Generating simulation data using a linear curve simplification and reverse simplification method
Simulation data can be generated using a linear curve simplification and reverse simplification technique. The linear curve simplification and reverse simplification technique can facilitate the generation of simulation data from existing data where the simulation data will exhibit the same curve pattern as the existing data but with a degree of variation. In this way, varying sets of simulation data for a particular environment can be easily generated.
1. A method, performed by a computing system that includes a simplification engine and a reverse simplification engine, for generating simulation data for testing a subject computing system based on existing data for the subject computing system by applying a simplification algorithm and a reverse simplification algorithm to the existing data to generate the simulation data with a controllable degree of individual distinction from the existing data such that the simulation data represents resource utilization variations that may actually occur when the subject computing system is executing, the method comprising:
obtaining, by the simplification engine, the existing data that comprises a series of points where each point represents resource utilization of the subject computing system at a particular time;
applying, by the simplification engine, the simplification algorithm to the existing data to convert the existing data into a reduced set of characteristic points, wherein the simplification engine employs a particular value of a parameter to calculate which of the points of the existing data are characteristic points;
receiving, by the reverse simplification engine and from the simplification engine, the reduced set of characteristic points and the particular value of the parameter;
calculating, by the reverse simplification engine, simulation data with a degree of individual distinction from the existing data by:
for each adjacent pair of characteristic points, defining a reverse simplification data sample zone using the particular value of the parameter;
for each defined reverse simplification data sample zone, generating a number of additional points that fall within the reverse simplification data sample zone; and
combining the number of additional points generated for each defined reverse simplification data sample zone along with the characteristic points to thereby produce the simulation data; and
testing the subject computing system using the simulation data.
2. The method of claim 1 , wherein each point in the series of points of the existing data represents one of:
actual resource utilization of the subject computing system obtained by monitoring the actual computing system's performance at each particular time; or
simulated resource utilization.
3. The method of claim 1 , wherein the parameter comprises a distance.
4. The method of claim 3 , wherein the distance is a perpendicular distance from a line between two points of the existing data.
5. The method of claim 4 , wherein the two points of the existing data are the first and last points.
6. The method of claim 3 , wherein the distance is a radial distance.
7. The method of claim 1 , wherein defining a reverse simplification data sample zone comprises:
identifying a first line between a first and a last characteristic point; and
identifying a second line that is parallel to the first line and is the farthest-spaced line from the first line that intersects a characteristic point.
8. The method of claim 7 , wherein defining a reverse simplification data sample zone further comprises:
identifying a first rectangle that has a first set of opposing sides that extend along the first and second lines and a second set of opposing sides that intersect the adjacent pair of characteristic points; and
identifying a second rectangle that has a first set of opposing sides that intersect the adjacent pair of characteristic points, has an axis of symmetry along a line segment between the adjacent pair of characteristic points, and that has a second set of opposing sides that are each spaced from the axis of symmetry by the particular value of the parameter;
wherein the reverse simplification data sample zone comprises the intersection of the first and second rectangles.
9. The method of claim 1 , wherein the number of additional points are generated randomly.
10. The method of claim 1 , further comprising:
receiving, by the reverse simplification engine, a decompression rate, wherein the number of additional points that are generated are based on the decompression rate.
11. The method of claim 1 , wherein the number of additional points that are generated are based on a number of points in the existing data.
12. The method of claim 1 , further comprising:
generating characteristic points from the simulation data; and
comparing the characteristic points generated from the simulation data to the characteristic points that were generated from the existing data to verify that the simulation data exhibits a degree of individual distinction from the existing data.
13. The method of claim 1 , wherein the simplification algorithm is the Ramer-Douglas-Peucker algorithm.
14. The method of claim 1 , further comprising:
applying, by the simplification engine, the simplification algorithm to the existing data to convert the existing data into a second reduced set of characteristic points, wherein the simplification engine employs a second particular value of the parameter, different from the first particular value of the parameter, to calculate which of the points of the existing data are characteristic points in the second reduced set of characteristic points;
receiving, by the reverse simplification engine and from the simplification engine, the second reduced set of characteristic points and the second particular value of the parameter;
calculating, by the reverse simplification engine, second simulation data with a second degree of individual distinction from the existing data by:
for each adjacent pair of characteristic points in the second reduced set of characteristic points, defining a reverse simplification data sample zone using the second particular value of the parameter;
for each defined reverse simplification data sample zone, generating a number of additional points that fall within the reverse simplification data sample zone; and
combining the number of additional points generated for each defined reverse simplification data sample zone along with the characteristic points in the second reduced set of characteristic points to thereby produce the second simulation data; and
testing the subject computing system using the second simulation data.
