Macroscopic model for large scale software system and its behavior

1. An apparatus for modeling a large software system, comprising:

a complex system, representative of the software system, composed of a large number of parts, which interact and work as an integrated whole system;

a processor;

a memory, coupled to the processor, the memory storing instructions adapted to be executed by the processor, the instructions including:

providing a logical system having a number of variables and operators that act on the variables, and a state space that is a product of all values that every variable assumes and that has a dimensional state which is a combination of possible values of the variables;

determining impacts of the variables on states to effect state transitions;

imposing an indicator on the impacts of variables on performance of the system during a given time interval; and

reducing interdependencies between variables to reduce state transition probabilities and correspondingly an average number of states traversed thereby improving reliability of the system through reduction of occupied states; and

wherein a statistical ensemble of variables is embedded in a logical environment that controls microscopic behavior of the variables that is representative of the complex system composed of a large number of parts, which interact and work as an integrated whole system; and

wherein macroscopic attributes are derived from the microscopic behavior; and

wherein, when the probability of the complex system being in a predefined state is high, the software is considered to be hardening.

2. The model according to claim 1 , wherein macroscopic attributes of the software system are at least logical volume, logical pressure, and logical temperature and are respectively analogous to volume, pressure, and temperature in a physical gas system.

3. The model according to claim 1 , wherein the software hardening is a framework for conceptual unification of software reliability improvement techniques including at least one of modularity, object orientation, and software reuse.

4. A macroscopic model of a large software system, comprising:

a processor;

a memory, coupled to the processor, the memory storing instructions adapted to be executed by the processor, the instructions including:

providing a logical system having a number of variables and operators that act on the variables, and a state space that is a product of all values that every variable assumes and that has a dimensional state which is a combination of possible values of the variables;

determining impacts of the variables on states to effect state transitions;

imposing an indicator on the impacts of variables on performance of the system during a given time interval; and

reducing interdependencies between variables to reduce state transition probabilities and corresponding an average number of states traversed thereby improving reliability of the system through reduction of occupied states.

5. The model according to claim 4 , wherein the system has a plurality of variables, and wherein the logical temperature of the system is an average dependency of all variables in the system.

6. The model according to claim 4 , wherein for constant logical temperature the logical pressure and logical volume are inversely related.

7. The model according to claim 4 , wherein when local activity of the software system is increased while global activity of the software system is reduced, keeping total activity constant, signifies an increase in logical pressure, then number of states traversed decreases leading to a software with reduced logical volume.

8. The model according to claim 4 , wherein when logical temperature is decreased, both the logical volume and the logical pressure decrease.

9. The model according to claim 4 , wherein when interdependencies of variables are reduced, potential impacts of the variables on each other are correspondingly decreased leading to a decrease in logical volume and reduced logical pressure.

10. The model according to claim 4 , wherein decreasing the logical temperature and the logical volume, results in decreasing the entropy.

11. The model according to claim 4 , wherein the logical volume is a sum of a potential impact contribution over all states admissible for traversal by a variable in the system.

12. The model according to claim 4 , wherein the logical pressure is the density of the variables with respect to a total number of states traversed.

13. The model according to claim 4 , wherein apparatus further includes a logical temperature algorithm that identifies the logical temperature.

14. The model according to claim 4 , wherein apparatus further includes a logical pressure algorithm that identifies the logical pressure.

15. The model according to claim 4 , wherein apparatus further includes a logical volume algorithm that identifies the logical volume.

16. A macroscopic model of a large software system, comprising:

a complex system composed of a large number of parts, which interact and work as an integrated whole system;

a processor;

a memory, coupled to said processor, said memory storing instructions adapted to be executed by said processor, the instructions including:

providing a logical system having a number of variables and operators that act on the variables, and a state space that is a product of all values that every variable assumes and that has a dimensional state which is a combination of possible values of the variables;

determining impact levels of the variables on states to effect state transitions;

imposing an indicator on the impacts of variables on performance of the system during a given time interval; and

reducing interdependencies between variables to reduce state transition probabilities and correspondingly an average number of states traversed thereby improving reliability of the system through reduction of occupied states;

wherein relationships are formed for distribution of n i variables with respect to impact levels E i of the large software system that is representative of the complex system composed of a large number of physical parts, which interact and work as an integrated whole system;

n i =( N/z ) g i . exp(−β. E i )

and

z

=

∑

i

⁢

g

i

·

exp

⁢

⁢

(

-

β

·

E

i

)

where

N=total number of variables

z=Partition function (we retain this term from thermodynamics)

β=constant dependent upon the environment

g i =average number of possible states at the impact level E i .

17. The model according to claim 16 , wherein the logical volume is a sum of potential impact contribution over all states admissible for traversal by a variable;

V

=

∑

i

⁢

s

i

·

ɛ

i

=

ɛ

·

∑

i

⁢

s

i

where,

V=logical volume

s i =the number of system states traversed by the variable x i

ε i =potential impact of the variable x i

ε=average potential impact of all variables

i=index enumerating all the variables.

18. The model according to claim 16 , wherein the logical temperature of the system is an inverse of the constant β and is a function of an average dependency of all variables in the system.

19. The model according to claim 16 , wherein the logical pressure is the density of the variables with respect to a total number of states traversed;

P

=

N

/

(

∑

all

·

var

⁢

s

i

)

=

N

·

ɛ

/

V

where,

V=logical volume

s i =the number of system states traversed by the variable x i

ε i =potential impact of the variable x i

ε=average potential impact of all variables

i=index enumerating all the variables.