Data server with event driven sampling

• US 5,809,238 A
• Filed: 06/23/1995
• Issued: 09/15/1998
• Est. Priority Date: 06/23/1994
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: Candle Distributed Solutions, Inc.
• Inventors: Yung, Alex, Greenblatt, Sam
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Alam, Hosain T.
• Application Number:
  US08/494,831
• Time in Patent Office: 1,180 Days
• Field of Search: 395/200.11, 395/610
• US Class Current: 709/202
• CPC Class Codes: G06F 11/34 Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/348 Circuit details, i.e. tracer hardware
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/86 Event-based monitoring

Abstract

A data server data base system applies a request for data to a data engine which launches a probe via network transport that causes the data to be collected on a remote platform and returned via network transport. The returned data is then applied to the requesting user application in the form of one or more rows of columnar data emulating the return of data from a conventional data base system including data tables. This operation may further be enhanced by the inclusion of Event processing in which the occurrence of an event based on a predicate test is determined by the data server without requiring attention of the user application. In particular, the user application stores a rule statement in the data server and then references that statement by name in a data inquiry. An event level data probe is invoked to collect the data referenced in the rule statement and the collected data is tested in accordance with the rule statement to determine if the event has occurred. Data passing the rule statement test is returned to the user application if the event is determined to have occurred unless Delta processing is invoked in which case only data representing a transition from Event true to Event false or Event false to Event true is returned.

Claims

1. A computer network comprising:

one or more user applications (U1 . . . Un);

a plurality of computer platforms (P1 . . . Pn);

a transport network interconnecting the computer platforms;

a data management engine on one of said plurality of computer platforms, said engine responsive to a request for data from a user application to invoke;

at least one data probe to cause the collection of the collected data from the appropriate computer platform and the return of the collected data in the form of one or more rows of columnar data;

an event manager for filtering collected data to return the collected data to the user application only upon the occurrence of a specified condition; and

data return means for applying the collected data from the data probe to the network transport for return to the user application.

2. The computer network claimed in claim 1 wherein the at least one data probe further comprises:

means specifically associated with the computer platform (P1 . . . Pn) from which the data is to be collected so that the operations of the data collection from the platform is transparent to the user application so that user applications compatible with the data management engine are usable to transparently retrieve data from heterogeneous computer platforms across the computer network.

3. The computer network claimed in claim 1, further comprising:

a data table responsive to the data probe for storage of the requested data and return of the requested data to the requesting user application.

4. The computer network claimed in claim 1, wherein the event manager further comprises:

a rule processor responsive to the filtered, collected data for returning at least a portion of the collected data to the user application.

5. The computer network claimed in claim 4, wherein the event manager further comprises:

a data buffer wherein the occurrence of the specified condition is determined by detecting that the number of rows of data stored in the data buffer is greater than zero.

6. The computer network claimed in claim 5, wherein the event manager further comprises:

an advisor probe responsive to the rule processor for collecting the data from the data buffer upon indication that the number of rows therein is greater than zero.

7. The computer network claimed in claim 6, wherein the event manager further comprises:

an advisor table for storing data processed by the advisor probe and for forwarding at least selected portions of that data to the requesting user application via the network transport means.

8. The computer network claimed in claim 4, wherein the event manager further comprises:

a rule table responsive to the network transport for storing rule definitions in accordance with a rule name, so that the user application may request specified data to be returned only upon the occurrence of the condition requirement specified in a particular rule definition by specifying the related rule name.

9. The computer network claimed in claim 8, wherein the event manager further comprises:

a nested event manager responsive to the request from the user application for returning the request to the rule processor as long as the request includes a rule name, wherein the user application may conveniently nest condition requirements within other condition requirements.

10. A method of managing data on a computer network comprising the steps of:

causing one or more user applications (U1 . . . Un) to request the collection of specified data;

interconnecting a plurality of computer platforms (P1 . . . Pn) with a transport network;

installing a data management engine on one of said plurality of computer platforms; and

causing said engine to respond to the request for data from the user application by launching at last one data probe to collect the requested data from the appropriate computer platform and return the collected data in the form of one or more rows of columnar data;

filtering the collected data to return the collected data to the user application only upon the occurrence of a specified condition; and

applying the collected data from the data probe to the network transport for return to the user application.

11. The method of managing data on a computer network claimed in claim 10 wherein the step of launching a data probe further comprises the step of:

associating the data probe directly with the computer platform (P1 . . . Pn) from which the data is to be collected so that the operations of the data collection from the platform is transparent to the user application wherein user applications compatible with the data management engine are usable to transparently retrieve data from heterogeneous computer platforms across the computer network.

12. The method of managing data on a computer network claimed in claim 11, further comprising the step of:

causing a data table to respond to the data probe by storing of the requested data and returning the requested data to the requesting user application.

13. The method of managing data on a computer network claimed in claim 10, further comprising the step of:

causing a rule processor to respond to the filtered, collected data by returning at least a portion of the collected data to the user application.

14. The method of managing data on a computer network claimed in claim 13 further comprising the step of:

for storing the filtered, collected data in a data buffer wherein the occurrence of the specified condition is determined by detecting that the number of rows of data stored in the data buffer is greater than zero.

15. The method of managing data on a computer network claimed in claim 14 further comprising the step of:

causing an advisor probe to respond to the rule processor by collecting the data from the data buffer upon indication that the number of rows therein is greater than zero.

16. The method of managing data on a computer network claimed in claim 15 further comprising the step of:

providing an advisor table for storing data processed by the advisor probe and for forwarding at least selected portions of that data to the requesting user application via the network transport means.

17. The method of managing data on a computer network claimed in claim 13 further comprising the step of:

providing a rule table responsive to the network transport for storing rule definitions in accordance with a rule name, wherein the user application may request that specified data is to be returned only upon the occurrence of the condition requirements specified in a particular rule definition by indicating the related rule name.

18. The method of managing data on a computer network claimed in claim 16 further comprising the step of:

providing a nested event manager responsive to the request from the user application for returning the request to the rule processor as long as the request includes a rule name, wherein the user application may conveniently nest condition requirements within other condition requirements.