 |
Claims  |
|
|
What is claimed is:
1. A method utilizing a computer means for automatically converting
structured analysis tool outputs from one or more of a variety of sources
into an integrated simulation model, said structured analysis tool outputs
residing in computer databases and containing at least one data flow
diagram representing a functional system, said at least one data flow
diagram having at least one data flow diagram element and at least one
dataflow, said method comprising the steps of:
(a) generating a standardized METAfile from said databases, said METAfile
containing inter-element relations in a format suitable for reconstruction
into a simulation model format on another host computer having a
simulation generation program residing therein, said METAfile
automatically generated by:
(i) accessing said computer databases of said structured analysis tools
outputs;
(ii) accessing said at least one data flow diagram element from said at
least one data flow diagram;
(iii) ascertaining and storing a first set of information related to said
data flow diagram element;
(iv) determining if said dataflows are entering or exiting said data flow
diagram element;
(v) ascertaining and storing a second set of information related to said
dataflow; and
(vi) determining if there are any remaining said dataflows, said data flow
diagram elements, and said data flow diagrams, if there are remaining said
dataflows repeat steps (iv) through (vi), if there are remaining said data
flow diagram elements repeat steps (ii) through (vi) and if there are
remaining said data flow diagrams repeat steps (i) through (vi);
(b) transporting said standardized METAfile into said another host computer
having said simulation generation program and automatically generating a
simulation model from said METAfile, said simulation model comprising
simulation source code statements; and,
(c) integrating actual source code into said simulation model by
automatically replacing components said simulation source code statements
with actual source code to form said integrated simulation model.
2. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 1, wherein said
step of accessing the data base for a data flow diagram comprises
searching through the data base of said structured analysis tools to
locate the generated data flow diagrams.
3. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 2, wherein said
step of accessing a data flow diagram element from said data flow diagram
comprises searching through the data base of said structured analysis
tools to locate said data flow diagram element corresponding to said data
flow diagram.
4. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 3, wherein said
step of ascertaining and storing said first set of information related to
said data flow diagram element comprises searching through the data base
of said structured analysis tool to locate all relevant information
pertaining to said data flow diagram element.
5. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 4, wherein said
all relevant information comprises a block type and one or more processing
time parameters associated with said data flow diagram element.
6. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 5, wherein said
block type and said processing time parameters are entered into a
parameter list for said data flow diagram element.
7. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 4, wherein said
step of determining if dataflows are entering or exiting said data flow
diagram element comprises searching through the data base of said
structured analysis tools to locate all input and output dataflows
associated with said data flow diagram element.
8. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 7, wherein said
step ascertaining and storing said second set of information related to
said dataflow comprises searching through the data base of said structured
analysis tools to locate all relevant information pertaining to said
dataflows.
9. The method for automatically converting structured analysis tool outputs
into an intergrated simulation model according to claim 8, wherein said
all relevant information comprises a data rate and a data size of said
dataflows.
10. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 9, wherein
said data rate and said data size are entered into a parameter list for
said dataflows.
11. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 8, wherein
said step of determining if there are any remaining dataflows, data flow
diagram elements and data flow diagrams comprises searching through the
data base of said structured analysis tools to determine if there are any
remaining dataflows associated with a particular data flow diagram
element, if there are any remaining data flow diagram elements associated
with a particular data flow diagram and if there are any remaining data
flow diagrams to convert.
12. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 11,
wherein said step of automatically generating a simulation model from said
METAfile comprises the steps of:
(a) accessing said METAfile for a METAfile element;
(b) determining a block type of said METAfile element;
(c) converting said METAfile element into a corresponding simulation
component based on the particular block type of said METAfile element;
(d) determining if said dataflows are entering or exiting said METAfile
element;
(e) ascertaining and storing said second set of information related to said
dataflows; and
(f) determining if there are any remaining said dataflows, if there are
remaining said dataflow steps (a) through (f) are repeated.
13. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 12,
wherein said step of accessing said METAfile for a METAfile element
comprises searching through said METAfile to locate said METAfile element.
14. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 13,
wherein said step of determining the block type of said METAfile element
comprises searching the METAfile to locate said parameter list for said
data flow diagram element to determine the particular block type of said
METAfile element.
15. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 14,
wherein said step of converting said METAfile element into a corresponding
simulation component based on the particular block type of said METAfile
element comprises creating a plurality of simulation blocks of said
simulation source code based on whether the block type is determined to be
a SOURCE, PROCESS, or SINK.
16. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 15,
wherein said step of determining if dataflows are entering or exiting said
METAfile element comprises the steps of:
(a) searching through said METAfile to locate all said input and output
dataflows associated with said METAfile element; and
(b) determining if said dataflow is an input dataflow or an output
dataflow.
17. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 16,
wherein said step of ascertaining and storing said second set of
information related to said dataflows comprises searching the METAfile to
locate said parameter list for said dataflow to determine all relevant
information pertaining to said dataflow.
18. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 17,
wherein said all relevant information comprises the data rate and the data
size of said dataflow, said relevant information is stored in tabular
form.
19. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 18,
wherein said step of determining if there are any remaining dataflows
comprises searching through the METAfile to determine if there are no
remaining dataflows and the METAfile contains more METAfile elements, then
the steps of claim 12 are repeated, if there are no remaining dataflows
and the METAfile contains no more METAfile elements then the process is
terminated.
20. The method for automatically converting structured analysis tool
outputs into an intergrated simulation model according to claim 19,
wherein the step of automatically integrating code into said simulation
model by replacing the simulation component representing code with actual
code comprises replacing one of said simulation blocks with a HELP block,
said HELP block accesses the actual source code that performs the specific
function.
21. A software design and modelling system for automatically generating
integrated simulation/source code models from structured analysis tools
outputs from one or more of a variety of sources, said structured analysis
tools outputs residing in computer databases and containing at least one
data flow diagram representing a functional system, said at least one data
flow diagram having one or more data flow diagram elements and one or more
dataflows entering or exiting said data flow diagram elements, said
software design and modelling system comprising:
(a) means for accessing said computer databases of said structured analysis
tools outputs to obtain therefrom said one or more data flow diagram
elements from said at least one data flow diagram;
(b) means for searching through said databases to obtain one or more first
sets of information corresponding to each of said one or more data flow
diagram elements, each of said first sets of information comprising block
type and processing time parameters associated with each respective said
data flow diagram element;
(c) means for determining if said one or more dataflows are entering or
exiting each of said one or more of said data flow diagram elements, said
means for searching through said databases to obtain therefrom one or more
second sets of information corresponding to said one or more dataflows
upon a determination that said one or more dataflows exists, each of said
second sets of information comprising data size and data rate parameters
associated with each respective said dataflows;
(d) means for automatically generating a standardized METAfile from said
one or more dataflow diagram elements, said one or more dataflows entering
or exiting therefrom, and said first sets of information and said second
sets of information corresponding therewith, said standardized METAfile
containing one or more interrelated elements in a format suitable for
reconstruction into a simulation model format on any host computer;
(e) means for transporting said standardized METAfile into said another
host computer;
(f) means residing on said another host computer for automatically
generating a simulation model from said standardized METAfile, said
simulation model comprising simulation source code statements; and,
(g) means for automatically integrating actual source code into said
simulation model by replacing components said simulation source code
statements with said actual source code to form said integrated
simulation/source code model.
22. The software design and modelling system of claim 21, wherein said at
least one structured analysis tools output is a data dictionary.
23. The software design and modelling system of claim 21, wherein said at
least one structured analysis tools output is a mini-specification.
24. The software design and modelling system of claim 23, wherein said
intergrated simulation/source code model is an operational behavioral
simulation model having a combination of simulation code and application
source code for testing the application source code in a simulation
environment. |
|
 |
|
 |
Claims |
|
|
|
Description |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a method for automatically generating a
simulation model from previously created structured analysis tools, and
more particularly, to a method for automatically generating simulation
models from previously created data flow diagrams and utilizing this
simulation model as a tool for the design, integration and testing of the
particular system being implemented by replacing components of the
simulation model with the actual hardware/software components in short
incremental steps. The simulation model can be created using any of the
standard simulation languages on a plurality of host computers.
2. Discussion of the Prior Art
The fields of structured analysis and simulation modeling are an
increasingly important technology due to the ever increasing complexity of
systems to be designed. The purpose of both structured analysis and
simulation is to design a realistic and accurate model of the particular
system to be designed. The design of a realistic and accurate system is
accomplished by performing an allocation of system functions. Structured
analysis performs an allocation of system functions in order to implement
the system, and simulation models perform an allocation of system
functions in order to execute and analyze a model of the system.
Basically, in the design of any multifaceted system, the use of structured
analysis and simulation is indispensable regardless of how simplistic the
system may appear to be. The concept behind these procedures is to first
break the problem or system design requirements into smaller and more
manageable parts, find a solution for each part, and then rebuild the
parts into a complete and functioning system.
