High-availability WWW computer server system with pull-based load balancing using a messaging and queuing unit in front of back-end servers

• US 6,023,722 A
• Filed: 07/23/1997
• Issued: 02/08/2000
• Est. Priority Date: 12/07/1996
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Colyer, Adrian Mark
• Primary Examiner(s): Lim, Krisna
• Application Number:
  US08/899,372
• Time in Patent Office: 930 Days
• Field of Search: 395/200.33, 395/200.65, 395/200.56, 395/200.32, 709/201, 709/203, 709/235
• US Class Current: 709/201
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/1014 based on the content of a request
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A high-availability computer server system, and method of operation thereof, capable of serving a large number of requests received from a plurality of computer client devices connected through a network to said server system, said requests specifically identifying said server system, said server system comprising: a messaging and queuing unit having an input connected to said network upon which said requests identifying said server system are received, and an output; and a plurality of server units connected in parallel to said output of said messaging and queuing unit, where each server unit pulls requests off of a queue in the messaging and queuing unit.

Claims

1. An Internet World Wide Web (WWW) computer server system which serves requests received from a plurality of WWW browser computer client devices connected through the Internet to said server system, said requests specifically identifying said server system, said server system comprising:

a messaging and queuing unit having an input connected to the Internet upon which said requests identifying said server system are received, and an output; and

a plurality of server units connected in parallel to said output of said messaging and queuing unit, wherein each server unit pulls requests off of a queue included in said messaging and queuing unit.

2. The system of claim 1 wherein said messaging and queuing unit includes means for assigning priority to received requests.

3. The system of claim 2 wherein said means for assigning priority assigns higher priority to text requests as compared to graphics requests.

4. The system of claim 1 wherein said messaging and queuing unit includes means for triggering an additional server unit as the number of unserved requests received by said messaging and queuing unit surpasses a threshold amount.

5. The system of claim 1 wherein said messaging and queuing unit includes means for sending a request to one of said plurality of server units in response to said server unit informing said messaging and queuing unit that said server unit is ready to serve a request.

6. A method of serving Internet World Wide Web (WWW) requests received from a plurality of WWW browser client computer devices via the Internet, each of said requests specifically identifying a specific WWW server system, said method comprising steps of:

storing, at said specific server system, said received requests into a messaging and queuing unit; and

sending requests from said messaging and queuing unit to a plurality of parallel-connected server units in response to the receipt of pull requests from the server units.