Method and apparatus for controlling an extensible computing system

(54) METHOD AND APPARATUS FOR
CONTROLLING AN EXTENSIBLE
COMPUTING SYSTEM

(75) Inventors: Ashar Aziz, Fremont, CA (US); Tom
Markson, San Mateo, CA (US);
Martin Patterson, Mountain View, CA
(US); Mark Gray, Mountain View, CA
(US)

(73) Assignee: Terraspring, Inc., Fremont, CA (US)

( * ) Notice: Subject to any disclaimer, the term ol this patent is extended or adjusted under 35 U.S.C. 154(b) by 202 days.

(21) Appl. No.: 09/630,440

(22) Filed: Aug. 2, 2000

Related U.S. Application Data

(63) Continuation-in-part of application No. 09/502,170, filed on Feb. 11, 2000

(60) Provisional application No. 60/213,090, filed on Jun. 20, 2000, and provisional application No. 60/150,394, filed on Aug. 23, 1999.

(51) Int. CI.7 G05B 19/18

(52) U.S. CI 700/3; 700/99; 700/101;

709/104; 709/105; 709/208; 709/209; 712/31

(58) Field of Search 700/3, 99, 101;

709/104, 105, 208, 209; 712/31

(56) References Cited

U.S. PATENT DOCUMENTS

4,591,967 A * 5/1986 Mattes et al 700/3

5,163,130 A 11/1992 Hullot

5,504,670 A * 4/1996 Barth et al 709/104

5,574,914 A 11/1996 Hancock et al.

5,590,284 A * 12/1996 Crosetto 712/31

5,659,786 A 8/1997 George et al.

5,951,683 A * 9/1999 Yuuki et al 709/208

6,446,141 Bl * 9/2002 Nolan et al 709/203

FOREIGN PATENT DOCUMENTS

EP 0 262 750 A2 4/1988

OTHER PUBLICATIONS

Rob Short, et al., "Windows NT Clusters for Availability and Scalability," 1997 IEEE, pp. 8-13.

Elliotte Rusty Harold, "XML: Extensible Markup Language," 1998, pp. 1-431.

Radek Vingralek, et al., "Snowball: Scalable Storage on Networks ol Workstations with Balanced Load," pp. 117-156, Distributed and Parallel Databases, vol. 6, No. 2, Apr. 1998, XP-002162201.

Armando Fox, et al., "Cluster-Based Scalable Network Services," pp. 78-91, Operating Systems Review (SIGOPS), US, ACM Headquarter, New York, vol. 31, No. 5, Dec. 1, 1997, XP-000771023.

* cited by examiner

Primary Examiner—Ramesh Patel
Assistant Examiner—-Thomas Pham

(74) Attorney, Agent, or Firm—Hickman Palermo Truong & Becker LLP; Edward A. Becker

A Virtual Server Farm (VSF) is created out of a wide scale computing fabric ("Computing Grid") which is physically constructed once and then logically divided up into VSFs for various organizations on demand. Allocation and control of the elements in the VSF is performed by a control plane connected to all computing, networking, and storage elements in the computing grid through special control ports. The control plane is comprised of a control mechanism hierarchy that includes one or more master control process mechanisms communicatively coupled to one or more slave control process mechanisms. The one or more master control process mechanisms instruct the slave control process mechanisms to establish VSFs by selecting subsets of processing and storage resources.

45 Claims, 18 Drawing Sheets