is checked whether or not the completion mark of the corresponding component in the prefetch component list 32 is “complete”. If the result of this check indicates that the completion mark is not “completed” but “not yet”, the server 22 requests the server 22 to transfer the data of the component. As a result of this request, when the client 20 receives the data of the component from the server 22 and stores the data in the cache memory 24, the pre-read component list 32 changes the completion mark of the corresponding component from “not” to “complete”. To "."

(11) Next, the client 20 deletes the pre-read data whose expiration date has passed. This is because the data held by the server 22 is updated as appropriate, so that an expiration date is provided for the pre-read data and the freshness of the content is maintained. FIG.
As shown in FIG. 3A and FIG. 3B, the expiration date is set for each of the pre-read data. In the example shown in FIG. 3, the expiration date of the pre-read data is set to one day, but it is also preferable to set a different expiration date for each data.

(12) The above processing is repeated. That is, the above processes (1) to (11) are repeated again.

As described above, according to the client 20 according to the present embodiment, by performing the text analysis, it is possible to pre-read the components included in the page.

According to the operation of the client 20 described above, the prefetch function is always executed.
It is also preferable that the prefetch function be turned on / off. That is, it is possible to always perform prefetching based on a user's instruction, or to perform an operation similar to the conventional operation without performing prefetching. This is because the prefetching operation increases the load on the network. Therefore, when it is desired to reduce the load on the network as much as possible, it is preferable not to perform the prefetching like the conventional client.

As described with reference to FIG. 3, the pre-read data has an expiration date, but is the page actually read or the page not actually read? It is also preferable to make the expiration dates different on the basis of

Now, in FIG. 3, specific examples of the contents of the prefetch text list 30 and the prefetch component list 32 have been described. These contents will be described in more detail.

As mentioned above, each list 3
Those in which “complete” is provided as a completion mark in 0 and 32 mean that the completion mark is stored in the cache memory 24 inside the client 20. Then, as described above, when the new URL is to be displayed to the user (to be viewed), the new URL is displayed.
The URL of the link destination page linked from L is added to the prefetch text list 30. Then, the URL of the component included in the added new URL is added to the prefetch component list 32 (see FIG. 3B).

Now, the expiration date has been described.
The client 20 according to the present embodiment deletes the data whose expiration date has passed. Here, deleting means setting the completion mark to “not yet”. It does not actually delete the data itself. The reason is that the number itself of the access frequency (the number of times of access) can be used when obtaining the priority or the like.

The example of the contents shown in FIGS. 3A and 3B represents the contents when browsing the hypertext as shown in FIG. As shown in FIG. 5, first, page A, which is the root, has A.P. html
Is described by the text. here . ht
The extension “ml” means that this file is a file described in HTML.

The text data A. html
Has B.B. html and C.I. html. As a result, a link to page B and page C is established (see FIG. 5). A. page representing this A page. ht
ml is image data therein. e. gif. Thereby, the image data e. gif is a component of the A page. here . g
The extension "if" means that this file is image data represented in the GIF format.

Thus, from page A, page B and C
There is a link to the page. In addition, B. describing page B. html, D.H. html is referred to, and a link to page D is established. In addition, B. describing page B. html contains the image data f. Gif is referenced. Thereby, the image data f. gif is a component of the B page.

C. Description of page C html, D.H. Reference to html is made, whereby a link to page D is established. On page C, image data g. g
if and image data h. gif is referred to as a component.

As described above, the page D is a page to which a link is provided from both the page B and the page C. D. text, which describes this D page htm
l internally contains image data i. gif. As a result, the image data i. gif is included.

In the example of the hypertext as shown in FIG. ht
The page A is displayed by reading ml. Then, the client 20 according to the present embodiment
A. Text that describes the page. The html is analyzed to detect the linked page. This page is Figure 5
The page B and the page C are shown in FIG. Therefore, the client 20 transmits this B. html and C.I. ht
First look ahead for the text portion of ml. FIG. 3 described above
The look-ahead text list 30 shown in (a) shows a state in which the text portions for the pages B and C are added to the list so as to be prefetched. Also, since the currently displayed page is page A, the page next to page A, that is, the linked page is page B and page C, and accordingly, page B. html and C.I. The next page mark is attached to both html. Next, the text portion B. html and the text portion of the C page, C.I. The contents of the html are analyzed, and the components constituting the B page and the C page are inspected. As a result, the client 20 displays the image data f.
gif, and the C page contains g. gif and h.
gif is found to be included.

Accordingly, the client 20 registers these image data in the prefetch component list 32 and starts prefetching these image data. In the example shown in FIG. 3B, the image data g.
gif, the pre-reading is completed and the image data f. gi
f and h. It is shown that gif is still being pre-read.

Now, the prefetching of the text portions of the pages B and C has been completed in this way, and during the prefetching of the constituent elements (image data) constituting the respective B and C pages, the user starts from the page A. When the button for the page C is clicked on the screen, the client 20 must display the page C to the user next. For this reason, the client 20 transmits f. Image data included in page B, which is a page other than the page (page C) specified by the user. Gif prefetch is interrupted. Then, only the prefetching of the components included in the page C, which is the page specified by the user, is continued.

That is, the client 20 transmits the image data f. gif is interrupted, and the image data h. It concentrates on gif prefetching. In this way, after the client 20 has obtained all the data relating to the text portion and the constituent elements of the C page, the client 20 displays the C page to the user.

