Efficient method and computer program for modeling and improving static memory performance across process variations and environmental conditions

• US 8,001,493 B2
• Filed: 08/27/2008
• Issued: 08/16/2011
• Est. Priority Date: 03/10/2005
• Status: Expired due to Fees

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: International Business Machines SA (International Business Machines Corporation)
• Inventors: Devgan, Anirudh, Joshi, Rajiv V.
• Primary Examiner(s): Dinh; Paul
• Assistant Examiner(s): Nguyen; Nha T
• Application Number:
  US12/199,161
• Publication Number:
  US 20080319717A1
• Time in Patent Office: 1,084 Days
• Field of Search: 716/54, 716/56, 716/106, 716/111, 716/112, 716/132, 716/133, 716/134, 716/135, 703/13, 703/19, 703/20
• US Class Current: 716/54
• CPC Class Codes: G06F 2111/08 Probabilistic or stochastic CAD
• CPC Class Codes: G06F 30/30 Circuit design

Abstract

An efficient method and computer program for modeling and improving stating memory performance across process variations and environmental conditions provides a mechanism for raising the performance of memory arrays beyond present levels/yields. Statistical (Monte-Carlo) analyses of subsets of circuit parameters are performed for each of several memory performance variables and then sensitivities of each performance variable to each of the circuit parameters are determined. The memory cell design parameters and/or operating conditions of the memory cells are then adjusted in conformity with the sensitivities, resulting in improved memory yield and/or performance. Once a performance level is attained, the sensitivities can then be used to alter the probability distributions of the performance variables to achieve a higher yield. Multiple cell designs can be compared for performance, yield and sensitivity of performance variables to circuit parameters over particular environmental conditions in order to select the best cell design.

Claims

1. A computer-performed method for characterizing a design of a memory cell, comprising:

pre-selecting subsets of circuit parameters of memory cell circuit elements, wherein the subsets correspond to a plurality of operational performance variables describing performance of the memory cell, wherein the subsets are pre-selected according to an expectation that first sensitivities of the operational performance variables to changes in the circuit parameters within their corresponding subsets are greater than second sensitivities of the performance variables with respect to changes in other circuit parameters that are not included in their corresponding subsets;

determining values of statistical descriptors, including at least mean values, for the associated circuit parameters for simulation by a simulation program within a computer system;

statistically simulating, by the computer program, the operational performance variables using the values of the statistical descriptors;

repeating the statistically simulating, while systematically varying the circuit parameters in the corresponding subsets by adjusting their associated mean values and while maintaining the mean values associated with the circuit parameters that are not included within the subsets at fixed values across the multiple repetitions of the statistically simulating until the circuit parameters have been varied across corresponding predetermined ranges;

computing the first sensitivities of the operational performance variables to the systematic variations of the mean values of the memory cell circuit parameters within their corresponding subsets from results of the statistically simulating by computing measures of partial derivatives of variations of the corresponding operational performance variables to the variations of the mean values of the memory cell circuit parameters; and

storing the computed sensitivities in a memory of the computer system.

2. The method of claim 1, wherein the pre-selecting is performed by:

performing multiple statistical analyses of a full set of devices within the memory cell over variations of circuit parameters for the full set of devices;

computing third sensitivities of the operational performance variables to said variations of the memory cell circuit parameters within the full set of devices; and

comparing the third sensitivities to a threshold criteria to select the subsets of parameters as those parameters for which a corresponding operational performance variable has a greater third sensitivity than the third sensitivity of other parameters within the parameters for the full set of devices.

3. The method of claim 2, further comprising performing the multiple statistical analyses, the computing and comparing on multiple cell designs of differing order and further comprising selecting one or more of the multiple cell designs for further analysis by determining which of the multiple cell designs have lesser values of the third sensitivities.

4. The method of claim 1, further comprising performing the selecting, determining, simulating, repeating and computing on multiple cell designs of differing order and further comprising selecting one or more of the multiple cell designs for further analysis by determining which of the multiple cell designs have lesser values of the first sensitivities.

5. The method of claim 1, further comprising adjusting one or more circuit parameter values within the subsets in conformity with a result of the computing the first sensitivities of the corresponding operational performance variables, whereby design of the memory cell is improved.

6. The method of claim 1, further comprising:

eliminating one or more circuit parameters from one or more of the subsets in conformity with a result of the computing; and

second repeating the determining, simulating, repeating and computing, wherein the second repeating is performed for the one or more subsets having a reduced number of parameters, and wherein the values of the eliminated circuit parameters are set to fixed values during the second repeating.

7. The method of claim 1, wherein the plurality of performance variables includes write stability and read stability, and wherein the simulating simulates the memory cell for each of write stability, write delay, read stability and read delay over the values of the circuit parameters within the corresponding subsets.

8. A workstation computer system comprising a processor for executing program instructions and a memory coupled to the processor for storing program instructions, the program instructions including program instructions for characterizing a design of a memory cell, the program instructions comprising program instructions for:

pre-selecting subsets of circuit parameters of memory cell circuit elements, wherein the subsets correspond to a plurality of operational performance variables describing performance of the memory cell, wherein the subsets are pre-selected according to an expectation that first sensitivities of the operational performance variables to changes in the circuit parameters within their corresponding subsets are greater than second sensitivities of the performance variables with respect to changes in other circuit parameters that are not included in their corresponding subsets;

determining values of statistical descriptors, including at least mean values, for the associated circuit parameters for simulation by a simulation program within a computer system;

statistically simulating, by the computer program, the operational performance variables using the values of the statistical descriptors;

repeatedly executing the program instructions for statistically simulating, while systematically varying the circuit parameters in the corresponding subsets by adjusting their associated mean values and while maintaining the mean values associated with the circuit parameters that are not included within the subsets at fixed values across the multiple repetitions of the statistically simulating until the circuit parameters have been varied across corresponding predetermined ranges;

computing the first sensitivities of the operational performance variables to the systematic variations of the mean values of the memory cell circuit parameters within their corresponding subsets from results of the statistically simulating by computing measures of partial derivatives of variations of the corresponding operational performance variables to the variations of the mean values of the memory cell circuit parameters; and

storing the computed sensitivities in a memory of the computer system.

