Volume 04 Issue 06-2024
50
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
ABSTRACT
The article discusses the justification and selection of criteria for evaluating the effectiveness of software used in
control computing facilities (CCF) for data transmission systems (DTS). It focuses on two main types of CCS with a
multi-step program execution pattern and a limited number of load sources. The effectiveness of in-house software
by using statistical modeling of CCF functioning is also evaluated in the article. It proposes a methodology for assessing
the in-house software's effectiveness for CCF of DTS, which would help determine how well the developed software
meets the basic requirements for CCF functioning.
KEYWORDS
Criterion, efficiency, in-house software, functionality, methodology, statistical modeling.
INTRODUCTION
One of the important tasks that arises at the stages of
system and system-technical design of a DTS
communication network is assessing the effectiveness
of the CCF of DTS software. It is necessary to evaluate
the effectiveness of that part of the software that
directly affects the time characteristics of the
functioning of control computing facilities and the
efficiency of the data transmission system as a whole.
This part of the in-house software is a complex of
programs that implement the functions assigned to
the control computing facilities. The in-house software
includes a complex of programs for functional tasks,
functional control, and an operating system
(dispatcher programs).
MAIN PART
To evaluate the effectiveness of the in-house software
of the CCF of DTS, a criterion must be selected that
would allow establishing the degree of compliance of
Research Article
EFFICIENCY EVALUATION OF IN-HOUSE SOFTWARE FOR CONTROL
COMPUTING FACILITIES OF DATA TRANSMISSION SYSTEMS
Submission Date:
June 11, 2024,
Accepted Date:
June 16, 2024,
Published Date:
June 21, 2024
Crossref doi:
https://doi.org/10.37547/ajast/Volume04Issue06-10
Khudoyberdiev Rahmatillo
Doctoral student of the Department of Multimedia Technologies, Tashkent University of Information
Technologies named after Muhammad al-Khwarizmi, Tashkent, Uzbekistan
Journal
Website:
https://theusajournals.
com/index.php/ajast
Copyright:
Original
content from this work
may be used under the
terms of the creative
commons
attributes
4.0 licence.
Volume 04 Issue 06-2024
51
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
the developed or being developed software with the
basic requirements for the functioning of the CCF. As
follows from the considered features of the operation
of the CCF of DTS /1,2,3,4/, compliance of the software
with the requirements to the CCF functioning is
understood not as ensuring fulfillment of the specified
functions (which is considered to be unconditionally
fulfilled), but as ensuring servicing of the requirements
with delays not exceeding the specified or maximum
permissible ones.
The choice of criterion depends on the type of CCF and
the mathematical model /5,6/ of its functioning. If all
requests for service are equivalent, then the following
criterion can be used: the probability that not all
requests received for service will be serviced untimely
must not exceed a certain acceptable value of
S
:
max
j
j
Э P T
t
S
=
.
(1)
The given criterion reflects the most essential
requirement for the functioning of the CCF determined
by the presence of strict restrictions on the magnitude
of delays in program execution. It is appropriate that
the software ensures the functioning of the CCF with
minimal costs for executing programs
( )
z
Y
:
min
z
Z
Y
=
(2)
Considering the mathematical models of the CCF
functioning, discussed in /7, 8/, the criterion for the
effectiveness of the in-house software of the CCF with
the multi-step pattern of program execution can be the
following functional:
1
i
k
i
oi
i
j m
Э
t
=
=
.
(3)
If the following conditions are met:
max
max
max
oi
i
i
i
j
j
t
t
T
t
(4)
at general loading of the CCF of the node with
programs of functional tasks and functional control:
*
1
(
)
1
i
k
j
j
i
i
j m
t
t
=
=
+
Based on the above, it is proposed to use the average
delay in the start of execution of all subroutines as a
criterion for the effectiveness of the in-house software
of the CCF of DTS with the multi-step pattern of
program execution, while the system must be in a
stationary mode
(
1)
and the requirements for the
operability of the CCF (conditions 4) must be met.
The use of criterion (3) with conditions (4) can serve to
assess the efficiency of the operating system, clarify
the distribution of programs and subroutines by
priority levels, and select a dispatching algorithm
(computer time distribution).
A criterion for the effectiveness of the in-house
software of the CCF with a limited number of load
sources can be the following functional:
0
1
n
i
i
Э T
=
=
(5)
if the following conditions are met:
0
0
max
max
Т
T
Т T
,
(6)
where
0
max
Т
- is the maximum allowable request
waiting time;
Volume 04 Issue 06-2024
52
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
max
Т
- is the maximum allowable request processing
time.
Thus, it is proposed to use the average total request
waiting time as a criterion for the effectiveness of the
in-house software of the CCF of DTS with a limited
number of load sources, while the requirements
related to the performance of the CCF - load
concentrators (conditions 6) must be met.