A feature of the present embodiment is that, when a user gives an instruction on a page to be browsed during a prefetch operation, only the prefetch for that page is continued, and the other pages are read. Look-ahead is to be interrupted. In this way, when the pre-reading process for a plurality of pages is performed, when the instruction of the page to be viewed is issued, the data reading process is performed intensively only on the specified page, Data can be read efficiently.

The client 20 always prefetches the data of the linked page from the currently displayed page. Therefore, since the C page is currently displayed according to an instruction from the user, the client 20 pre-reads the D page linked from the C page this time.

As shown in FIG. 5, the D page has a text portion of D.D. html, and the image data included in the D page is i. gif, this D.G.
html and i. Gif is prefetched.

As described above, FIGS. 6A and 6B show the contents of the prefetch text list 30 and the prefetch component list 32 immediately after the C page is displayed according to the instruction from the user. b) respectively.

FIG. 6A shows the contents of the look-ahead text list 30, but since the currently displayed page is the C page, C.I. The priority for html is set to “0”. On the other hand, since there is no link from page C to page A or page B,
A. html or B.I. No priority is set for html or the like, and "-" is set. Immediately after page C is displayed, prefetching to page D is being performed. Since the pre-reading of html is being performed, the completion mark is set to “not yet”.

FIG. 6B shows a prefetch component list 3
2 are shown. As shown here,
Immediately after displaying the C page, the look-ahead constituent element list 32 newly includes image data i. gif is added to the image data i. Gif is prefetched. Therefore, this i. “Not yet” is set as the completion mark for the gif. The image data f. Since the pre-read of the gif has not been completed yet, “not yet” is set in the completion mark.

Further, as described above, it is preferable to set different expiration dates for the text portion and different components of different pages. This is because some pages change frequently, while others rarely change. In the example shown in FIG. 6, an example is shown in which the contents of the D page are contents that are hardly changed. In this case, D
Text portion related to page D. html or image data i. The expiration date for the gif is preferably set to be longer for other text parts and components. For example, in the example of FIG. 6, 10 days is set as the long expiration date. By setting such a long expiration date, the cache memory 24
The period during which the data inside is valid can be made longer, and the frequency of prefetching can be reduced. Therefore, it is possible to display the D page more efficiently.

Embodiment 3 Next, a more detailed operation of the client 20 according to the present invention will be described based on a flowchart. First, FIG. 7 shows a general flowchart of the client 20 according to the present embodiment.
First, as shown in this flowchart, in step S6-1, the client 20 receives a page request from a user.

Next, in step S6-2, it is checked whether or not the client 20 is currently performing the prefetching process. If the pre-reading process is being performed, step S6
The process proceeds to -3, and the prefetching process is stopped. On the other hand, if it is determined in step S6-2 that the process is not the prefetching process, the process immediately proceeds to step S6-4.

As described above, if the pre-reading process is being performed, the pre-reading process is stopped. If the pre-reading process is not being performed, the process directly proceeds to step S6-4, and the page display process is performed. In this step, the page requested by the user as the browse page is displayed.

Note that the client 2 according to the present embodiment
In the case of 0, prefetching from the currently displayed page to the linked page is always executed in the background, so to speak. Therefore, after a new page is displayed in step S6-4 in FIG.
The prefetching process for the new page is automatically started, but since the prefetching process for the new page is performed in the background, it does not explicitly appear in the flowchart in FIG.

Hereinafter, the details of the page display processing in step S6-4 in FIG. 7 will be described with reference to FIGS. 8 and 9, including the prefetching processing.

First, in step S7-1 in FIG. 8, it is checked whether or not the page specified by the user is registered in the prefetch text list 30.
If the result of this check is that the page does not exist in the pre-read text list 30, the page currently being displayed is added to this list (step S7-2). After the addition to the list, the process proceeds to step S7-3. On the other hand, if the page already exists in the pre-read text list 30 in step S7-1, the process immediately proceeds to step S7-3.

In step S7-3, the UR of the page specified by the user from the pre-read text list 30
A row (record) having the same URL as L is selected.
This line is called line i in the text.

Next, in step S7-4, the number of accesses (access frequency) in the row i is incremented. This is for counting the number of accesses to the page.

Next, in S7-5, a check is made as to whether or not the completion mark is attached to the row i, in other words, whether or not the completion mark "complete" is set. As a result of the inspection, if “complete” is added, step S7-
The process proceeds to 8 to display a text portion. On the other hand, if “complete” is not set, step S7
The process shifts to -6, and requests the server 22 for text data.

Next, in step S7-7, after obtaining the text data, a completion mark is added to the row i, that is, "complete" is set as the completion mark.

In this way, after the text data is obtained and stored in the cache memory 24, step S
At 7-8, a text portion is displayed.

Next, in step S7-9, a link to a component included in the currently displayed text portion is extracted, and a current page component list is created. FIG. 9 is an explanatory diagram of an example of specific contents of the current page component list 34. If the currently displayed page is the page A in FIG. 5, for example, the image data e. gif is a component of this current page. The number N of elements in the current page constituent element list 34 differs depending on each page.

In step S7-10, 1 is set to the loop variable n.

First, in step S7-11, it is checked whether or not the component n (n is an integer from 1 to N) currently exists in the page component list 34. If there is, the process proceeds directly to step S7-13. If it does not exist, the component n is added to the look-ahead configuration list in step S7-10, and
The process shifts to 7-13.