Analysis is the study of a problem prior to taking action to solve the
problem. Structured analysis is a particular form of analysis which uses a
set of standardized tools to partition the problem into smaller parcels
such that they become more manageable. The primary tools of structured
analysis are data flow diagrams, data dictionaries, and
mini-specifications. Structured diagramming techniques support a top-down,
structured development approach to problem solving with various levels of
decomposition and thus various levels of detail for the particular system
can be achieved. Early structured analysis tools were largely long hand
procedures utilizing pencil and paper techniques. However, with the advent
of widely available, multipurpose graphics terminals, structured analysis
diagramming has become a quick and efficient automated process. A variety
of vendors, including Cadre and Tektronix, have easy to use, commercially
available structured analysis programs. These automated tools provide an
efficient means for creating and updating structured analysis outputs
which are the first step in generating a finished product from the initial
problem or design requirements.
Simulation and modelling are widely accepted techniques whereby prior to
actually building or constructing the particular product, a simulation or
model is constructed to see if the product functions as envisioned and
whether or not it is a feasible design. Modelling is the older of the two
techniques and involves actually building a prototype of the product. This
prototype or model may or may not be a fully working model or even built
to scale, but rather, it typically is used to be representative of the
particular product. This model would be subjected to various tests in
order to determine if the design was sound and feasible. As time
progressed, and the use of computers was becoming more prevalent, software
routines were written to simulate the workings of hardware and software
components that comprised the particular product. Today, simulation is a
design-aid tool that has been in existence for nearly thirty years and is
a highly exact science. This breakthrough has had an incredible impact on
designing techniques; namely, by being able to simulate the product on a
computer, a model that in all probability would not function, would not
have to be built. The product could be simulated on the computer, tested
and have a majority of the bugs worked out before the prototype was built.
Although simulation time is somewhat expensive and time consuming, it
represents an improvement in having to continuously build prototypes to
work out the problems encountered.
The design of a particular system starts with the basic system concepts,
the general idea of what is to be designed and what is expected of the
design. Once the system concepts are fairly well established, the design
of the system follows two diverging paths; namely, the hardware
development path and the software development path. Basically, the two
paths are similar in concept, but differ in the actual implementation of
the various stages along the paths. The software development path starts
with the system software requirements analysis and proceeds through
software requirements analysis, preliminary design, detailed design,
coding unit and CSC integration testing, and ends with the computer
software configuration item testing. The hardware development path starts
with the system hardware requirements analysis and proceeds through
hardware requirements analysis, preliminary design, detailed design,
fabrication and ends with the hardware configuration item testing. The
basic tool utilized to go from the system concepts stage to the
configuration item testing stage is structured analysis. Once this phase
is completed, the next step would be to use system modelling through
simulation to test the system, and finally to build, fabricate or code the
finished product, namely, the completed and functional system. Currently,
the use of structured analysis and system modelling have been separate and
distinct steps in the design process, one to be used after the desired
result is achieved with the other.
SUMMARY OF THE INVENTION
The present invention is directed to a method for automatically converting
structured analysis tool outputs into an executable simulation model. The
method comprises the steps of automatically generating a METAfile from a
data base of the structured analysis tools and automatically generating a
simulation model from the generated METAfile. The METAfile contains
inter-element relations in a format suitable for reconstruction into
simulation model format and the simulation model comprises simulation
source code statements. The method for automatically converting structured
analysis tool outputs into an executable simulation model further
comprises the step of automatically integrating actual code into said
simulation model by replacing the simulation component representing code
with actual code. As a consequence of the entire procedure, at the end of
the conversion the actual system will have been implemented and tested
using the model as the basis for the implementation.
The method for automatically converting structured analysis tool outputs
into an executable simulation model is, as described above, a two step
process. The first step in the process is converting the outputs from a
variety of structured analysis tools into a standard METAfile. A METAfile
Conversion Program, MCP, is utilized to accomplish this first step. The
MCP accesses the data bases of the various tools and collects the
necessary data. Once the data is collected, the MCP reorganizes the data
into a standard METAfile format for further use. The second step in the
process is the conversion of the METAfile into an executable simulation
model. A Simulation Generation Program, SGP, is utilized to accomplish
this second step. The SGP accesses the METAfile and in a step by step
process converts the various blocks of data contained in the METAfile into
simulation code utilizing any of the standard simulation languages.