15. The method of claim 1 , further comprising:
sending the characteristic points and the particular value of the parameter to another reverse simplification engine.
16. One or more computer storage media storing computer executable instructions which when executed by one or more processors implement a method for generating simulation data for testing a subject computing system based on existing data for the subject computing system by applying a simplification algorithm and a reverse simplification algorithm to the existing data to generate the simulation data with a controllable degree of individual distinction from the existing data such that the simulation data represents resource utilization variations that may actually occur when the subject computing system is executing, the method comprising:
obtaining, by a simplification engine, the existing data that comprises a series of points where each point represents resource utilization of the subject computing system at a particular time;
applying, by the simplification engine, the simplification algorithm to the existing data to convert the existing data into a reduced set of characteristic points, wherein the simplification engine employs a particular value of a parameter to calculate which of the points of the existing data are characteristic points;
receiving, by a reverse simplification engine and from the simplification engine, the reduced set of characteristic points and the particular value of the parameter;
calculating, by the reverse simplification engine, simulation data with a degree of individual distinction from the existing data by:
for each adjacent pair of characteristic points, defining a reverse simplification data sample zone using the particular value of the parameter;
for each defined reverse simplification data sample zone, generating a number of additional points that fall within the reverse simplification data sample zone; and
combining the number of additional points generated for each defined reverse simplification data sample zone along with the characteristic points to thereby produce the simulation data; and
testing the subject computing system using the simulation data.
17. The computer storage media of claim 16 , wherein defining a reverse simplification data sample zone comprises:
identifying a first line between a first and a last characteristic point; and
identifying a second line that is parallel to the first line and is the farthest-spaced line from the first line that intersects a characteristic point.
18. The computer storage media of claim 17 , wherein defining a reverse simplification data sample zone further comprises:
identifying a first rectangle that has a first set of opposing sides that extend along the first and second lines and a second set of opposing sides that intersect the adjacent pair of characteristic points; and
identifying a second rectangle that has a first set of opposing sides that intersect the adjacent pair of characteristic points, has an axis of symmetry along a line segment between the adjacent pair of characteristic points, and that has a second set of opposing sides that are each spaced from the axis of symmetry by the particular value of the parameter;
wherein the reverse simplification data sample zone comprises the intersection of the first and second rectangles.
19. A method, performed by a computing system that includes a simplification engine and a reverse simplification engine, for generating multiple sets of simulation data for testing a subject computing system based on existing data for the subject computing system by applying a simplification algorithm and a reverse simplification algorithm to the existing data to generate each of the multiple sets of simulation data with a controllable degree of individual distinction from the existing data such that each of the multiple sets of simulation data represents resource utilization variations that may actually occur when the subject computing system is executing, the method comprising:
obtaining, by the simplification engine, the existing data that comprises a series of points where each point represents resource utilization of the subject computing system at a particular time;
applying, by the simplification engine, the simplification algorithm to the existing data to convert the existing data into multiple reduced sets of characteristic points, wherein the simplification engine employs a different value of a parameter to calculate which of the points of the existing data are characteristic points in each of the multiple reduced sets of the characteristic points;
receiving, by the reverse simplification engine and from the simplification engine, the multiple reduced sets of characteristic points and the corresponding values of the parameter;
calculating, by the reverse simplification engine, multiple sets of simulation data where each set of simulation data exhibits a different degree of individual distinction from the existing data, each set of simulation data being calculated using a corresponding reduced set of characteristic points and a corresponding value of the parameter by:
for each adjacent pair of characteristic points in the corresponding reduced set of characteristic points, defining a reverse simplification data sample zone using the corresponding value of the parameter;
for each defined reverse simplification data sample zone, generating a number of additional points that fall within the reverse simplification data sample zone; and
combining the number of additional points generated for each defined reverse simplification data sample zone along with the characteristic points in the corresponding reduced set of characteristic points to thereby produce the set of simulation data; and
testing the subject computing system using at least one of the sets of simulation data.
20. The method of claim 19 , wherein defining a reverse simplification data sample zone comprises:
identifying a first line between a first and a last characteristic point in the corresponding reduced set of characteristic points;
identifying a second line that is parallel to the first line and is the farthest-spaced line from the first line that intersects a characteristic point;
identifying a first rectangle that has a first set of opposing sides that extend along the first and second lines and a second set of opposing sides that intersect the adjacent pair of characteristic points; and
identifying a second rectangle that has a first set of opposing sides that intersect the adjacent pair of characteristic points, has an axis of symmetry along a line segment between the adjacent pair of characteristic points, and that has a second set of opposing sides that are each spaced from the axis of symmetry by the corresponding value of the parameter;
wherein the reverse simplification data sample zone comprises the intersection of the first and second rectangles.