Next, in step S7-13, it is checked whether or not the completion mark is set for the currently focused component n in the prefetch component list 32. Here, the completion mark is set when “complete” is set as the completion mark. As a result of this inspection, when “complete” is set, the process proceeds to step S7-16, and the component is displayed. On the other hand, if “complete” is not set as the completion mark, the process proceeds to step S7-14, and requests the server 22 for the data of the component. In step S7-15, the completion mark is set after the data of the component n is stored in the cache memory 24. Specifically, “complete” is set as the completion mark. After the completion mark is set, the process proceeds to step S7-16, and the component n is displayed.

Next, in step S7-17, the loop variable n is incremented. That is, 1 is added to the loop variable n.

Next, in step S7-18, it is checked whether or not the loop variable n is larger than N. N is the number of components included in the current page. Therefore, this check is a check whether or not the display has been completed for all the constituent elements. If the loop variable n is larger than N, it is determined that the display has been completed for all the constituent elements, and the process proceeds to the next processing.

In the example shown in FIG. 8, an example in which the processing is sequentially performed for each component has been described. However, it is preferable to perform the processing in parallel for a plurality of components. By simultaneously requesting a plurality of components for content as described above, more efficient prefetch processing can be realized.

The process described above with reference to FIG. 8 is an operation of displaying a browse page specified by a user, apart from the configuration of a list in the middle, and the operation itself is the same as that of the conventional client 10. Operation. The processing subsequent to the display processing of FIG. 8 is shown in the flowchart of FIG.

First, in step S9-1 of the flowchart of FIG. 10, the next page link included in the currently displayed page is extracted. Then, the current page prefetch list 36 is created based on this link. This current page prefetch list 36 has already been described with reference to FIG. Here, the number of lines (that is, the number of records) of the current page prefetch list 36 is represented by M. FIG.
5 shows the contents of the current page look-ahead list 36 in the case where the currently displayed page is the page A in FIG. As shown in FIG. 5, the currently displayed page A is provided with a link to page B and a link to page C as next page links. Therefore, when the current page is the page A as shown in FIG. html and C.I. htm
The URLs of the text portions of the two pages 1 are shown. In this case, the number of rows M is 2.

Next, in step S9-2, 1 is substituted for the loop variable m.

Based on the loop variable m, a loop process from step S9-3 to step S9-7 described below is performed.

First, in S9-3, the next page m
It is checked whether (m is a positive integer from 1 to M) exists in the look-ahead text list 30. As a result of this inspection, if there is no such page, the next page m is added to the pre-read text list 30 in step S9-4, and then the process proceeds to step S9-5. Step S9-
3, the next page m is already a pre-read text list 30
If it exists, the process directly proceeds to step S9-5.

In step S9-5, a next page mark is given to a page in the prefetched text list 30 that corresponds to the page next to the currently displayed page. Here, the next page means a page linked from the currently displayed page (for example, page A) as described above. For example, if the currently displayed page is the page A, the page linked from the page A is the page B as shown in FIG.
Page and page C. Therefore, in this case, FIG.
As shown in FIG. html and C.I. ht
The next page mark is added to the two text portions of ml.

In the step S9-6, the loop variable m
Is incremented. That is, 1 is added to this m.

Next, in step S9-7, it is checked whether or not the loop variable m is larger than M. With this check, it is checked whether or not the loop processing has been completed for all of the next pages currently stored in the page prefetch list 36. If m is not larger than M as a result of the inspection, the processing has not been completed for all of the next pages currently stored in the page prefetch list 36, so the processing shifts to step S9-3 and the loop processing continues. Is done.

On the other hand, if m is larger than M in step S9-7, it is determined that the loop process has been completed for all the next pages, and the process proceeds to the next step S9-8.

In the example shown in FIG. 10, the processing is sequentially performed for each next page. However, it is preferable to perform the processing for all the next pages in parallel.

In step S9-8, FIG.
Among the lines stored in the look-ahead text list 30 shown in FIG. 3, the lines with the next page mark have the priority Y (Y = Y = Y) in descending order of the access frequency (access frequency).
1, 2, 3,... M) are set. Here, Y = 1 means the highest priority. When the priority column is “0” as described above, it means that the page corresponding to the text of the line is the currently displayed page.

In the step S9-9, the loop variable y
Is set to 1. Thereafter, the loop processing from step S9-10 to step S9-15 is repeated based on the loop variable y. According to this loop variable, the following processing is performed in descending order of priority. Of course, as described above, when the priorities for all the pages are the same, it is preferable to perform the processing for all the pages in parallel.

In the flowchart shown in FIG. 10, the processing is sequentially performed for each page in the order of priority.

First, in step S9-10, a line m with priority = y is found from the prefetch text list 30. The y is an initial value of 1 as described above, and is a loop variable that is sequentially incremented as described later. Therefore, in this step S9-10, the line with the highest priority is found first. Then, each time the processing proceeds, the processing for the lines of lower priority is performed one after another.

In step S9-11, it is checked whether or not the found row m is marked as completed. As a result of this inspection, when the completion mark is attached (that is, the completion mark is “complete”), the process proceeds to step S <b> 9-14. On the other hand, if the completion mark has not been added, the process proceeds to step S9-12, and the text data of the page is requested to the server 22. If text data is obtained as a result of this request, the contents are stored in the cache memory 24, and a completion mark is added to the line m in step S9-13. Specifically, "complete" is given as a completion mark.