This method is feasible because of the similarity in relationship between
the components used in structured analysis to define the functions of the
system being designed, and the components used in simulation to define the
functions of the system being simulated. The most common structured
analysis tool used is the data flow diagram. It comprises five basic
elements namely; sources, sinks, transformations or processes, data flows
and files. As the description of this invention progresses, it will be
shown that there is a one-to-one correspondence between the number of
elements in data flow diagrams and simulation models, and it is this
correspondence that makes the conversion possible. It should be noted that
any and all structured analysis components can and will be used, as
required, to fully define the system being simulated, including process
specifications, structure charts, etc.
The present invention provides a method for automatically generating a
source code simulation model from previously created data flow diagrams.
The executable simulation model can be implemented in any of the standard
simulation languages such as Simscript, Network II.5 and GPSS. In having
the simulation model implemented in any of the standard languages,
compliance with any government simulation requirement can be met with
increased ease and speed. Once the METAfile is created, using the METAfile
Conversion Program, from the data flow diagrams, it can be transferred to
any host computer system having the simulation generation program
resident. The automatic conversion direct from the data flow diagrams
assures a correct and accurate representation of the system to be
designed. The entire process is less costly then the current approach for
creating comparable models because a simulation expert will not be needed
to construct the model. In addition, since it is an automatic process, the
time to generate the model will be greatly reduced and thus the design of
the model can be deferred till the data flow diagrams become available.
The present invention also provides a means and method for using the
simulation model as a tool for the design, integration and testing of the
system being implemented by replacing components of the simulation model
with the actual hardware/software components in short incremental steps.
The incremental incorporation of the actual components, hardware/software,
of the system into the model simplifies overall system integration and
testing. This incremental process provides for a time saving and cost
effective method of accurately testing system operation one step at a
time.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram illustrating the method utilized in the present
invention;
FIG. 2 is a structured analysis context diagram for a given system;
FIG. 3 is a structured analysis top-level data flow diagram for the given
system of FIG. 2;
FIG. 4 is a structured analysis consolidated data flow diagram representing
the consolidation of the context diagram and the top-level data flow
diagram for the given system;
FIG. 5 is a structured analysis child data flow diagram of process PR1 in
the top-level data flow diagram for the given system;
FIG. 6 is a structured analysis consolidated data flow diagram representing
the consolidated diagram of FIG. 4 and the child data flow diagram of FIG.
5;
FIG. 7 is a flow chart representation of the METAfile Conversion Program of
the present invention;
FIG. 8 is a data flow diagram for a telemetry system;
FIG. 9 is a GPSS Simulation Model for the telemetry system of FIG. 8;
FIG. 10 is a flow chart representation of the Simulation Generation Program
of the present invention;
FIG. 11 is a GPSS Simulation Model for the telemetry system of FIG. 8
integrating object code for a given model block.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The present invention is directed to a method for automatically generating
a simulation model from previously created data base outputs from a
variety of structured analysis tools. The method further includes a
procedure for using this simulation model as a tool for the design,
integration and testing of the system being implemented by replacing
components of the simulation model with the actual hardware/software
components in short incremental steps. As a consequence of the entire
procedure, at the end of the conversion the actual system will have been
implemented and tested using the model as the basis for the
implementation.
Analysis in general terms is the study of a particular problem prior to
taking any action towards solving the problem. Structured analysis is a
form of analysis which utilizes certain tools to partition the problem
into smaller parcels such that they are more easily managed. The basic
tools of structured analysis are data flow diagrams, data dictionaries,
and mini-specifications. The data flow diagram is basically a network
representation of a system, wherein the system may be automated, manual or
a combination of automated and manual. The data flow diagram portrays the
system in terms of its component parts. The data dictionary contains the
set of definitions of all the data flow diagram elements. The
mini-specification is a tool for defining small portions of the data flow
diagrams. The mini-specification can use such other tools as structured
English, decision tables and decision trees to describe the logic and
policy in the data flow diagrams. A complete description of the
aforementioned structured analysis tools is given as the workings of the
invention are disclosed.
Simulation is a design-aid tool that has been in existence in the computer
field for almost thirty years. Simulation packages are comprised of
programs that enable a user to describe or model a system, hardware and
software, prior to implementation in order to provide vital information
about the system to assist in decision making. Simulation models provide
assistance in design decision making by highlighting system problem areas,
determining hardware/software bottlenecks, providing critical system
statistical information and providing processor and memory sizing
information.
