WO2005119436A2

WO2005119436A2 - Thin packing method for a thin client terminal

Info

Publication number: WO2005119436A2
Application number: PCT/FR2005/050399
Authority: WO
Inventors: Damien Mercier
Original assignee: Insia
Priority date: 2004-06-01
Filing date: 2005-06-01
Publication date: 2005-12-15
Also published as: WO2005119436A3; EP1759288A2; FR2870953A1; FR2870953B1

Abstract

The invention relates to a method of packing and/or unpacking digital files, which is performed on a device containing, at least, a random access memory (RAM), a processor, a network interface, an I/O interface (input/output) and an operating system (OS) comprising, at least, a kernel and a file system. The inventive method is characterised in that it comprises the following steps, namely: a first step in which a packet is constructed, whereby at least one digital file and command information are incorporated into the packet and comprise an indication of the future location of said file within the file system, a start-up command and a stop command; a step comprising the compression or reduction of the size of the packet; a step in which the packet is transmitted or copied so that it can be accessed by the operating system of the device if it was not already accessible; a packet start step consisting in invoking command information contained in the above-mentioned command information; a step consisting in copying the digital file(s); and a later step in which the digital file is used by associating a means belonging to the operating system and at least one piece of information contained in the command information.

Description

LIGHT PACKAGE FOR LIGHT CLIENT TERMINAL

The present invention relates to the field of shaping digital computer files.

The present invention relates more particularly to the packaging and unpacking of digital files in a system architecture comprising, preferably, thin clients. In general, this involves providing applications to thin clients in the form of ready-to-install packages and performing a temporary installation of these applications on said thin clients. Faced with the growing share of IT in the corporate world, the thin client / server architecture, offering characteristics of ease of management, increased performance and security or even lower management costs, has strongly developed, especially in large companies.

Known in the prior art, digital file packaging and unpacking solutions for installing applications on thin clients implementing packages such as that described in PCT patent application WO 00 / 77614 (Sun Microsystems). A package is copied to a local non-volatile medium on the terminal and then executed in order to install the contained application, on said local storage medium. Said application is permanently installed, and started as many times as desired. The disadvantages of such solutions are numerous. These packages are suitable for heavy IT environments that are greedy in computing power, in RAM and very bulky. Certain solutions known in the prior art use the Java computer language (trademark registered by Sun Microsystems). This language is said to be "interpreted". A program written in this language requires, for its execution on the client device or workstation, the operation of a virtual machine previously installed. This virtual machine uses significant resources both in terms of memory and in terms of CPU.

The prior art already knows, by American patent application US 6 574 618 (AppStream), a method and a system making it possible to execute a continuous application on a client system by means of a virtual file system installed at the client and designed to be considered by the operating system as a local storage device containing all the application files required by the application. Before the continuous processing, the application files are divided into mini-sequences, generally corresponding to various parts of the application files, and these mini-sequences are transmitted to the client by the server, preferably by means of an algorithm. predictive to determine an optimal transmission order. After receiving an initial set of mini-sequences, the application is executed from the virtual file system. Loads, in terms of files, transmitted by the operating system to the virtual file system are interpreted to determine which mini-sequences correspond to the load request, then the appropriate data is sent back. In the absence of a required mini sequence, a request for mini sequences is sent to the server and the virtual file system maintains the busy status until the required mini sequences be provided. The execution of the application is dependent on the proper functioning of the network. This prior art method also requires numerous treatments (creation of mini-sequences, acknowledgments on reception, etc.).

The present invention intends to remedy the drawbacks of the prior art by proposing an industrial process in the field of information and communication systems, allowing the temporary installation of an application on a random access memory, for example but not necessarily of the type "Flash" through the use of packages. The applications are kept in “to install” form. Thus, each time the package is started, the application is installed within the operating system. Likewise, each time the package is stopped, the application is uninstalled. Such a package is called a "light package". This solution offers easy administration of applications, increased performance on the part of the terminals since they are not "polluted" by unused applications, a simplification and a reduction in the decibels of said terminals. As the terminal is less stressed, the fans are no longer necessary. In addition, this type of terminal offers robustness when the client terminal suddenly turns off. This solution also goes against current lessons in this technological field, where the constant increase in the size of computer applications, the capacity and performance of components is required.

To this end, the invention relates in its most general sense, to a method of packaging and / or unpacking of digital files implemented on a device comprising, at least, a random access memory (RAM), a processor, a network interface, an I / O interface

(input / output) and an operating system (OS) composed, at least, of a kernel and a file system, said method being characterized in that it comprises: • a first stage of construction of a package comprising incorporating within the package at least one digital file and control information, which includes an indication of the future location of said file in a file system, a start command and a stop command; • a step of compressing or reducing the size of said package; • a step of transmitting or copying the packet to make it accessible by the operating system of said device, if it was not already accessible; • a step of starting the packet consisting in calling control information contained in said control information; • a step of copying the digital file (s); • a subsequent step of using the digital file by associating a means of the operating system and at least one piece of information contained in the order information.