In step S9-14, the loop variable y is incremented. That is, 1 is added to the loop variable y.

In step S9-15, it is checked whether or not the loop variable y is larger than M. If not, step S9-1 is performed to continue the loop processing.
The process returns to 0. On the other hand, when the loop variable y is larger than M, the processing as shown in FIG. In the flowchart shown in FIG. 10 described above, prefetching was performed on the text portion of each page.
As described above, in this embodiment, all the data of the text portion is first pre-read for all the pages.

After all the prefetching of the text portion is completed, prefetching of the components (image data and audio data) of each page is performed. FIG. 11 is a flowchart showing the operation related to the prefetching of the components.

First, in step S10-0, 1 is set to the loop variable y.

Next, in step S 10-1, the line m whose priority = y is selected from the prefetched text list 30.
Find

In step S10-2, the row m
By extracting the text part of the above and analyzing the text file, the link to the component included in the page is extracted. The next page component list 38 is created based on this link. In the text, the number of elements in the next page component list 38 is represented by N.

Next, in step S10-3, 1 is set to n as a loop variable. This n loop variable is a loop variable for counting constituent elements, and the above-described loop variable y is a loop variable representing priority.

In step S10-4, a check is first made to determine whether or not the component n (the initial value of n is 1 and is an integer up to N) exists in the prefetch component list 32. As a result of this inspection, if there is no component n
Is added to the prefetch component list 32. After this addition, the process moves to step S10-6. On the other hand, if the component n already exists in the prefetch component list 32 in step S10-4, the process immediately proceeds to step S10-4.
The process shifts to 10-6.

In step S10-6, it is checked whether or not the completion mark is set to the component n in the prefetch component list 32, that is, whether or not the completion mark "complete" is set. . As a result of this test,
If it has been set, the process proceeds to step S10-9. On the other hand, if the completion mark is not set,
The process proceeds to step S10-7, and requests the component data of the component n from the server 22. Upon receiving the component data from the server 22 in response to this request,
It is stored in the cache memory 24, and the corresponding completion mark is set in step S10-8. That is, "complete" is set as the completion mark. After acquiring the component data in this way, step S
The process shifts to 10-9.

In step S10-9, the loop variable n is incremented, that is, 1 is added.

Next, in step S10-10, it is checked whether or not the loop variable n is larger than N. This check is a check whether or not the processing has been completed for all the constituent elements n. If n is not larger than N, the processing shifts to step S10-4, and the same processing is repeated. On the other hand, if n is larger than N, the next step S
The process shifts to 10-11.

In FIG. 11 of the present embodiment,
Although the processing is performed on each component n in order, it is also preferable to perform the processing on all the components N in parallel.

In step S10-11, the loop variable y representing the priority is incremented. That is, 1 is added to y.

In step S10-12, it is checked whether or not y is larger than M. If y is not larger, it is determined that prefetching has not yet been completed for all the components having the priorities. Step S10-
The process proceeds to 1 and the same process is repeated. On the other hand, if y is larger than M, it is determined that the prefetching has been completed for all the constituent elements, and the prefetching process is completely terminated.

As described above, in the present embodiment, when displaying the page specified by the user, the data of the text portion and the component of the page linked from the newly displayed page are displayed. Is read ahead from the server 22 and stored in the cache memory 24. For this reason, it is possible to perform prefetching using the idle time of the network, and when the browse page specified by the user is included in the prefetched data, the cache memory can be used without requesting the server 22 for data. The page can be displayed to the user using the data inside 24. Therefore, it is possible to realize a content prefetching method capable of promptly responding to a user's instruction.

Although the present embodiment has been described with reference to the flowchart in which instructions are given in order for each page and processing is performed in order for each component, a plurality of pages or a plurality of components are used. It is also preferable to perform the treatment simultaneously. As described above, by performing the parallel processing, the prefetch can be performed more quickly.

[0161]

As described above, according to the present invention, a page to which a link is set is read ahead from a page displayed by a user's instruction, so that the user can use the idle time of the network effectively. A content look-ahead method that can quickly respond to an instruction from a user is realized.

According to the present invention, since the expiration date of the page is checked, the contents of the pre-read page can be kept fresh.

Further, according to the present invention, the expiration date is set individually according to the update frequency of each page, so that a more flexible content prefetching method is realized.

Further, according to the present invention, the expiration date is individually set according to the update frequency of the components of each page, so that a more flexible content prefetching method is realized.

Further, according to the present invention, since the pre-reading is performed in parallel, the idle time of the network can be more effectively utilized.

Further, according to the present invention, when a plurality of contents are read out in parallel and the user gives an instruction to view any of the pages, the instruction is given. Since only the reading of the read page is continued, a content prefetching method capable of responding to a user's instruction more quickly is realized.

Further, according to the present invention, since the contents of the text data are read out earlier than the image data other than the text data, only the rough display contents of the page can be quickly displayed.

Further, according to the present invention, reading of linked pages is not performed in parallel, but is performed sequentially for each page. Therefore, prefetching can be reliably performed for each page.

Further, according to the present invention, when reading out (pre-reading) of a content, and when a user gives an instruction of a page to be browsed, reading of the designated page is started, and pre-reading is started. Operation is interrupted. Therefore, a content prefetching method that can quickly respond to a user's instruction is realized.