The suggested criteria (3) and (5), along with their
corresponding conditions (4) and (6), have clear
physical interpretations and consider the key
characteristics of the operation of DTS computing
facilities. These criteria are advantageous due to their
relative simplicity: when the values of the quantities
involved are known, the calculations using formulas (3)
and (5) are quite straightforward.
If there is no analytical queuing model for a given CCF,
it is necessary to use the method of statistical modeling
of the CCF functioning on general-purpose digital
computers (DCs) /9/. To solve the problem of analyzing
the CCF of DTS as a queuing system (QS) using the
method of modeling of the CCF, all the basic
techniques that are general for studying other classes
of queuing systems are applicable. At the same time, a
method that requires the creation of a generalized
model of the CCF as a queuing system is preferable to
a method based on the use of a model of the command
system of the CCF. The first of these methods is
associated with significantly less computer time; in
addition, to assess the time characteristics of the CCF
and the effectiveness of its in-house software, the
accuracy of the results achieved when using the
second method is not necessary.
When using the method that requires the creation of a
generalized model of the functioning of the CCF, the
information processing is not reproduced in it, but only
the principles of the timing parameters of programs
and subroutines are considered. In this case, the time
intervals occupied by the execution of each subroutine
are simulated. A program capable of reproducing the
functioning of the CCF must include blocks that
reproduce the operation of the dispatching algorithms
discussed above, blocks that generate input flows of
requests at individual service phases, and blocks
designed to obtain statistical characteristics of the CCF
functioning.
If the operating durations of subroutines and CCF
programs are not constant, then the modeling
program must also contain blocks that form service
durations in individual phases as random variables with
the necessary distribution laws. It should be noted that
for the modeling method under consideration, the
processing of information by various programs and
subroutines is not significant; only the operating time
of the corresponding programs (subroutines) is
important. The latter is specified as part of the initial
data for modeling as a constant value or as an
exponential function with parameter (for the CCF with
a limited number of load sources).
The simulation can be performed in three steps. In the
first step, the functioning of the in-house software is
simulated separately for each dispatching algorithm.
As a result of the modeling, the statistical
characteristics of the process of executing programs in
the CCF should be determined (distribution functions
for the waiting time of requests in the queue, functions
of the stay of requests in each phase and the system as
a whole, as well as their central moments). Using the
simulation results, a dispatch algorithm is selected that
ensures that conditions (4) or (6) are met.
In the second step, the distribution of subroutines by
priority levels is clarified and modeling is performed for
several options for the distribution of subroutines by
Volume 04 Issue 06-2024
53
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
priority. At the same time, the possibility of combining
several subroutines into one priority level and
rearranging individual subroutines is considered. A
distribution option is selected that has obvious
advantages from the point of view of the physical
nature of the tasks being performed and, at the same
time, allows us to ensure that conditions (4) and (6) are
met.
In the third step, the statistical characteristics of the
functioning of the CCF are finally clarified with the
selected dispatching algorithm and the refined
distribution of programs and subroutines by priority
levels. If conditions (4) and (6) are met with a reserve,
then at this step, the possibility of increasing the
number of functional tasks for a given intensity of
requests or increasing the intensity of flow for a given
volume of functional tasks is considered. The
effectiveness of in-house software is assessed
according to criteria (3) or (5).
As an example, Fig. 1 shows the block diagram of the
algorithm for simulating the functioning of the CCF of
the data transmission system, presented in the form of
a phase-based QS with a queue at each phase, with a
constant service time at each phase and with two
incoming Poisson flows of requests. The modeling
algorithm is compiled by principles known in the field
of statistical research of complex systems and
therefore is not explained here.
Volume 04 Issue 06-2024
54
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
Fig.1
. Block diagram of the simulation algorithm
Start
Form t
m
Z+1
Initial data input λ λ
1
τ
i
K Z
*
n=1
Z=0
C
i
=0
t
sojourn
i
t
completion i
= t
begin
i
+τ
i
t
begin i
=t
m
i<K
Z<Z
*
λ
completion
i+1
i=i+1
By algorithm П=2
t
completion
i
=t
m
+τ
i
+τ
queue i
τ
queue
i
=t
c
-t
m
t
c
l+1
m=m+1
τ
Output
Processing
Shut down
Shut down
Volume 04 Issue 06-2024
55
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
The following notations are used in the diagram in Fig.