According to a first variant, said device does not include any other memory. According to another variant, said device comprises an additional memory of non-volatile type (ROM).

According to one embodiment, said ROM memory is of the EEPROM (Electrically Erasable Programmable Read-Only Memory) type. Advantageously, said ROM memory is of the “flash” type.

Preferably, said device is a client device. According to a variant, said client device is "light".

Advantageously, said digital file is an application. According to a variant, said digital file is a configuration file.

According to a particular embodiment, said control information, called during the step of starting the packet, is saved during this step. According to a second embodiment, the step of saving said control information, called during the step of starting the packet, is carried out in random access memory during this step.

Advantageously, the operating system is installed on said random access memory. According to a variant, said file system is of the type operating on a RAM type memory and is not simply mounted in memory.

Advantageously, said package includes an identification header. According to a variant, said header comprises at least one digital signature. According to a particular implementation mode, the operating system is based on a UNIX type environment, preferably GNU LINUX or FreeBSD. According to another embodiment, the method further comprises a step of adapting said digital file to a specific configuration of said device. According to a variant, said adaptation step is carried out substantially at the time of the call to the command.

The invention also relates to an information system for implementing the method comprising at least one server and at least one client station, linked together by means of a computer or communication network. According to a first variant, the information system for implementing the method comprises at least two clients, connected to each other by means of a computer or communication network. According to a second variant, the information system for implementing the method comprises at least two servers, linked together by means of a computer or communication network.

The invention also relates to a computer device comprising at least one processor and a memory for implementing the method. The invention will be better understood with the aid of the description, given below for purely explanatory purposes, of an embodiment of the invention, with reference to the appended figures in which: FIG. 1 represents the various technical stages of the industrial method of packaging and unpacking of digital files according to the present invention; and - Figure 2 shows the technical structure of a package of digital files used in the method of the present invention.

A client / server architecture implements at least one thin client and one server. The thin client is connected to the server by a means of communication, such as a corporate network. The thin client user wants to run a business application. According to a particular embodiment, the thin client is composed of a processor, a storage memory (ROM or Read-Only Memory) for example of the “flash” type, a random access memory (RAM or Random Access Memory) ), a network interface and I / O interfaces to connect the keyboard, mouse and screen. Note that the presence of a ROM memory is only optional. It is understood that, in the following description, the tasks implementing the ROM memory could implement the RAM memory. An operating system is installed on said thin client. Said operating system is preferably based on a UNIX type environment, for example GNU Linux, FreeBSD or Solaris (UNIX solution from Sun Microsystems).

The remote server forms the basis of the company's applications (and / or configurations). This centralization increases security and facilitates updates and administration of applications. Applications are put in the form of packages, with reference to FIG. 2. For this purpose, all the digital files necessary for the installation of an application are grouped together, then are compressed into a single file: the archives (3). Sometimes this file may undergo encryption to increase security. The result is usually the largest part of the package. For the operation of the package, this section is associated with the intermediate section of said package: the script (2). This contains command information such as, for example, the future location of the application file (s) from the package in the file system of the thin client operating system, such as a command to start the installation of said application or of a command to stop the installation. This control information can be combined in a script and encrypted so that it is protected. Said file system is a structured organization of data on a storage system, volatile (RAM) or permanent (ROM), allowing the backup and localization of data. This contains a tree structure, that is to say an organization of the data into folders and sub-folders in which the data is stored. The file system chosen will preferably be of the type compatible with a UNIX environment, for example EXT2, EXT3, ReiserFS or TmpFS. The tree structure will preferably correspond to a UNIX type tree structure, with, for example, a "/ pkg" folder in which the light packages will be decompressed. Finally, the header (1) is an optional part, at the start of the package. It contains package identification information, but may also contain a digital signature guaranteeing package integrity security.

In a variant, the package is not application-based but contains configuration files. These configuration files can be used, for example, to change the language of an application. The package may also contain an application and associated configuration files.

The package thus obtained corresponds to a single application. It is then put on the server and made accessible to the thin client.

