Cost management of software application portfolio

A computer program product, comprising a computer readable storage medium having a computer readable program code stored therein, said computer readable program code containing instructions configured to be executed by a processor of a computer system to implement a method for reducing a cost of ownership of a portfolio of N software applications, said method comprising the steps of:

• identifying a regression function for each software application of the N software applications, wherein each regression function expresses an approximate cost of ownership of the application in terms of at least one parameter and at least one undetermined coefficient, wherein N is at least 2;
• selecting J applications A1, A2, . . . , Aj from the N applications such that a remaining N-J applications of the N applications are unselected, wherein J<N, wherein the coefficients in the regression functions for the N-J unselected applications constitute a subset of the coefficients in the regression functions for the J selected applications, and wherein Fj represents the approximate cost of ownership of the application Aj for j=1, 2, . . . , J;
• determining the coefficients in the regression functions for the J selected applications so as to minimize a function H=ΣjWj|Fj−Gj|P, wherein Gj is an actual cost of ownership of application Aj for j=1, 2, . . . , J, wherein Σj represents a summation over j from j=1 to j=J, wherein P>0, and wherein Wj are predetermined weights subject to Wj>0, wherein said determining the coefficients comprises computing said coefficients, said computing comprising utilizing a numerical value of each parameter in each regression function relating to the J applications;
• calculating a regression total cost of ownership CREGR of the N applications, said calculating comprising summing the N regression functions associated with the N applications including utilizing the computed coefficients and a numerical value of each parameter relating to the N applications;
• ascertaining whether CREGR differs from CEST by more than a tolerance selected from the group consisting of a pre-determined cost differential and a predetermined percent, wherein CEST is an estimate of a total cost of ownership of the N applications;
• if CREGR differs from CEST by more than the tolerance then iterating until CREGR does not differ from CEST by more than the tolerance, wherein each iteration of the iterating comprises selectively re-executing the identifying, selecting, determining, calculating, ascertaining, and iterating steps;
• computing a first total cost of ownership (T1) of the N applications as a first sum of the N regression functions associated with the N applications, using the computed coefficients and a first numerical value for each parameter in the N regression functions;
• choosing at least one control parameter from the parameters in the N regression functions such that the remaining parameters in the N regression functions are unselected;
• changing the numerical value for the at least one control parameter from the first numerical value to a second numerical value;
• computing a second total cost of ownership (T2) of the N applications as a second sum of the N regression functions associated with the N applications, using the computed coefficients, the first numerical value for each unselected parameter, and the second numerical value for each selected parameter;
• determining that T2 is less than T1;
• transmitting output data to an output device of the computer system, wherein the output data is selected from the group consisting of T2, the second numerical value for each selected parameter, and a combination thereof;
• maintaining each application of the N applications and utilizing and maintaining all instances of each application of the N applications at the second total cost of ownership (T2) with each control parameter of the at least one control parameter being at its second numerical value.

A computer program product, comprising a computer readable storage medium having a computer readable program code stored therein, said computer readable program code containing instructions configured to be executed by a processor of a computer system to implement a method for reducing a cost of ownership of a portfolio of N software applications, said method comprising the steps of:

• identifying a regression function for each software application of the N software applications, wherein each regression function expresses an approximate cost of ownership of the application in terms of at least one parameter and at least one undetermined coefficient, wherein N is at least 2;
• selecting J applications A1, A2, . . . , Aj from the N applications such that a remaining N-J applications of the N applications are unselected, wherein J<N, wherein the coefficients in the regression functions for the N-J unselected applications constitute a subset of the coefficients in the regression functions for the J selected applications, and wherein Fj represents the approximate cost of ownership of the application Aj for j=1, 2, . . . , J;
• determining the coefficients in the regression functions for the J selected applications so as to minimize a function H=ΣjWj|Fj−Gj|P, wherein Gj is an actual cost of ownership of application Aj for j=1, 2, . . . , J, wherein Σj represents a summation over j from j=1 to j=J, wherein P>0, and wherein Wj are predetermined weights subject to Wj>0, wherein said determining the coefficients comprises computing said coefficients, said computing comprising utilizing a numerical value of each parameter in each regression function relating to the J applications;
• calculating a regression total cost of ownership CREGR of the N applications, said calculating comprising summing the N regression functions associated with the N applications including utilizing the computed coefficients and a numerical value of each parameter relating to the N applications;
• ascertaining whether CREGR differs from CEST by more than a tolerance selected from the group consisting of a pre-determined cost differential and a predetermined percent, wherein CEST is an estimate of a total cost of ownership of the N applications;
• if CREGR differs from CEST by more than the tolerance then iterating until CREGR does not differ from CEST by more than the tolerance, wherein each iteration of the iterating comprises selectively re-executing the identifying, selecting, determining, calculating, ascertaining, and iterating steps;
• computing a first total cost of ownership (T1) of the N applications as a first sum of the N regression functions associated with the N applications, using the computed coefficients and a first numerical value for each parameter in the N regression functions;
• choosing at least one control parameter from the parameters in the N regression functions;
• changing the numerical value for each control parameter from the first numerical value to a second numerical value such that a second total cost of ownership (T2) of the N applications is less than T1, wherein T2 is a second sum of the N regression functions associated with the N applications using the computed coefficients, the first numerical value for each unselected parameter, and the second numerical value for each chosen parameter, wherein T2 is minimized in the changing step with respect to the control parameters;
• transmitting the second numerical value for each control parameter to an output device of the computer system;
• maintaining each application of the N applications and utilizing and maintaining all instances of each application of the N applications at the second total cost of ownership (T2) with each control parameter of the at least one control parameter being at its second numerical value.