Further, according to the present invention, the reading of the interrupted page is resumed when the page linked to the page is displayed again to the user. Therefore, the reading of the content can be performed more efficiently.

When the reading of the contents of the page is resumed, the expiration date of the partially read data is checked, and if the expiration date has passed, the halfway data is discarded, and the data is read. Can be held from the beginning, so that the data can be kept fresher.

Further, according to the present invention, when reading each page in order, if the order is determined based on the priority of each page, a page with a higher priority can be read with a high probability.

According to the present invention, the linked pages are divided into a plurality of groups according to the priority. And
Because we decided to read multiple pages in parallel for each group, we did not increase network load significantly,
In addition, it is possible to prefetch the contents of the page efficiently.

Further, according to the present invention, by performing the above-described grouping based on the priority, it is possible to prefetch a page having a higher priority with a higher probability, and thus to respond quickly to a user's instruction. A content prefetching method is realized.

In addition, according to the present invention, the priority is determined by the frequency of access to the page.
Therefore, more efficient prefetching can be performed by performing prefetching more quickly for pages that are frequently accessed.

In addition, according to the present invention, it is a simple method to express this access frequency in particular by the number of accesses. The number of accesses can be easily measured by incrementing the counter in the list each time the page is accessed. Therefore, this access frequency can be measured by a simple method.

Further, according to the present invention, content prefetching is performed only when the prefetching operation is permitted. Thus, when it is not desired to increase the load on the network, the read operation is prohibited, so that the load on the network can be appropriately adjusted.

[Brief description of the drawings]

FIG. 1 is a sequence diagram illustrating transmission and reception of data between a client and a server according to the present embodiment.

FIG. 2 is a flowchart illustrating an operation of a client according to the present embodiment.

FIG. 3 is an explanatory diagram showing an example of specific contents of a prefetch text list and a prefetch composition list.

4 is an explanatory diagram showing specific contents of a current page prefetch list 36. FIG.

FIG. 5 is an explanatory diagram illustrating an example of a link relationship between pages.

FIG. 6 is an explanatory diagram showing specific contents of a prefetch text list 3 and a prefetch component list immediately after a C page is displayed.

FIG. 7 is a general flowchart illustrating an operation of the client 20 according to the present embodiment.

FIG. 8 is a flowchart illustrating a detailed operation of a page display process in FIG. 7;

FIG. 9 is an explanatory diagram showing specific contents of a current page component list.

FIG. 10 is a flowchart illustrating a detailed operation of a page display process in FIG. 7;

FIG. 11 is a flowchart illustrating a detailed operation of a page display process in FIG. 7;

FIG. 12 is an explanatory diagram showing specific contents of a next page component list 38.

FIG. 13 is a sequence diagram illustrating a state of data transmission and reception between a conventional client and a server.

[Explanation of symbols]

10 clients, 12 servers, 14a, 14b
Page request, 16a, 16b page content, 2
0 client, 22 server, 24 cache memory, 26a, 26b, 26c page request, 28a,
28b, 28c page content, 30 prefetch text, 32 prefetch component list, 36 current page prefetch list, 34 current page component list, 38
Next page component list, 100 page instruction receiving step, 102 inspection step, 104 cache memory read step, 106 server read step, 108 linked page prefetch step.

[Procedure amendment]

[Submission date] January 14, 1999

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0010

[Correction method] Change

[Correction contents]

[0010]

According to the present invention, there is provided a page instruction receiving step of receiving from a user an instruction of a page to be viewed by the user, and a page indicated by the instruction received in the page instruction receiving step is stored in a cache memory. A checking step of checking whether the specified page exists in the cache memory, and in the checking step, if the specified page does not exist in the cache memory, requesting the specified page to a server; In a server reading step of displaying a page transmitted from a server to a user, and in the checking step, when the specified page exists in the cache memory, the content of the specified page is read from the cache memory.
Out read, and the cache memory read step of displaying the read-out page to the user, said examine the content of the page that is displayed to the user at the server read step or the cache memory read step, the link is Zhang from the display pages its dependent content landing page request to the server, see contains a landing page prefetching storing the contents of the destination pages that are sent to the cache memory from the result of the request the server, the Landing Page Look-Ahead
The step of reading is the text in the content of the landing page.
Text data must precede non-text data
It is characterized by reading .

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0011

[Correction method] Change

[Correction contents]

[0011] Further, the present invention is directed to a page which a user wants to view.
Page instruction receiving step of receiving a page instruction from the user.
Received in the page instruction receiving step.
Whether the page indicated by the instruction exists in the cache memory
An inspection step for inspecting whether or not
The specified page exists in the cache memory.
If not, send the specified page to the server
Request, and as a result of the request, the page transmitted from the server
A server readout step for displaying a page to the user, and
In the inspection step, the specified page is
If it exists in the cache memory,
Read the content of the specified page from the
Cache memory for displaying read pages to users
A reading step, and the server reading step or
A user in the cache memory reading step
Inspect the content of the page displayed on the
Link destination page link from the page
Content to the server, and as a result of the request,
The content of the landing page sent from the server
Prefetching linked pages stored in the cache memory
And the step of prefetching the linked page
Reads the contents of multiple linked pages
It is characterized by being executed on a line .

