ing upon the user’s budget they can choose flavours
too.
For this evaluation, a simulator had been cre-
ated named flavour selector where users can input the
range of SAU, BW, and total cost depending upon
their requirement our simulator will propose the ap-
proximate flavor. For example, if the SAU range
given by the client is 100-120 our selector suggests
flavor A from Table 5. Another case is a range be-
tween 130-160 then our selector suggests flavour B
also whenever the user inputs range value then it will
suggest customizing flavor, i.e. flavour E. Similar to
BW, if the client provides the range of bandwidth be-
tween 18-20 Mbps the selector will propose flavour B
from Table 6. Another interesting case is using to-
tal cost. For the total cost, we added the value of
Resources Cost (RC
ca
) and License Cost (LC
ca
) as
shown in column (9
th
) in Table 5 and Table 6. So,
depending upon the customer’s budget simulator pro-
posed flavour between SAU and BW. For example:
if the client’s budget range is from 800-900ke, the
simulator provides flavour D from Table 5. Further-
more, if the range is from 600-700ke then it could be
from flavour C from SAU or BW table. So, to avoid
the confusion of choosing between SAU and BW the
simulator asks the client preference between SAU and
BW. Thus, depending upon the client’s needs simula-
tor provides the result either from BW, SAU or cost.
Ë Resources Unknown Users (RUU): These are
the users who don’t have the estimated knowledge
about resources requirements (SAU, BW) for their
system. So for these types of users, a simulator was
created to provide the users with several choices. At
first, users need to provide their range after which the
simulator will propose the least SAU value from the
Table 5. If the user is not satisfied with that proposed
then they can processed further simulator will pro-
pose from BW Table 6, least range from BW. If this
range is also not satisfactory to the client requirement
then the simulator proposed the mean value from the
SAU flavour table, if this also fails to meet the user’s
needs the simulator proposed the mean value from
the BW flavour table. After this, the simulator pro-
posed the highest value of SAU and BW from SAU
and BW flavour table respectively. So the simulator
proposed from least to maximum flavour value from
tables based on SAU, BW. Thus, our aim here is to
provide as many options as possible to the user. Ad-
ditionally, the offer can be made concerning the total
cost as performed in RKU.
3.2.3 Scenario 3: Node Analysis in Real Scenario
For this analysis, we used the two techno-economic
friendly models to estimate the LC, RC. They are
Table 5: SAU Flavour Table for Scenario 2.
Flavour SAU LC
(Ke)
vStorage
(TB)
vRAM vCPU Redundancy Resources
cost
(Ke)
Total
cost
(Ke)
A 100 250 4 2 GB 2 1 100 350
B 150 350 4 3 GB 3 2 150 500
C 200 450 6 4 GB 4 2 215 665
D 250 550 7 5 GB 5 2 275 825
E 350 650 8 5 GB 5 2 325 975
F Customize your needs
Table 6: BW Flavour Table for Scenario 2.
Flavour BW
(Mbps)
LC
(Ke)
vStorage
(TB)
vRAM vCPU Redundancy Resources
cost (K
e)
Total
cost
(Ke)
A 15 200 2 2 GB 2 1 170 370
B 20 250 3 2 GB 3 2 200 450
C 30 355 4 3 GB 4 2 285 640
D 50 435 4 3 GB 4 2 338 773
E 65 549 5 4 GB 4 2 420 969
F Customize your needs
capacity and consumption. To adapt these models
from a business point of view we have considered
some thresholds, constraints related to license and re-
sources like license threshold, resources threshold, li-
cense factors, etc.
• Capacity: The capacity analysis is similar to pre-
paid service where a certain amount of cost is
paid upfront to a certain capacity (license refer-
ence for our research) of VNF. When it surpasses
the threshold extra costs will be incurred. The
threshold can be a license or resource or both.
In this research both LC and RC were estimated
using unit cost and license reference using Equa-
tions (9) and (11). In this mode, once the capacity
is increased it cannot be reversed even if the con-
sumption (SAU/BW) is lower than the threshold.
• Consumption: Clients will pay for the resources
they had consumed or will consume during a cer-
tain time. As a consequence, there is no contrac-
tual threshold limiting the user’s ability to con-
sume resources (SAU, BW). It was calculated us-
ing Equations (11) and (12).
• LC threshold: This is the threshold for calculating
license cost. LC threshold was implemented for
both LC
ca
and LC
cp
. In this research, wherever li-
cense is being used it includes both (LC
ca
, LC
cp
).
It is defined at the time of negotiation of the con-
tract based on estimated needs. If the uses exceed
the threshold, the cost of a license is increased by
1.5, 2, 3, . . . , n, which is known as license factor
σ. The license threshold is based on LR.
• RC threshold: It is the threshold in the resources.
Whenever a threshold is exceeded, it requires a
careful evaluation to understand whether the ex-
ceeded threshold can be covered by a single re-
source or more. If it can be covered with one re-
source then our research will use resources fac-
tor (δ)=1, and if requires more than one it will be
from 1,5, 2, 3, 4, 5, . . . , n. This threshold is also
dependent upon the LR. Alike the LC threshold, it
CLOSER 2022 - 12th International Conference on Cloud Computing and Services Science
250