Method and system for managing resources among different clients for an exclusive use

1. A method for sharing resources of a server among different clients, the server containing a maximum number m of resources among which a predetermined number w of the resources are initialized and kept ready for use for the clients and are contained in a pool, which method comprises the steps of:

releasing at a time T 4 a resource of the server, wherein a status of the resource released being changed from “in use” to “available”;

determining a size of the pool at the time T 4 ; and

if the pool contains p resources at the time T 4 , wherein m≧p>w, then terminating p−w resources among available resources of the pool if and only if a number p_u of the “in-use” resources at the time T 4 is inferior to the predetermined number w; or

otherwise, maintaining the resources of the pool initialized independently of time.

2. The method according to claim 1 , which further comprises performing an allocation of the resource to a client by the further steps of:

a) receiving by the server of a client request for using one of the resources;

b) determining a number of available resources for the client in the pool; and

if at least one of the resources is available in the pool, then allocating the one resource to the client and changing a status of the resource allocated from “available” to “in use”, otherwise;

if each of the resources of the pool is “in use”, then determining if the maximum number m of the resources is used;

if the maximum number m of the resources of the server is not used, then initializing a new resource and adding the new resource to the pool;

if the maximum number m of the resources of the server is used and if a timeout is not reached, then returning to step b; and

if the maximum number m of the resources of the server is used and if the timeout is reached, then returning to the client request an error.

3. The method according to claim 1 , which further comprises prioritizing an allocation of the resource to the client or a request in dependence on a rank of the client or the request.

4. The method according to claim 1 , which further comprises setting at least one condition for allocation of the resource to the client.

5. A system for sharing resources among different clients, the system comprising:

a server having a maximum number m of resources among which a predetermined number w of said resources are initializable for being kept ready for use by the clients in a pool;

a communication system for communicating with the clients;

a managing device for managing said resources, said managing device having at least a processor and a memory, said managing device being able to:

change a status of a resource of said pool in dependence on an assignment of said resource to a client or a release of said resource at a time T 4 ;

to add a new resource to the pool;

to terminate said resource from the pool;

said memory containing instructions to:

determine a number p of said resources contained in said pool at the time T 4 , and to determine among said p resources, a number p_u of said resources that are “in use” at the time T 4 ;

terminate p−w resources among the p−p_u available resources of said pool if and only if at said time T 4 , m≧p>w and p_u<w; and

maintain said resources of said pool initialized independently of the time if at said time T 4 , p≦w or p_u>w.

6. The system according to claim 5 , wherein said server has a databank with a prioritizing list of at least one of the clients or the requests.

7. The system according to claim 5 , wherein said server is able to timestamp a reception of a request.