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0012

[Correction method] Change

[Correction contents]

[0012] Further, the present invention provides the above-mentioned linked page prefetching.
The reading step is for the user to read out the content.
Viewed one of multiple landing pages
Link when the instruction was issued
Continue reading the previous page, and select the page where the instruction was issued.
The reading of the linked pages other than the page is interrupted .

[Procedure amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0013

[Correction method] Change

[Correction contents]

Further, according to the present invention, a page which a user wants to view is provided.
Page instruction receiving step of receiving a page instruction from the user.
Received in the page instruction receiving step.
Whether the page indicated by the instruction exists in the cache memory
An inspection step for inspecting whether or not
The specified page exists in the cache memory.
If not, send the specified page to the server
Request, and as a result of the request, the page transmitted from the server
A server readout step for displaying a page to the user, and
In the inspection step, the specified page is
If it exists in the cache memory,
Read the content of the specified page from the
Cache memory for displaying read pages to users
A reading step, and the server reading step or
A user in the cache memory reading step
Inspect the content of the page displayed on the
Link destination page link from the page
Content to the server, and as a result of the request,
The content of the landing page sent from the server
Prefetching linked pages stored in the cache memory
And the step of prefetching the linked page
Reads the content of multiple landing pages,
Execute in order for each linked page, and
Before the reading of the contents of the page is completed, the user
When the user gives an instruction on the page to be viewed, the link
The reading of the previous page is interrupted .

[Procedure amendment 6]

[Document name to be amended] Statement

[Correction target item name] 0014

[Correction method] Change

[Correction contents]

[0014] In addition, the present invention, the landing page look-ahead step, the user, phosphorus to the landing page
When browsing to a linked source page that has been marked
Displays the content of the link source page to the user
And the reading of the linked page whose reading was interrupted
Dispensing is resumed .

[Procedure amendment 7]

[Document name to be amended] Statement

[Correction target item name] 0015

[Correction method] Change

[Correction contents]

Further, in the present invention, the step of prefetching the linked page includes the step of preliminarily reading the linked page.
When resuming the reading of the data, the
Check the expiration date of the read result in the middle of the destination page,
If the expiration date has passed, the landing page
If the expiration date has not expired
Is the part where reading of the linked page was interrupted
The reading is continued from this point.

[Procedure amendment 8]

[Document name to be amended] Statement

[Correction target item name] 0016

[Correction method] Change

[Correction contents]

Further, according to the present invention, a page which a user wants to view is provided.
Page instruction receiving step of receiving a page instruction from the user.
Received in the page instruction receiving step.
Whether the page indicated by the instruction exists in the cache memory
An inspection step for inspecting whether or not
The specified page exists in the cache memory.
If not, send the specified page to the server
Request, and as a result of the request, the page transmitted from the server
A server readout step for displaying a page to the user, and
In the inspection step, the specified page is
If it exists in the cache memory,
Read the content of the specified page from the
Cache memory for displaying read pages to users
A reading step, and the server reading step or
A user in the cache memory reading step
Inspect the content of the page displayed on the
Link destination page link from the page
Content to the server, and as a result of the request,
The content of the landing page sent from the server
Prefetching linked pages stored in the cache memory
And the step of prefetching the linked page
Will read the content of multiple landing pages
Execute in order for each linked page
The order of reading the page content is determined by the client
Determined by the priority of each landing page determined by the side
It is characterized by doing.

[Procedure amendment 9]

[Document name to be amended] Statement

[Correction target item name] 0017

[Correction method] Change

[Correction contents]

Further, the present invention provides a method for displaying a page which a user wants to view.
Page instruction receiving step of receiving a page instruction from the user.
Received in the page instruction receiving step.
Whether the page indicated by the instruction exists in the cache memory
An inspection step for inspecting whether or not
The specified page exists in the cache memory.
If not, send the specified page to the server
Request, and as a result of the request, the page transmitted from the server
A server readout step for displaying a page to the user, and
In the inspection step, the specified page is
If it exists in the cache memory,
Read the content of the specified page from the
Cache memory for displaying read pages to users
A reading step, and the server reading step or
A user in the cache memory reading step
Inspect the content of the page displayed on the
Link destination page link from the page
Content to the server, and as a result of the request,
The content of the landing page sent from the server
Prefetching linked pages stored in the cache memory
And the step of prefetching the linked page
Classifies multiple landing pages into a predetermined number of groups.
Client splits the content of multiple landing pages
Are read out in order for each of the groups,
Read the content of the landing page in each group
Extrusion is performed for multiple linked pages included in the group.
This is characterized in that it is executed in parallel to the

[Procedure amendment 10]

[Document name to be amended] Statement

[Correction target item name] 0018

[Correction method] Change

[Correction contents]

[0018] In addition, the present invention, the landing page look-ahead step, a plurality of the landing page, client
Predetermined in order of priority given to landing page
Grouping by the number of pages
It is characterized by forming a loop .

[Procedure amendment 11]

[Document name to be amended] Statement

[Correction target item name] 0019

[Correction method] Change

[Correction contents]

In the present invention, the step of prefetching the linked page includes the step of assigning the priority to a past page.
It is characterized by being determined by the access frequency .

[Procedure amendment 12]

[Document name to be amended] Statement

[Correction target item name] 0020

[Correction method] Change

[Correction contents]

Further, according to the present invention, the access frequency may be
It is characterized by the number of accesses .

