Electrical computers and digital processing systems: multicomput – Computer-to-computer session/connection establishing
Reexamination Certificate
1997-11-18
2001-03-06
Rinehart, Mark (Department: 2756)
Electrical computers and digital processing systems: multicomput
Computer-to-computer session/connection establishing
C709S223000, C709S224000, C709S226000
Reexamination Certificate
active
06199111
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a network computing system for updating and adding data in a distributed client-server system through a network, and its method.
2. Description of the Related Art
In the conventional information system it was common to process a single huge database on a transaction by transaction basis. However, today, with the development of a computing system and networking, it has become necessary to construct an information system in which data is frequently updated and added among widely distributed systems. With the fusion between a LAN (Local Area Network) and a global network, data processing in a form of network computing centering around network is also being realized.
FIG. 1A
is a block diagram of the conventional client-server system. The system in
FIG. 1A
comprises clients
1
, application servers
5
(application servers A, B and C) and a database
6
. The application servers A, B and C share the database
6
among them, and are provided with a server A exclusive communication module
2
, a server B exclusive communication module
3
and a server C exclusive communication module
4
respectively, in order to communicate with the clients
1
.
A client
1
is provided with one of the server A exclusive communication module
2
, the server B exclusive communication module
3
, and the server C exclusive communication module
4
, and updates the database
6
, etc. connected to a specific application server
5
having a corresponding communication module.
Furthermore, with the recent huge increase in size of the database
6
and the decentralization of systems, the database
6
has been distributed as shown in FIG.
1
B. In the system of
FIG. 1B
, each of the application servers A, B and C manages an individual database
6
. Also in this system, the clients
1
are connected only to specific application servers
5
using one of the server A exclusive communication module
2
, the server B exclusive communication module
3
, and the server C exclusive communication module
4
.
For this reason, when a client
1
accesses the data in the database
6
of a server
5
different from the application server
5
of the connected party, replication of data is made between the relevant databases
6
, or a remote procedure call (RPC) is made between the application servers
5
. In the case of the replication, data is copied between different databases
6
, and in the case of an RPC, a function call is made and the necessary data are exchanged between the application servers
5
.
What is common to the systems shown in
FIGS. 1A and 1B
is that a client
1
is designed to be connected with a specific application server
5
by using communication modules
2
,
3
and
4
.
On the Internet, which has recently become popular, a world wide web (WWW) has been developed and is widely used as a means for sourcing information. The characteristics of a WWW are that any client (WWW browser) can be connected to any server (WWW server) by a common protocol called hypertext transfer protocol (HTTP). A WWW is provided with a hyperlink function for embedding a link to a hypertext markup language (HTML) document in a server, in a HTML document in another server. Here, a HTML document corresponds to a file sent from a server to a client.
FIG. 1C
is a flowchart showing the operation of a client and a server on a WWW. When the client starts operating and the user clicks a link on the screen (step S
1
), the client is connected to the corresponding server using the HTTP (step S
2
).
Thus, the corresponding HTML document is sent from the server to the client (step S
3
), and the connection with the server is cut (step S
4
) and the data of the HTML document is displayed on the screen (step S
5
). When the user clicks a displayed link on the screen again, the operation of steps S
1
to S
5
is repeated.
However, the conventional information system described above has the following problems.
In such a conventional client-server system as shown in
FIGS. 1A
or
1
B, since a client
1
can be connected with a specific server
5
only, the function of other servers
5
distributed on the network cannot be directly used.
For this reason the server
5
of the connected party of the client
1
has to be connected with another server
5
and has to ask for a necessary process, which generates many transactions between the servers
5
. When replication is needed between databases
6
, too, this process becomes complicated in the same way.
In this way, since the control of the processes between servers
5
and databases
6
becomes complicated, it is impossible to construct such a large-scale distributed system that allows servers
5
and databases
6
to be spread all over the world, such as in a WWW.
The conventional WWW also has a problem that it is provided with only a simple function of sending an HTML document on a client's request. Particularly, since its session is cut after every transmission of one page in HTTP, the communicating state (connecting state) cannot be held. Accordingly, it is difficult to develop a client-server system for performing a real interaction only with the basic functions of a WWW.
Furthermore, since in a WWW the creation/modification of a page and link is left to each individual, the entire configuration of a WWW cannot be determined. For example, it has the following problems.
1. It is difficult to create the entity data (object) of a new page on an arbitrary distributed server.
2. In registering a new object the link from an existing object cannot be updated/managed.
3. When deleting an object, a destination of a link is often lost (dead link).
4. Although a program such as a WWW robot is developed as a means for retrieving information desired by a user, a search using a keyword must be performed for all objects, because there is no linking configuration between the objects.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a network computing system in which the update and addition of data can be easily made, in a distributed client-server system through a network and its method.
The network computing system of the present invention manages distributed data in a distributed system through a communication network.
In the first aspect of the present invention a client side is provided with a common communication unit and a connection management unit. The common communication unit controls the communication between a client and a plurality of servers on the network and can access the plurality of servers commonly. When a client is connected to the first server out of the plurality of servers, the connection management unit connects the client to the second server out of the plurality of servers, if necessary, while holding the connecting state of the client with the first server, and enables the client to receive services from the second server.
The common communication unit is not for communicating with a specific server as it is conventionally, and it can communicate with all the servers on the network commonly. For example, it is assumed that it has become necessary to ask the second server for a process when a client is connected to the first server under the control of the common connection unit. In this case the client is connected to the second server under the control of the connection management unit while it is still connected with the first server, and the client receives services from the second server.
In the second aspect of the present invention a server side is also provided with a common communication unit and a connection management unit. The common communication unit controls the communication between the first server and a client on the network, and receives an access from the client. The connection management unit connects the client to the second server, if necessary, while it holds the connecting state of the client with the first server, and sends out link information for receiving services from the second server to the client.
The common communica
Hara Hirotaka
Kanaya Nobuyuki
Matsutsuka Takahide
Okubo Takao
Uehara Sanya
Fujitsu Limited
Rinehart Mark
Romero Almari
Staas & Halsey , LLP
LandOfFree
Client-led network computing system and its method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Client-led network computing system and its method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Client-led network computing system and its method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2499508