There are basically two types of simulations; namely, "behavioral" and
"real time". "Real time" simulations are concerned with testing actual
system hardware and software in a test environment that approximates the
actual system environment to determine if the system operates as specified
and meets requirements. This entails implementing a test environment that
interacts with the system under test similarly to the manner in which the
actual interfacing systems would interact. This test environment would
consist of fairly well-defined hardware and software components.
"Behavioral" models are concerned with simulating the behavior of the
system to be implemented in order to determine system traffic and possible
bottlenecks. This information is then used to assist in determining design
criteria. To create the behavioral model, system processes, storage
devices and transfer devices are defined in relation to one-another, and
the model developed from these relationships. Other information is
required to complete the model, such as data input distributions that are
required to determine data arrival times, expected data processing times,
data base organization and size, and other additional information as
required to complete the model. As more information about the system's
characteristics are uncovered, the model is refined, and ultimately
resembles and acts similarly to the proposed system.
The present invention is concerned with developing an operational
"behavioral" simulation model, with all required inputs/outputs and
processing elements by utilizing previously created structured analysis
data bases comprised of data flow diagrams, data dictionaries and other
components of the structured analysis data base as provided by those
automated systems. The present invention also allows for the replacement
of components of the simulation model with the actual software code as it
is completed and tested. In this manner, the model will incrementally be
replaced by the actual system components until it becomes that system,
fully integrated and tested.
A data flow diagram is comprised of five basic elements, sources of data
into the system, sinks or recipients of data leaving the system,
transforms or processes which convert data, data flows which define the
data passing between transformations, and files in which data is stored.
Simulation models are comprised of three basic elements, processing
elements which are characterized by their instruction set and cycle time,
transfer devices which are characterized by their data transfer rate, data
transfer protocol and connections, and data storage devices which are
characterized by capacity, access time, and access method. The first two
elements of the data flow diagrams, sources and sinks can be readily
simulated by the use of GPSS GENERATE and TERMINATE blocks. A complete
description of these blocks is given in subsequent paragraphs. The three
remaining elements of a data flow diagram can be simulated with the three
elements comprising a simulation model. Basically, processing elements are
substituted for transformations, data transfer devices are substituted for
data flows and data storage devices are substituted for files. As a
result, it is possible to design a behavioral simulation model utilizing
data flow diagrams as the basis for the design. Given that it is possible
to design a simulation model utilizing data flow diagrams, then it is also
possible that the system be designed so it is capable of automatically
generating a behavioral simulation model from data flow diagrams with a
new model capable of being produced automatically from the structural
analysis tool used to produce the data flow diagram code. This enables
updating of the simulation model automatically to the latest version of
the data flow diagram model, ensuring the model is truly representative of
the actual system. One such-tool that could be adapted is the Teamwork
System used for generating Yourdon-like data flow diagrams with an
interface implemented between it and the simulation model.
Referring to FIG. 1, there is shown a block diagram of the overall
simulation conversion process. Block 10 represents the various structured
analysis inputs that are to be transformed into a simulation model. The
structured analysis inputs of block 10 are entered into the data base
through block 20 which represents the various automated structured
analysis tools. The data base output of the various systems included in
block 20 are input into block 100. Block 100 represents the output of a
plurality of METAfile Conversion Programs utilized to automatically
convert the data base created by any of the automated tools of block 20 to
a standard METAfile format. The METAfile will contain inter-element
relationships in a format suitable for reconstruction into simulation
model format. The METAfile is shown as block 30 in the figure. The
METAfile 30 can be realized on a variety of computers such as the IBM
Series 14300 Model 1 and the DEC VAX 8600. The METAfile 30 is then fed
into the Simulation Generation Program represented by block 400. The
Simulation Generation Program 400 converts the METAfile 30 into a basic
simulation model which can be in any of the standard simulation source
languages. The simulation model is shown in the figure as block 40. The
final step in the overall process, the integration of actual software into
the simulation model, is represented in the figure by block 44. As can be
seen from the figure, block 44 is comprised of any of the standard
simulation source languages from block 40 and actual source code from
block 42. The actual source code can be in any language such as FORTRAN or
ADA.