[Procedure amendment 13]

[Document name to be amended] Statement

[Correction target item name] 0021

[Correction method] Deleted

[Procedure amendment 14]

[Document name to be amended] Statement

[Correction target item name] 0022

[Correction method] Deleted

[Procedure amendment 15]

[Document name to be amended] Statement

[Correction target item name] 0023

[Correction method] Deleted

[Procedure amendment 16]

[Document name to be amended] Statement

[Correction target item name] 0024

[Correction method] Deleted

[Procedure amendment 17]

[Document name to be amended] Statement

[Correction target item name] 0025

[Correction method] Deleted

[Procedure amendment 18]

[Document name to be amended] Statement

[Correction target item name] 0026

[Correction method] Deleted

[Procedure amendment 19]

[Document name to be amended] Statement

[Correction target item name] 0028

[Correction method] Change

[Correction contents]

Embodiment 1 FIG. 1 is a sequence diagram illustrating the operation of the content prefetching method according to the preferred embodiment of the present invention. In this figure,
Similar to FIG. 1 3 representing the conventional method with the client 20,
The transmission and reception of data with the server 22 are shown.

[Procedure amendment 20]

[Document name to be amended] Statement

[Correction target item name] 0033

[Correction method] Change

[Correction contents]

Next, the client 20 checks the priority of the linked page. To set the priority for this landing page, refer to the past history, etc.
For those that are likely to be subsequently viewed within each landing page it is performed so as to set a higher priority. The specific operation of setting the priority will be described later in detail.

[Procedure amendment 21]

[Document name to be amended] Statement

[Correction target item name] 0043

[Correction method] Change

[Correction contents]

In particular, when the link destination page is read out in parallel, if the user gives an instruction of a new browse page during the prefetching, the prefetching of all pages is performed.
But there is no page read-ahead even during the Application has been completed. Therefore, only the prefetching of the browse page designated by the user is continued, and the prefetching of the other pages is interrupted.

[Procedure amendment 22]

[Document name to be amended] Statement

[Correction target item name] 0056

[Correction method] Change

[Correction contents]

[0056] That is, the read-ahead in parallel with the first high priority this landing page. After the prefetching of the ten linked pages is completed, the prefetching is performed in parallel on the ten linked pages with the next highest priority. As described above, if the linked pages are grouped by a predetermined number (for example, 10) and the content of the page is prefetched in parallel for each group,
Effective prefetching can be performed while preventing the network load from becoming excessive.

[Procedure amendment 23]

[Document name to be amended] Statement

[Correction target item name] 0058

[Correction method] Change

[Correction contents]

First, in a page instruction receiving step 100, the client 20 receives an instruction of a page to be browsed from a user. This indication is also if you enter the URL of the page the user directly, there is also a case where an instruction by clicking on the link button from a predetermined page is carried out.

[Procedure amendment 24]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

Further, the client 20 adds a link destination page linked from the currently displayed page to the current page look-ahead list. The currently diagram of the stored contents of the page prefetch Listing 3 6 shown in FIG. As shown in FIG. 4, the current page prefetch Listing 3 6 is a list obtained by arranging landing page URL. Two pages are currently stored in the page prefetch list 3-6 in FIG. 4.

[Procedure amendment 25]

[Document name to be amended] Statement

[Correction target item name] 0162

[Correction method] Change

[Correction contents]

Also, the content of the text data is
Read before image data other than text data
Configuration so that only the rough contents of the page
It can be displayed quickly .

[Procedure amendment 26]

[Document name to be amended] Statement

[Correction target item name] 0163

[Correction method] Change

[Correction contents]

According to the present invention, read-ahead read
Is performed in parallel, so when the network is idle
The time can be used more effectively .

[Procedure amendment 27]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

Also, according to the present invention, a plurality of contents
If reading is performed in parallel, the user
When you give an instruction to view one of the pages
In this case, only reading of the indicated page was continued.
As a result, a content prefetching method that can respond to a user's instruction more quickly is realized.

[Procedure amendment 28]

[Document name to be amended] Statement

[Correction target item name] 0165

[Correction method] Change

[Correction contents]

Also, according to the present invention, the link destination page
Reading is performed in parallel for each page instead of in parallel.
To do. For this reason, read ahead for each page
Can be done .

[Procedure amendment 29]

[Document name to be amended] Statement

[Correction target item name] 0166

[Correction method] Change

[Correction contents]

Also, according to the present invention, an interrupted page
Is linked to the page
Resumes when the page is presented to the user again
It is. As a result, reading content more efficiently
Can be done .

[Procedure amendment 30]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

Also, according to the present invention, the content of a page
When resuming the reading of data,
Inspect the expiration date and if it has expired
Discard data up to the middle of
, The data can be kept with fresher contents .

[Procedure amendment 31]

[Document name to be amended] Statement

[Correction target item name] 0168

[Correction method] Change

[Correction contents]

Also, according to the present invention, each page is sequentially
When reading, the order of priority depends on the priority of each page.
Once the number is determined, pages with high priority are read with high probability
Can be put out .

[Procedure amendment 32]

[Document name to be amended] Statement

[Correction target item name]

[Correction method] Change

[Correction contents]

Also, according to the present invention, the linked page
Divided into multiple groups according to priority. And
I decided to read multiple pages for each group in parallel
Therefore, without significantly increasing the load on the network,
It is possible to prefetch the content of the page efficiently and
I can .