In one embodiment, the package is then transferred via the network or simply copied to the local storage means of the thin client, for example a hard disk. Protocols such as HTTP, NFS (file sharing on Unix), Samba (file sharing on Windows), FTP, TFTP, SSH, SSL / TLS, can be used to perform this transfer or copy operation. In a variant, the package is stored ephemeral, that is to say as long as the client terminal is on, on the RAM. In another embodiment, the package is already present on the ROM (flash) of the thin client terminal. A version comparison process can then be implemented to check the version of the package and update the package. This process uses digital signatures, for example the MD5 hash. A file containing the digital signatures of the reference lightweight packages is stored on the server. The thin client calculates the digital signatures of its packages, if they are not already contained in the headers of said packages. It then compares said signatures with those contained in the file. If there is a difference in signatures between a package from the thin client and the reference package, the update of this package can be done automatically, i.e. by transferring the most recent version of the server to the client's hard drive. lightweight in place of the old package. In a third embodiment, the package remains on a remote server or machine and nevertheless remains accessible by the thin client.

The thin client operating system has a program for starting and stopping packages. This program resides in the execution of a command to start the package, for example "pkg nom start" and a second command to stop the package, for example "pkg nom stop". When the package starts, the header is read and checked. Then the command to start the package calls up the command information contained in the package script. This script performs a first step which consists of copying said script and its command lines into the memory of the thin client. Then the script copies the digital files from the package to the RAM in the file system of the operating system. The installation of the application is then carried out in a traditional manner on the RAM in the file system of the operating system.

The application thus installed, is used as an application in the traditional sense by executing the main program of the application. When the application is finished, the package is stopped via the program. The installation script is called to obtain the information necessary to uninstall the application. Then all the application data is uninstalled so that only the package in its initial version remains. The system and method according to the invention also allow a user to perform remote processing on the package, that is to say through a communication network. These treatments can for example consist in starting, stopping an application.

The invention is described in the foregoing by way of example. It is understood that a person skilled in the art is able to carry out different variants of the invention without going beyond the scope of the patent.

Claims

1. Method for packaging and / or unpacking digital files implemented on a device comprising, at least, a random access memory (RAM), a processor, a network interface, an I / O interface (input / output) and an operating system (OS) composed, at least, of a kernel and a file system, said method being characterized in that it comprises: • a first stage of construction of a package consisting in incorporating into the within the package at least one digital file and control information, which includes an indication of the future location of said file in a file system, a start command and a stop command; • a step of compressing or reducing the size of said package; • a step of transmitting or copying the packet to make it accessible by the operating system of said device, if it was not already accessible; • a step of starting the packet consisting in calling control information contained in said control information; • a step of copying the digital file (s); • a subsequent step of using the digital file by associating a means of the operating system and at least one piece of information contained in the order information.

2. Method for packaging and / or unpacking digital files according to claim 1, characterized in that said device does not include any other memory.

3. Method for packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that said device comprises an additional memory of non-volatile type (ROM).

4. Method for packaging and / or unpacking digital files according to claim 3, characterized in that said ROM memory is of EEPROM (Electrically Erasable Programmable Read-Only Memory) type.

5. A method of packaging and / or unpacking digital files according to claim 4, characterized in that said EEPROM memory is of the “flash” type.

6. Method for packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that said device is a client device.

7. Method for packaging and / or unpacking digital files according to claim 6, characterized in that said client device is "light".

8. A method of packaging and / or unpacking digital files according to claims 1 to 7, characterized in that said digital file is an application.

9. A method of packaging and / or unpacking digital files according to claims 1 to 7, characterized in that said digital file is a configuration file.

10. Method for packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that said control information, called during the step of starting the packet, is saved during this step.

11. Method for packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that the step of saving said control information, called during the step of starting the package, is carried out in RAM during this step.

12. A method of packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that the operating system is installed on said random access memory.

13. Method for packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that said file system is of the type operating on a RAM type memory and is not simply mounted in memory .

14. A method of packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that said package comprises an identification header.

15. A method of packaging and / or unpacking digital files according to claim 14, characterized in that said header comprises at least one digital signature.

16. Method for packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that the operating system is based on a UNIX type environment, preferably GNU LINUX or FreeBSD.

17. A method of packaging and / or unpacking digital files according to any one of the preceding claims, characterized in that it further comprises a step of adapting said digital file to a specific configuration of said device.

18. A method of packaging and / or unpacking digital files according to claim 17, characterized in that said adaptation step is carried out substantially at the time of the call to the command.

19. Information system for the implementation of the method according to any one of the preceding claims, characterized in that it comprises at least one server and at least one client station, linked together by means of a computer network or communication.

20. Information system for implementing the method according to any one of claims 1 to 17, characterized in that it comprises at least two clients, connected to each other by means of a computer or communication network.

21. Information system for implementing the method according to any one of claims 1 to 17, characterized in that it comprises at least two servers, connected to each other by means of a computer or communication network.

22. Computer device comprising at least one processor and a memory for implementing the method according to at least one of claims 1 to 17.