A METAfile is a standardized output file produced after analysis of the
structured analysis data base files that are created by the various
vendors in order for them to regenerate the respective data flow diagrams,
data dictionaries, mini-specifications, and state transition diagrams that
are used in structured analysis design. These data base files are analyzed
by the METAfile Conversion Program 100, MCP, shown in FIG. 7, which
reformats the files so they may be used as input to the Simulation
Generation Program, SGP 400, shown in FIG. 10. A complete description of
these two programs is given in subsequent paragraphs. The reason for
standardizing the output of the MCP 100 is so that which ever data base
input source, Teamwork, Tektronix, or other system, is used, the converted
output from each of those systems will be in precisely the same format so
that only one SGP is necessary to compile that output into the desired
simulation source language. In having a standardized METAfile, only one
SGP need to be coded regardless of the input source; however, one MCP has
to be coded for each different structured analysis vendor input. The
reason for having different MCP's for different structured analysis tools
is because there are differences between various vendors tools in the way
they internally store their information. Once the METAfile is created, it
can be transported and used as input on any processing system that has an
SGP running on it. Numerous examples of METAfiles are given in subsequent
paragraphs.
Referring now to FIG. 2, there is shown a typical structured analysis
context diagram 50, which is one form of a data flow diagram. The system
60 to be designed, denoted SY, receives inputs from a source 70, denoted
SO, and outputs data to a sink 80, denoted SI. Dataflow 1, DF1,
represented by vector 71 and Dataflow 2, DF2, represented by vector 79
represent the data being transferred into and out of SY 60. The context
diagram 50 follows the standard data flow diagram representation for the
various elements; namely, processes and systems are represented by circles
or bubbles, data sources and data sinks are represented by boxes, and data
flows are represented by named vectors. The one additional element which
is not shown in this figure is a data file which is represented by
straight parallel lines or a single straight line. This context diagram 50
represents a simplistic system and the only difference between this system
and the most complicated system is the number of sources and sinks that
lead into and out of the system respectively. Although named SO and SI,
the names of the actual source 70 and sink 80 will be distinguishable from
one another by a convenient naming scheme. The context diagram 50 is drawn
on a graphics terminal by the Cadre Teamwork tool which as stated
previously is a standard structured analysis tool. After completion, the
diagram is saved and stored by Teamwork in its internal files. These
internal files are used by Teamwork when the context diagram 50 is
recalled for purposes of modifying the diagram or just for viewing it. In
a manner similar to the way in which Teamwork uses these internal files to
reconstruct the context diagram, the MCP of FIG. 7 uses them to generate a
standard METAfile.
Referring to FIG. 3, there is shown a top-level data flow diagram 52 of the
system, 60, represented by the context diagram of FIG. 2. Whereas the
context diagram 50 depicts the domain or scope of the task, the top level
data flow diagram 52 depicts the specific functions of which the system
60, in the context diagram 50 is comprised. As is shown in FIG. 3, the
system 60 is comprised of two functions, process 1, PR1, and process 2,
PR2, respectively. Utilizing the standard convention for data flow
diagrams, PR1 is represented by a first bubble 62 and PR2 is represented
by a second bubble 64. An internal file 66, or data base denoted FI1, is
represented in the dataflow diagram by two parallel lines. DF1 71 is shown
entering process 1 bubble 62 and DF2 79 is shown exiting process 2 bubble
64. The remaining dataflows, DF3 73, DF4 75 and DF5 77 are all internal to
system 60 of FIG. 2. It is important to note that only the data flows
input to the specific process being decomposed are depicted in the
diagram.
The system user need not convert each and every data flow diagram prior to
METAfile conversion. The system user is given the capability of generating
a METAfile for any level of decomposition desired. As stated previously,
in order to generate a METAfile, the MCP of FIG. 7 analyzes the internal
Teamwork files and accesses those data base files required to rebuild the
desired data flow diagrams. The MCP initiates conversion with the context
diagram 50, or other specified data flow diagram. For each element that is
contained in a data flow diagram, the MCP generates a series of records
containing the name of the element and its various input and output data
flows. Table 1 given below represents the METAfile for the context diagram
50 of FIG. 2, this METAfile is the output of the first step in the MCP
conversion process. The element name, wherein an element may be a source,
sink, process or file, is depicted followed by the data flows entering or
exiting the element.
TABLE 1
______________________________________
METAfile for Context Diagram
______________________________________
SO (SOURCE, DATA GENERATION TIME)
DF1 ("O", SY, DATA RATE, DATA SIZE)
SY (PROCESS, PROCESS TIME)
DF1 ("I", SO, DATA RATE, DATA SIZE)
DF2 ("O", SI, DATA RATE, DATA SIZE)
SI (SINK)
DF2 ("I", SY, DATA RATE)
______________________________________
The basic pa | | |