[Procedure amendment 33]

[Document name to be amended] Statement

[Correction target item name] 0170

[Correction method] Change

[Correction contents]

Further, according to the present invention, the above grouping
By priority, high priority pages
Can be read ahead with a higher probability,
That can respond quickly to the instructions of
A look-ahead method is realized .

[Procedure amendment 34]

[Document name to be amended] Statement

[Correction target item name] 0171

[Correction method] Change

[Correction contents]

According to the present invention, the priority is
Is determined by the frequency of access to the page.
Therefore, faster for frequently accessed pages
By performing prefetching in advance, more efficient prefetching can be performed.

[Procedure amendment 35]

[Document name to be amended] Statement

[Correction target item name] 0172

[Correction method] Change

[Correction contents]

According to the present invention, in particular,
It is a simple method to express the frequency by the number of accesses. This
Access to the page
Each time the counter in the list is incremented
Thus, the measurement can be easily performed. Obedience
Thus, the access frequency can be measured by a simple method .

[Procedure amendment 36]

[Document name to be amended] Statement

[Correction target item name] 0173

[Correction method] Deleted

[Procedure amendment 37]

[Document name to be amended] Statement

[Correction target item name] 0174

[Correction method] Deleted

[Procedure amendment 38]

[Document name to be amended] Statement

[Correction target item name] 0175

[Correction method] Deleted

[Procedure amendment 39]

[Document name to be amended] Statement

[Correction target item name] 0176

[Correction method] Deleted

[Procedure amendment 40]

[Document name to be amended] Statement

[Correction target item name] 0177

[Correction method] Deleted

Claims (17)

[Claims] 1. A page instruction receiving step of receiving an instruction of a page that a user wants to view from the user, and checking whether a page indicated by the instruction received in the page instruction receiving step exists in a cache memory. In the checking step, when the specified page does not exist in the cache memory, the specified page is requested to a server, and the page transmitted from the server as a result of the request is determined. In the server reading step to be displayed to the user, and in the checking step, when the specified page exists in the cache memory, the content of the specified page is read from the cache memory, and the read page is used. Reading the cache memory to be displayed to the user, Inspecting the content of the page displayed to the user in the step of reading out the cache memory or the step of reading out the cache memory, requesting the server for the content of the linked page to which the displayed page is linked, and the result of the request. A link destination page prefetching step of storing the content of the link destination page transmitted from the server in the cache memory. 2. The cache memory reading step examines an expiration date of the content of the indicated page in the cache memory, and only when the expiration date has not passed, the cache memory reads out the instruction from the cache memory. 2. The content prefetching method according to claim 1, wherein the content of the page is read and the read page is displayed to a user. 3. The content prefetching method according to claim 2, wherein the expiration date is determined independently for each page. 4. The content prefetching method according to claim 2, wherein the expiration date is determined separately and independently for each component of each page. 5. The content prefetching method according to claim 1, wherein in the step of prefetching linked pages, reading of contents of a plurality of linked pages is executed in parallel. 6. The linked page prefetching step includes: when a user issues an instruction to view any of a plurality of linked pages from which the content is being read, the instruction is issued. 6. The content prefetching method according to claim 5, wherein the reading of the linked page is continued, and the reading of the linked pages other than the page for which the instruction is issued is interrupted. 7. The linked page pre-reading step, wherein when reading the contents of the plurality of linked pages in parallel, reading the text data in the contents of the plurality of linked pages before the data other than the text data. 6. The content prefetching method according to claim 5, wherein: 8. The linked page prefetching step according to claim 1, wherein the reading of the contents of a plurality of linked pages is sequentially executed for each linked page. Content prefetching method. 9. The linked page pre-reading step includes: interrupting the reading of the linked page when the user gives an instruction of a page to be viewed before the reading of the content of the linked page is completed. 9. The content prefetching method according to claim 8, wherein the content is prefetched. 10. The linked page prefetching step includes: when a user instructs browsing of a linked source page to which a link to the linked page is provided, the user reads the content of the linked source page. 10. The content prefetching method according to claim 6, wherein the content is prefetched and the reading of the linked page whose reading has been interrupted is resumed. 11. The linked page pre-reading step includes: when resuming reading of the linked page whose reading has been interrupted, checking the expiration date of a read result in the middle of the linked page whose reading has been interrupted; When the expiration date has passed, the linked page is read again, and when the expiration date has not passed, the reading is continued from the portion where the reading of the linked page was interrupted. The content prefetching method according to claim 10. 12. The content prefetching method according to claim 8, wherein in the link page prefetching step, the order of reading the contents of the plurality of link pages is determined based on the priority of each link page. 13. The linked page prefetching step divides a plurality of linked pages into a predetermined number of groups, and sequentially reads out the contents of the plurality of linked pages for each of the groups. 5. The content prefetching method according to claim 1, wherein the reading of the content of the linked pages included in the group is executed in parallel for a plurality of linked pages included in the group. 14. The linked page prefetching step forms the group by grouping a plurality of linked pages in a predetermined number of pages in the order of priority assigned to the linked pages. 14. The content prefetching method according to claim 13, wherein: 15. The content prefetching method according to claim 12, wherein in the link destination page prefetching step, the priority is determined based on a frequency of access to a page in the past. 16. The content prefetching method according to claim 15, wherein the access frequency is an access frequency. 17. The content prefetching method according to claim 1, wherein in the linked page reading step, the content of the linked page is prefetched only when the prefetch operation is permitted.