Method for analysis of the time response of complex distributed systems

• US 20030120450A1
• Filed: 11/27/2002
• Published: 06/26/2003
• Est. Priority Date: 11/30/2001
• Status: Active Grant

• Current Assignee: Symtavision GmbH (DXC Technology Company)
• Original Assignee: Technische Universitat Braunschweig Carolo-wilhelmina Institut Fur Datentechnik Und Kommunikationsnetze
• Inventors: Ernst, Rolf, Richter, Kai
• Granted Patent: US 6,947,868 B2
• Time in Patent Office: Days
• US Class Current: 702/109
• CPC Class Codes: G06F 2119/12 Timing analysis or timing optimisation
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06F 30/33 Design verification, e.g. functional simulation or model checking
• CPC Class Codes: G06F 30/3308 using simulation
• CPC Class Codes: G06F 30/3312 Timing analysis

Abstract

A method for analysis of the time response of complex distributed systems with basic components (B) which have data processing program modules and/or electronic circuit modules, having the following steps:

a) Determination of the input event models (EM_Y) for recording input event streams (EE) at the input of the respective basic components (B) in the distributed system;

b) Determination of transformation functions (EMIF_X→Y, EAF_X→Y) for matching the representation of input event streams (EE) at the input of one basic component (B) in an input event model (EM_Y) which is suitable for that basic component (B);

c) Local analysis of the basic components (B) using the associated input event models (EM_Y) in order to determine the time responses of the individual basic components (B);

d) Determination of the time response of the complex distributed system from the determined time responses of the individual basic components (B).

Claims

1. A method for analysis of the time response of complex distributed systems with basic components (B) which have data processing program modules and/or electronic circuit modules, having the following steps:

a) Determination of the input event models (EM_Y) for recording input event streams (EE) at the input of the respective basic components (B) in the distributed system;

b) Determination of transformation functions (EMIF<sub>X→

Y</sub>, EAF<sub>X→

Y</sub>) for matching the representation of input event streams (EE) at the input of one basic component (B) in an input event model (EM_Y) which is suitable for that basic component (B);

c) Local analysis of the basic components (B) using the associated input event models (EM_Y) in order to determine the time responses of the individual basic components (B);

d) Determination of the time response of the complex distributed system from the determined time responses of the individual basic components (B).

2. The method as claimed in claim 1, further comprising the step of determining transformation functions (EMIF_X→

Y) for transformation of the representation of input event streams (EE) by means of parameters for associated input event models (EM_Y), in order to match input event streams (EE) to the input event models (EM_Y) of the respectively associated basic components (B).

3. The method as claimed in claim 1, further comprising the step of determining transformation functions (EAF_X→

Y) for transformation of the event streams (EE, AE), in order to match the event streams (EE, AE) to event models (EM_Y) of the respectively associated basic components (B), and local analysis of the basic components (B) using the transformed event streams (EE, AE).

4. The method as claimed in claim 1, further comprising the step of determining determination of output event models (EM_x) in order to record output event streams (AE) at the output of the respective basic components (B) in the distributed system, and determination of the transformation functions (EMIF_X→

Y, EAF<sub>X→

Y</sub>) of associated directly subsequent basic components (B) as a function of the output event models (EM_x).

5. The method as claimed in claim 1, further comprising the step of determining combination of the equation systems for the event models (EM_X EM_y) and the transformation functions (EMIF_X→

Y, EAF<sub>X→

Y</sub>) into an overall equation system, and global analysis of the complex distributed system using the overall equation system and at least one defined input event stream (ES₁₁) in the complex distributed system.

6. The method as claimed in one claim 1, further comprising the step of inserting additional components into the system in order to produce transformation functions (EAF_X→

Y).

7. The method as claimed in claim 1, wherein additional components are data buffer elements and/or timers.

8. The method as claimed in claim 1, further comprising the step of determining the analysis method and adaptation of the transformation functions ((EMIF_X→

Y, EAF<sub>X→

Y</sub>) being carried out iteratively.

10. A computer program product for analysis of the time response of complex distributed systems with program code means for carrying out the method as claimed in claim 1.

9. The method as claimed in 1, further comprising the step of determining classification of time models for the event models (EM_X, EM_y) and derivation of transformation functions (EMIF_X→

Y, EAF<sub>X→

Y</sub>) from the time model classes.