1:
- intensity of the total flow of requests;
1
- intensity of the priority flow of requests (service
with relative priority);
K - number of QS phases;
i
t
- duration of service in the
i
-th phase;
*
Z
- required number of implementations;
m
t
- the moment the request enters for service at the
next phase;
Z
- number of the current implementation;
n - number of the incoming stream (n=1,2);
Z
- queue length during the service phase;
i
C
- number of the service device in the
i
-th phase;
t
begin i
- begin time of service in the
i
-th phase;
τ
sojourn i
- the time of request sojourn in the
i
-th phase;
t
queue i
- queue time for a request before the
i
-th phase;
c
t
- takedown time from servicing the previous request;
t
completion i
- completion time of service in the
i
-th phase;
Using the algorithm shown in Fig. 1, modeling of the
functioning of the CCF was conducted with several sets
of initial data
1
,
, ,
i
t k
. As a result, histograms of the
waiting time of requests of two flows and the total
flow in the queue before each phase, the time the
requests remain in the same flows at each service
phase, and the average values, variances, and
coefficients of variation of these values were obtained.
Fig. 2
. Histogram of the time the priority flow remains in the first phase
Volume 04 Issue 06-2024
56
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
The results obtained made it possible to evaluate the
functioning of the CCF. The assessment showed that
max
oi
i
t
t
. In Fig. 2, as an example, a histogram of the
time the priority flow remains in the first service phase
is shown.
Thus, the following methodology is proposed for
assessing the effectiveness of the in-house software of
the CCF of DTS, which includes the following steps:
1. Determination of the characteristics of request flows
entering the processor (subsystem) of the CCF.
2. Selection and justification of the mathematical
model of the functioning of the CCF.
3. Solving the problem of analyzing a queuing system
that interprets the functioning of the CCF under
consideration. In this step, the duration of execution of
programs and subroutines (3) and (5) of values
t
queue i
and
0
Т
are estimated.
4. Checking the fulfillment of conditions (4) and (6). If
these conditions are met, then go to step 5. If the
conditions are not met, then the developed software
of the CCF must be reprocessed. The latter may consist
of reprogramming individual parts of the software to
reduce the duration of their execution or the use of
dispatching algorithms that ensure a reduction in the
average delay times of subroutines. After processing is
completed, the fulfillment of conditions (4) or (6) is
checked again.
5. Calculations using formulas (3) or (5).
Note that it makes sense to evaluate the effectiveness
of the in-house software of the CCF only if there are
several software implementations that differ at least
slightly from each other. However, in any case, when
developing software, it is necessary to check the
fulfillment of conditions (4) and (6).
CONCLUSIONS
Thus:
- the selection and justification of the criteria for the
effectiveness of the software for the CCF of DTS was
made for two main types of CCF - with a multi-step
pattern of program execution and a limited number of
load sources;
- using the method of statistical modeling of the CCF
functioning, the effectiveness of the in-house software
was assessed;
- a methodology was developed for assessing the
effectiveness of the in-house software of the CCF of
DTS, the use of which makes it possible to establish the
degree of compliance of the developed or being
developed software with the basic requirements for
the CCF functioning.
REFERENCES
1.
Khudoiberdiev R.F. Functions and features of
operation of control computing facilities of data
transmission networks. Muhammad Al-Khorazmiy
avlodlari Ilmiy-amaliy va akhborot-takhliliy journal.
3(21)/2022.
–
P. 90-95.
2.
Stroganov R.P. Control machines and their
application. - M.: Higher School, 1986. - 240 p.
3.
Suleimanov A.A., Khudoyberdiev R.F., Mirzaeva
M.B. On the issue of assessing the throughput
capacity of the control computing facilities of data
transmission networks. Uzbek journal Problems of
Informatics and Energy No. 5. 2021 ISSN 2010-7242.
4.
Tarakanov A.V. Features of control computers //
Bulletin of the Volga University named after V.N.
Tatishchev. Series "Informatics". Issue 18.
Volume 04 Issue 06-2024
57
American Journal Of Applied Science And Technology
(ISSN
–
2771-2745)
VOLUME
04
ISSUE
06
Pages:
50-57
OCLC
–
1121105677
Publisher:
Oscar Publishing Services
Servi
5.
Mirzaeva M.B., Suleymanov A.A. On the analysis of
the functioning of computing facilities in
automated communication n
etworks. “Al
-Fargoniy
avlodlari” electronic journal. 2
-son 1(6)2024.
–
P.
274-280.
6.
Parsiev S.S. Development of mathematical models
of a telecommunications network with priority
service. Higher school, Scientific and practical
journal, Publishing House “Infinity”, UFA, 2017. –
P.
84-86.
7.
Khudoyberdiev R.F. Analytical models of operation
of control computing facilities of data transmission
systems
with
one
processor.
Raqamli
transformatsiya va sun'iy intellekt ilmiy jurnali.
2023, 3(2023).
–
P. 98-102.
8.
Khudoyberdiev R.F., Sh.A.Sadullaeva. Analytical
models of program execution processes in
multiprocessor control computing facilities of data
transmission systems. Muhammad Al-Khorazmiy
avlodlari Ilmiy-amaliy va akhborot-tahliliy journal.
2(24)/2023.
–
P. 50-53.
9.
Kagan B.M. Electronic computers and systems. -
M.: Energo-atomizdat, 1985.
–
552 p.
