NON-INTRUSIVE DATA LOGGING

• US 20120072399A1
• Filed: 09/21/2011
• Published: 03/22/2012
• Est. Priority Date: 04/16/2003
• Status: Active Grant

• Current Assignee: The MathWorks, Inc.
• Original Assignee: The MathWorks, Inc.
• Inventors: Taitel, Howard, Kositsky, Michael
• Granted Patent: US 8,631,013 B2
• Time in Patent Office: Days
• US Class Current: 707/687
• CPC Class Codes: G06F 16/2237 Vectors, bitmaps or matrices
• CPC Class Codes: G06F 17/40 Data acquisition and logging for input to computer G06F3/00

Abstract

Mediums, methods, and systems are provided for efficiently logging data. A model may include one or more logging points which process data, the data being stored in a log associated with the logging point. The logging point may request that a logging object store the data point. The logging object may include a reference to a vector for storing the data point. When two or more logging objects are associated with the same logged data points, the two or more logging objects may share the same vector. If an object logs a point which is not present in a shared vector, the object may update the object'"'"'s reference so that the object references a different existing vector, or the object may create a new vector. The vectors may be compressed and/or made circular to achieve improved efficiency.

Claims

1. A non-transitory computer-readable medium storing computer readable instructions for:

creating a first object;

the first object interacting with a first logging location in an arbitrary model,

the first object including a first reference,

the first object having a first counter,

the first object creating a first vector having a first data point for the first object;

creating a second object;

the second object interacting with a second logging point in an arbitrary model,

the second object including a second reference, and

the second object having a counter for indexing into a vector;

determining whether the second object shares at least the first data point of the data in the first vector with the first object;

indexing the counter when the second object shares the first data point; and

creating a second vector when the second object does not share the first data point.

2. The medium of claim 1, wherein the data is time data.

3. The medium of claim 1, wherein the first reference and the second reference are pointers to a vector.

4. The medium of claim 1, wherein the arbitrary model is a block diagram model of a dynamic system.

5. The medium of claim 1, wherein the first vector initially stores uniform data and is a compressed first vector.

6. The medium of claim 1, wherein:

the first vector comprises at least N data points; and

the medium further comprises instructions for;

receiving a request that the second object logs an N^th data point having a value V,

indexing the counter to the value N, and

determining whether the N^th data point in the first vector has the value V, wherein;

if the N^th data point in the first vector has the value V, the first vector is shared with the second object, and

if the N^th data point in the first vector does not have the value V, the second vector is created having the first (N-1) data points from the first vector and the value V as an N^th data point.

7. The medium of claim 1, wherein the data is streaming data in which data points become known during an execution of the arbitrary model.

8. A computer implemented method comprising:

creating a first object;

the first object interacting with a first logging location in an arbitrary model,

the first object including a first reference,

the first object having a first counter,

the first object creating a first vector having a first data point for the first object;

creating a second object;

the second object interacting with a second logging point in an arbitrary model,

the second object including a second reference, and

the second object having a counter for indexing into a vector;

determining whether the second object shares at least the first data point of the data in the first vector with the first object;

indexing the counter when the second object shares the at least a portion of the data; and

creating a second vector when the second object does not share the at least a portion of the data.

9. A system comprising:

a storage for storing;

a first object;

the first object interacting with a first logging location in an arbitrary model,

the first object including a first reference,

the first object having a first counter,

the first object creating a first vector having a first data point for the first object;

a second object;

the second object interacting with a second logging point in an arbitrary model,

the second object including a second reference, and

the second object having a counter for indexing into a vector; and

a processor for;

determining whether the second object shares at least the first data point of the data in the first vector with the first object;

indexing the counter when the second object shares the at least a portion of the data; and

creating a second vector when the second object does not share the at least a portion of the data.

10. A non-transitory computer-readable medium storing computer readable instructions for:

monitoring a plurality of objects, the objects for logging data from a plurality of logging locations in a model;

monitoring a plurality of log vectors, where the plurality of log vectors is fewer than the plurality of objects;

receiving a logging request at one of the plurality of objects, the request being for a data point to be logged with data in one of the plurality of log vectors;

determining whether the data point is in one of the plurality of log vectors; and

sharing a log vector having the data point with the one of the plurality of objects at which the logging request was received.

11. A non-transitory computer-readable medium storing computer readable instructions for:

creating a plurality of logging objects, the plurality of logging objects;

comprising a first logging object and a second logging object, and

being exposed to an environment that generates information to be stored in a log;

creating one or more vectors, the one or more vectors;

comprising a first vector, and

storing the information;

accepting a request from the environment that the first logging object log a data point;

logging the data point in the first vector;

identifying that the first logging object can share the first vector with the second logging object; and

sharing the first vector with the second logging object.

12. The medium of claim 11, wherein:

the first logging object and the second logging object each comprise a pointer to one of the one or more vectors; and

sharing the first vector with the second logging object comprises causing the pointer of the second logging object to point to the first vector.

13. The medium of claim 11, wherein the medium further stores instructions for:

simulating a block diagram model, the block diagram model comprising one or more logging points;

receiving an instruction from one of the logging points to log information; and

logging the information in one of the one or more vectors.

14. The medium of claim 11, wherein the logging points are configured by a user.

15. The medium of claim 11, wherein the first vector initially stores uniform data and is a compressed first vector.

16. The medium of claim 15, wherein logging the data point in the first vector causes the stored uniform data to become non-uniform, and the medium further stores instructions for converting the compressed first vector into a non-compressed first vector.

17. The medium of claim 11, wherein the medium further stores instructions for specifying a maximum number of data points allowed for the first vector.

18. The medium of claim 17, wherein logging the data point in the first vector causes the first vector to exceed the maximum number of data points allowed for the first vector, and the medium further stores instructions for converting the first vector into a circular vector that logs the point using a circular buffer.

19. The medium of claim 11, wherein the plurality of logging objects are exposed to the environment through an Application Programming Interface (API).

20. The medium of claim 19, wherein the API specifies a function for logging a data point “

d”

using one of the logging objects.

21. The medium of claim 11, further storing instructions for identifying that the first logging object can share the first vector with the second logging object by comparing the first vector with a vector associated with the second logging object.

22. The medium of claim 11, wherein the data is time data.

23. A computer-implemented method comprising:

creating a plurality of logging objects, the plurality of logging objects;

comprising a first logging object and a second logging object, and

being exposed to an environment that generates information to be stored in a log;

creating one or more vectors, the one or more vectors;

comprising a first vector, and

storing the information;

accepting a request from the environment that the first logging object log a data point;

logging the data point in the first vector;

identifying that the first logging object can share the first vector with the second logging object; and

sharing the first vector with the second logging object.

24. A non-transitory computer-readable medium storing computer readable instructions for:

creating a first object;

the first object interacting with a first logging location in a first arbitrary model,

the first object including a first reference,

the first object having a first counter,

the first object creating a first vector having a first data point for the first object;

creating a second object;

the second object interacting with a second logging point in a second arbitrary model,

the second object including a second reference, and

the second object having a counter for indexing into a vector;

determining whether the second object shares at least the first data point of the data in the first vector with the first object;

indexing the counter when the second object shares the first data point; and

creating a second vector when the second object does not share the first data point.