A computer system comprising a processor and a computer readable memory unit coupled to the processor, said memory unit containing program code configured to be executed by the processor to implement a method for reducing a cost of ownership of a portfolio of N software applications, said method comprising the steps of:

• identifying a regression function for each software application of the N software applications, wherein each regression function expresses an approximate cost of ownership of the application in terms of at least one parameter and at least one undetermined coefficient, wherein N is at least 2;
• selecting J applications A1, A2, . . . , Aj from the N applications such that a remaining N-J applications of the N applications are unselected, wherein J<N, wherein the coefficients in the regression functions for the N-J unselected applications constitute a subset of the coefficients in the regression functions for the J selected applications, and wherein Fj represents the approximate cost of ownership of the application Aj for j1, 2, . . . , J;
• determining the coefficients in the regression functions for the J selected applications so as to minimize a function H=ΣjWj|Fj−Gj|P, wherein Gj is an actual cost of ownership of application Aj for j=1, 2, . . . , J, wherein Σj represents a summation over j from j=1 to j=J, wherein P>0, and wherein Wj are predetermined weights subject to Wj>0, wherein said determining the coefficients comprises computing said coefficients, said computing comprising utilizing a numerical value of each parameter in each regression function relating to the J applications;
• calculating a regression total cost of ownership CREGR of the N applications, said calculating comprising summing the N regression functions associated with the N applications including utilizing the computed coefficients and a numerical value of each parameter relating to the N applications;
• ascertaining whether CREGR differs from CEST by more than a tolerance selected from the group consisting of a pre-determined cost differential and a predetermined percent, wherein CEST is an estimate of a total cost of ownership of the N applications;
• if CREGR differs from CEST by more than the tolerance then iterating until CREGR does not differ from CEST by more than the tolerance, wherein each iteration of the iterating comprises selectively re-executing the identifying, selecting, determining, calculating, ascertaining, and iterating steps;
• choosing at least one control parameter from the parameters in the N regression functions such that the remaining parameters in the N regression functions are unselected;
• specifying permissible values for each control parameter of the at least one control parameter, wherein the permissible values for each control parameter are independently selected from the group consisting of a discrete set of values, at least one continuous range of values, and a combination thereof
• expressing a total cost of ownership (T) of the N applications as a sum of the N regression functions associated with the N applications such that T is a function of the at least one control parameter, wherein the computed coefficients and a numerical value for each unselected parameter appears in T;
• and
• determining a value for each control parameter of the at least one control parameter so as to minimize T subject to the permissible values;
• transmitting the value for each control parameter to an output device of the computer system;
• maintaining each application of the N applications and utilizing and maintaining all instances of each application of the N applications at the minimized T with each control parameter of the at least one control parameter being at its determined value at which T is minimized.

A computer system comprising a processor and a computer readable memory unit coupled to the processor, said memory unit containing program code configured to be executed by the processor to implement a method for reducing a cost of ownership of a portfolio of N software applications, said method comprising the steps of:

• identifying a regression function for each software application of the N software applications, wherein each regression function expresses an approximate cost of ownership of the application in terms of at least one parameter and at least one undetermined coefficient, wherein N is at least 2;
• selecting J applications A1, A2, . . . , Aj from the N applications such that a remaining N-J applications of the N applications are unselected, wherein J<N, wherein the coefficients in the regression functions for the N-J unselected applications constitute a subset of the coefficients in the regression functions for the J selected applications, and wherein Fj represents the approximate cost of ownership of the application Aj for j>1, 2, . . . , J;
• determining the coefficients in the regression functions for the J selected applications so as to minimize a function H=ΣjWj|Fj−Gj|P, wherein Gj is an actual cost of ownership of application Aj for j=1, 2, . . . , J, wherein Σj represents a summation over j from j=1 to j=J, wherein P>0, and wherein Wj are predetermined weights subject to Wj>0, wherein said determining the coefficients comprises computing said coefficients, said computing comprising utilizing a numerical value of each parameter in each regression function relating to the J applications;
• calculating a regression total cost of ownership CREGR of the N applications, said calculating comprising summing the N regression functions associated with the N applications including utilizing the computed coefficients and a numerical value of each parameter relating to the N applications;
• ascertaining whether CREGR differs from CEST by more than a tolerance selected from the group consisting of a pre-determined cost differential and a predetermined percent, wherein CEST is an estimate of a total cost of ownership of the N applications;
• if CREGR differs from CEST by more than the tolerance then iterating until CREGR does not differ from CEST by more than the tolerance, wherein each iteration of the iterating comprises selectively re-executing the identifying, selecting, determining, calculating, ascertaining, and iterating steps;
• computing a first total cost of ownership (T1) of the N applications as a first sum of the N regression functions associated with the N applications, using the computed coefficients and a first numerical value for each parameter in the N regression functions;
• choosing at least one control parameter from the parameters in the N regression functions such that the remaining parameters in the N regression functions are unselected;
• changing the numerical value for the at least one control parameter from the first numerical value to a second numerical value;
• computing a second total cost of ownership (T2) of the N applications as a second sum of the N regression functions associated with the N applications, using the computed coefficients, the first numerical value for each unselected parameter, and the second numerical value for each selected parameter;
• determining that T2 is less than T1;
• transmitting output data to an output device of the computer system, wherein the output data is selected from the group consisting of T2, the second numerical value for each selected parameter, and a combination thereof;
• maintaining each application of the N applications and utilizing and maintaining all instances of each application of the N applications at the second total cost of ownership (T2) with each control parameter of the at least one control parameter being at its second numerical value.

A computer system comprising a processor and a computer readable memory unit coupled to the processor, said memory unit containing program code configured to be executed by the processor to implement a method for reducing a cost of ownership of a portfolio of N software applications, said method comprising the steps of:

• identifying a regression function for each software application of the N software applications, wherein each regression function expresses an approximate cost of ownership of the application in terms of at least one parameter and at least one undetermined coefficient, wherein N is at least 2;
• selecting J applications A1, A2, . . . , Aj from the N applications such that a remaining N-J applications of the N applications are unselected, wherein J<N, wherein the coefficients in the regression functions for the N-J unselected applications constitute a subset of the coefficients in the regression functions for the J selected applications, and wherein Fj represents the approximate cost of ownership of the application Aj for j=1, 2, . . . , J;
• determining the coefficients in the regression functions for the J selected applications so as to minimize a function H=ΣjWj|Fj−Gj|P, wherein Gj is an actual cost of ownership of application Aj for j=1, 2, . . . , J, wherein Σj represents a summation over j from j=1 to j=J, wherein P>0, and wherein Wj are predetermined weights subject to Wj>0, wherein said determining the coefficients comprises computing said coefficients, said computing comprising utilizing a numerical value of each parameter in each regression function relating to the J applications;
• calculating a regression total cost of ownership CREGR of the N applications, said calculating comprising summing the N regression functions associated with the N applications including utilizing the computed coefficients and a numerical value of each parameter relating to the N applications;
• ascertaining whether CREGR differs from CEST by more than a tolerance selected from the group consisting of a pre-determined cost differential and a predetermined percent, wherein CEST is an estimate of a total cost of ownership of the N applications;
• if CREGR differs from CEST by more than the tolerance then iterating until CREGR does not differ from CEST by more than the tolerance, wherein each iteration of the iterating comprises selectively re-executing the identifying, selecting, determining, calculating, ascertaining, and iterating steps;
• computing a first total cost of ownership (T1) of the N applications as a first sum of the N regression functions associated with the N applications, using the computed coefficients and a first numerical value for each parameter in the N regression functions;
• choosing at least one control parameter from the parameters in the N regression functions;
• changing the numerical value for each control parameter from the first numerical value to a second numerical value such that a second total cost of ownership (T2) of the N applications is less than T1, wherein T2 is a second sum of the N regression functions associated with the N applications using the computed coefficients, the first numerical value for each unselected parameter, and the second numerical value for each chosen parameter, wherein T2 is minimized in the changing step with respect to the control parameters;
• transmitting the second numerical value for each control parameter to an output device of the computer system;
• maintaining each application of the N applications and utilizing and maintaining all instances of each application of the N applications at the second total cost of ownership (T2) with each control parameter of the at least one control parameter being at its second numerical value.