Method and computing system for transparence computing on the computer network

• US 7,467,293 B2
• Filed: 07/19/2005
• Issued: 12/16/2008
• Est. Priority Date: 11/12/2004
• Status: Active Grant

• Current Assignee: Tsinghua University
• Original Assignee: Tsinghua University
• Inventors: Wei, Li, Zhang, Yaoxue, Zhou, Yuezhi, Kuang, Wenyuan, Xu, Guangbin, Guo, Guanfei, Yang, Huajie
• Primary Examiner(s): Chen; Tse
• Assistant Examiner(s): Stoynov; Stefan
• Application Number:
  US11/184,466
• Publication Number:
  US 20060117172A1
• Time in Patent Office: 1,246 Days
• Field of Search: 713/1
• US Class Current: 713/1
• CPC Class Codes: G06F 9/4416 Network booting; Remote initial program loading [RIPL]

Abstract

A computing system based on Transparence Computing consists of clients and a server. There is no disk and no preinstalled OS on the client; a remote booting chip is set on the client'"'"'s mainboard for downloading and loading the OS. Boot-supporting services, OS kernel mirrors, and applications are preset on the server; After being powered, the client broadcasts the boot request to the network and the server allocates network parameters for it. The client downloads an OS-selecting script from the server. After the user selects the needed OS, the client downloads and loads the selected OS from the server. Hence, the client fulfills the user'"'"'s computing requirements just as a normal PC. The computing work is implemented on the client-side, during which the client can acquire corresponding resources such as applications, data, and documents from the server on demand. The client can be provided with private storage across the network by presetting a service in the mirror for creating virtual disks on the server, and thus data generated during or after the computing can be persistently stored on the server'"'"'s storage.

Claims

1. A self-organizing method for server clusters in a system based on transparence computing, the system comprising clients and servers distributed randomly on computer networks, wherein the clients are network computers, PDAs, home appliances or mobile terminals, wherein said client has no preinstalled OS, said server stores all the OS kernel mirrors and initial applications for booting the client, all the servers on the same broadcast-reachable network belong to one server cluster, a server cluster consists of one master network computer server (MNCS) and some slave network computer servers (SNCSs), a client can only be served by a server in the server cluster, and servers of the system self-organize themselves into server clusters according to the following steps:

Step I starting a process that is responsible for listening to messages from other servers

Step II if the server has not received any heart-beating signal for a time interval of Td, it affirms itself as the MNCS in the cluster, starting to broadcast the heart-beating signal to the network periodically and listening to service requests from clients, wherein Td is a timed value of the heart-beating period;

otherwise it affirms itself as a SNCS, and starts to listen to service requests from clients;

the SNCS sends a reflective signal to the MNCS when receiving a heart-beating signal;

Step III when receiving the first reflective signal from a SNCS, the MNCS adds it to the cluster;

the MNCS dynamically maintains the status information of the server cluster according to status information in the SNCS'"'"' reflective signal;

wherein;

if the MNCS has not received any reflective signal from a SNCS for a time interval of Td, it will try to contact the SNCS, and if the MNCS fails to contact the SNCS after a certain number of attempts, the MNCS removes the SNCS from the cluster;

if the SNCS has not received any heart-beating signal for a time interval of Td, the SNCS will perform the following steps;

Step 1 broadcasting a contesting signal;

Step 2 if any heart-beating signal broadcasted by the MNCS is received in a time period of Td, the processing is then ended;

Step 3 otherwise, if no contesting signal from other SNCS is received during the same period, the SNCS then upgrades itself as the MNCS, initiates the cluster status information with its own binding client status information, starts to broadcast heart-beating signal to the network, and ends the processing;

Step 4 if any contesting signal was received from other SNCS during the same period, the SNCS then makes a delay, and if any heart-beating signal is received during the delay, the SNCS confirms itself as a SNCS, and ends the processing, otherwise going to step 1.

2. The self-organizing method according to claim 1 wherein

if the MNCS receives a heart-beating signal from another MNCS, it sends a collision indication signal to that MNCS, which contains status information about the server cluster;

if a MNCS A receives a collision indication signal from another MNCS B, then A compares the number of servers in its cluster with that of B'"'"'s; and

if the A'"'"'s server cluster has servers no less than B does, A adds information of B'"'"'s server cluster into its own, adding B as a SNCS and sending a collision ACK signal to B;

if any MINCS receives a collision ACK, it then affirms itself as an SNCS, updating related status information and stopping broadcasting the heart-beating signal.

3. The self-organizing method according to claim 2 wherein said client obtains services provided by the server cluster according to the following steps:

Step a broadcasting the booting request to the network after being powered on;

Step b when a SNCS receives the booting request and finds that the client is one of its affiliated clients, the service is then provided to the client;

when the MNCS receives the booting request and finds that the client belongs to none of the servers in the server cluster according to maintained cluster status information, it will bind the client to a server as its affiliated client and notify that server, and the server then provides the client with the service;

Step c after loading the downloaded OS kernel mirror and initial application, the client broadcasts to the network again, its binding server will return a response, and then the booting procedure of the client is accomplished;

Step d after being booted, the client sends corresponding requests to its binding server for the service if it needs services, such as customization on demand, or application downloading;

if a client cannot get requested services from the server, it can be rebooted manually or automatically upon detecting that the client cannot get requested services from the server.

4. The self-organizing method according to claim 1 wherein said client obtains services provided by the server cluster according to the following steps:

Step a broadcasting the booting request to the network after being powered;

Step b when a SNCS receives the booting request and finds that the client is one of its affiliated clients, the service is then provided to the client;

when the MNCS receives the booting request and finds that the client belongs to none of the servers in the server cluster according to maintained cluster status information, it will bind the client to a server as its affiliated client and notify that server, and the server then provides the client with the service;

Step c after loading the downloaded OS kernel mirror and initial application, the client broadcasts to the network again, its binding server will return a response, and then the booting procedure of the client is accomplished;

Step d after being booted, the client sends corresponding requests to its binding server for the service if it needs services, such as customization on demand, or application downloading;

if a client cannot get requested services from the server, it can be rebooted manually or automatically upon detecting that the client cannot get requested services from the server.