9. The workstation computer system of claim 8, wherein the program instructions for pre-selecting further comprise program instructions for:

performing multiple statistical analyses of a full set of devices within the memory cell over variations of circuit parameters for the full set of devices;

computing third sensitivities of the operational performance variables to said variations of the memory cell circuit parameters within the full set of devices; and

comparing the third sensitivities to a threshold criteria to select the subsets of parameters as those parameters for which a corresponding operational performance variable has a greater third sensitivity than the third sensitivity of other parameters within the parameters for the full set of devices.

10. The workstation computer system of claim 9, wherein the program instructions for performing the multiple statistical analyses, the computing and comparing are executed for multiple cell designs of differing order and further comprising program instructions for selecting one or more of the multiple cell designs for further analysis by determining which of the multiple cell designs have lesser values of the third sensitivities.

11. The workstation computer system of claim 8, wherein the program instructions for selecting, determining, simulating, repeatedly executing, and computing are executed for multiple cell designs of differing order and further comprising program instructions for comprising selecting one or more of the multiple cell designs for further analysis by determining which of the multiple cell designs have lesser values of the first sensitivities.

12. The workstation computer system of claim 8, further comprising program instructions for adjusting one or more circuit parameter values within the subsets in conformity with a result of the computing the first sensitivities of the corresponding operational performance variables, whereby design of the memory cell is improved.

13. The workstation computer system of claim 8, further comprising program instructions for:

eliminating one or more circuit parameters from one or more of the subsets in conformity with a result of the computing; and

second repeatedly executing the program instructions for determining, simulating, repeatedly executing and computing, wherein the second repeatedly executing is performed for the one or more subsets having a reduced number of parameters, and wherein the values of the eliminated circuit parameters are set to fixed values during repetitions of the simulating.

14. The workstation computer system of claim 8, wherein the plurality of performance variables includes write stability and read stability, and wherein the simulating simulates the memory cell for each of write stability, write delay, read stability and read delay over the values of the circuit parameters within the corresponding subsets.

15. A computer program product comprising non-transitory computer-readable media storing program instructions for execution on a workstation computer, the program instructions for characterizing a design of a memory cell, the program instructions comprising program instructions for:

pre-selecting subsets of circuit parameters of memory cell circuit elements, wherein the subsets correspond to a plurality of operational performance variables describing performance of the memory cell, wherein the subsets are pre-selected according to an expectation that first sensitivities of the operational performance variables to changes in the circuit parameters within their corresponding subsets are greater than second sensitivities of the performance variables with respect to changes in other circuit parameters that are not included in their corresponding subsets;

determining values of statistical descriptors, including at least mean values, for the associated circuit parameters for simulation by a simulation program within a computer system;

statistically simulating, by the computer program, the operational performance variables using the values of the statistical descriptors;

repeatedly executing the program instructions for statistically simulating, while systematically varying the circuit parameters in the corresponding subsets by adjusting their associated mean values and while maintaining the mean values associated with the circuit parameters that are not included within the subsets at fixed values across the multiple repetitions of the statistically simulating until the circuit parameters have been varied across corresponding predetermined ranges;

computing the first sensitivities of the operational performance variables to the systematic variations of the mean values of the memory cell circuit parameters within their corresponding subsets from results of the statistically simulating by computing measures of partial derivatives of variations of the corresponding operational performance variables to the variations of the mean values of the memory cell circuit parameters; and

storing the computed sensitivities in the memory.

16. The computer program product of claim 15, wherein the program instructions for pre-selecting further comprise program instructions for:

performing multiple statistical analyses of a full set of devices within the memory cell over variations of circuit parameters for the full set of devices;

computing third sensitivities of the operational performance variables to said variations of the memory cell circuit parameters within the full set of devices; and

comparing the third sensitivities to a threshold criteria to select the subsets of parameters as those parameters for which a corresponding operational performance variable has a greater third sensitivity than the third sensitivity of other parameters within the parameters for the full set of devices.

17. The computer program product of claim 16, wherein the program instructions for performing the multiple statistical analyses, the computing and comparing are executed for multiple cell designs of differing order and further comprising program instructions for selecting one or more of the multiple cell designs for further analysis by determining which of the multiple cell designs have lesser values of the third sensitivities.

18. The computer program product of claim 15, wherein the program instructions for selecting, determining, simulating, repeatedly executing and computing are executed for multiple cell designs of differing order and further comprising program instructions for comprising selecting one or more of the multiple cell designs for further analysis by determining which of the multiple cell designs have lesser values of the first sensitivities.

19. The computer program product of claim 15, further comprising program instructions for adjusting one or more circuit parameter values within the subsets in conformity with a result of the computing the first sensitivities of the corresponding operational performance variables, whereby design of the memory cell is improved.

20. The computer program product of claim 15, further comprising program instructions for:

eliminating one or more circuit parameters from one or more of the subsets in conformity with a result of the computing; and

second repeatedly executing the program instructions for the determining, simulating, repeatedly executing and computing, wherein the second repeatedly executing is performed for the one or more subsets having a reduced number of parameters, and wherein the values of the eliminated circuit parameters are set to fixed values during repetitions of the simulating.