Capacity planning method and apparatus
The present disclosure relates to capacity planning methods and apparatus. One example method includes matching a distribution model based on a quantity of service packets in each transmission time interval within specified duration to obtain a matched first distribution model, matching a distribution model based on a length of the service packets to obtain a second distribution model, and performing bandwidth control based on the first distribution model, the second distribution model, a distribution parameter of the first distribution model, and a distribution parameter of the second distribution model.
1. A capacity planning method, comprising:
receiving, by a capacity planning apparatus, a distribution parameter value of a first distribution model and a distribution parameter value of a second distribution model from a service measurement apparatus, wherein the first distribution model and the second distribution model are respectively a distribution model matching a quantity of service packets in each transmission time interval within specified duration and a distribution model matching a length of the service packets that are obtained by the service measurement apparatus; and
performing, by the capacity planning apparatus, bandwidth control based on the first distribution model, the second distribution model, the distribution parameter value of the first distribution model, and the distribution parameter value of the second distribution model.
2. The method according to claim 1 , wherein the performing, by the capacity planning apparatus, bandwidth control based on the first distribution model, the second distribution model, the distribution parameter value of the first distribution model, and the distribution parameter value of the second distribution model comprises:
determining, by the capacity planning apparatus, a user experienced rate distribution model based on the first distribution model, the second distribution model, a downlink transmission rate of a base station, and the transmission time interval; and
performing, by the capacity planning apparatus, the bandwidth control based on the user experienced rate distribution model, the distribution parameter value of the first distribution model, the distribution parameter value of the second distribution model, and a quality of service requirement parameter value.
3. The method according to claim 2 , wherein the quality of service requirement parameter value is preset bandwidth utilization; and
wherein the performing, by the capacity planning apparatus, the bandwidth control based on the user experienced rate distribution model, the distribution parameter value of the first distribution model, the distribution parameter value of the second distribution model, and a quality of service requirement parameter value comprises:
determining, by the capacity planning apparatus, a busy-hour average user experienced rate value based on the user experienced rate distribution model, the distribution parameter value of the first distribution model, the distribution parameter value of the second distribution model, and the preset bandwidth utilization; and
performing, by the capacity planning apparatus, the bandwidth control based on the busy-hour average user experienced rate value.
4. The method according to claim 3 , wherein the first distribution model is a Zeta distribution model, a distribution model parameter of the first distribution model includes s, the second distribution model is a Pareto distribution model, and distribution model parameters of the second distribution model include m and α;
the user experienced rate distribution model is:
Pr
(
R
U
>
x
)
≈
{
(
1
-
p
0
)
ζ
(
s
-
1
)
m
a
(
R
τ
-
E
[
S
]
)
ζ
(
s
)
(
α
-
1
)
(
τ
x
-
ζ
(
s
-
1
)
ζ
(
s
)
m
+
m
)
-
α
+
1
,
x
<
R
,
0
,
x
≥
R
wherein Pr( ) is the user experienced rate distribution model, R U is a user experienced rate, the user experienced rate at a moment t is
R
U
(
t
)
=
min
{
R
,
Q
(
t
)
τ
}
,
R is the downlink transmission rate of the base station, Q(t) is a length of a queue on the base station at the moment t, the queue is used to buffer the service packets, T is the transmission time interval, ζ( ) is a Riemann function, E[S] is an expected value of a quantity of bits that arrive within one transmission time interval, and p 0 is a probability that a quantity of service packets that arrive is zero; and
the capacity planning apparatus determines the busy-hour average user experienced rate value according to the following formula:
R
¯
busy
=
(
R
τ
-
ζ
(
s
-
1
)
ζ
(
s
)
m
+
m
)
2
-
α
-
(
m
-
ζ
(
s
-
1
)
ζ
(
s
)
m
)
2
-
α
t
(
2
-
α
)
(
ε
R
τ
-
ζ
(
s
-
1
)
ζ
(
s
)
m
+
m
)
-
α
+
1
,
wherein R busy is the busy-hour average user experienced rate value, and ε is the preset bandwidth utilization, and 0≤ε≤1.
5. The method according to claim 3 , wherein the performing, by the capacity planning apparatus, the bandwidth control based on the busy-hour average user experienced rate value comprises:
increasing bandwidth if a difference between the busy-hour average user experienced rate value and the downlink transmission rate of the base station is greater than a first difference threshold; or
decreasing bandwidth if a difference between the busy-hour average user experienced rate value and the downlink transmission rate of the base station is less than a second difference threshold.
6. The method according to claim 2 , wherein the quality of service requirement parameter value is a preset user experienced rate satisfaction degree; and
wherein the performing, by the capacity planning apparatus, the bandwidth control based on the user experienced rate distribution model, the distribution parameter value of the first distribution model, the distribution parameter value of the second distribution model, and a quality of service requirement parameter value comprises:
determining, by the capacity planning apparatus, a lower limit value of the user experienced rate based on the user experienced rate distribution model, the distribution parameter value of the first distribution model, the distribution parameter value of the second distribution model, and the preset user experienced rate satisfaction degree; and
performing, by the capacity planning apparatus, the bandwidth control based on the lower limit value of the user experienced rate.
7. The method according to claim 6 , wherein the first distribution model is a Zeta distribution model, a distribution model parameter of the first distribution model includes s, the second distribution model is a Pareto distribution model, and distribution model parameters of the second distribution model include m and a;
the user experienced rate distribution model is:
Pr
(
R
U
>
x
)
≈
{
(
1
-
p
0
)
ζ
(
s
-
1
)
m
a
(
R
τ
-
E
[
S
]
)
ζ
(
s
)
(
α
-
1
)
(
τ
x
-
ζ
(
s
-
1
)
ζ
(
s
)
m
+
m
)
-
α
+
1
,
x
<
R
,
0
,
x
≥
R
wherein Pr( ) is the user experienced rate distribution model, R U is the user experienced rate, the user experienced rate at a moment t is
R
U
(
t
)
=
min
{
R
,
Q
(
t
)
τ
}
,
R is the downlink transmission rate of the base station, Q(t) is a length of a queue on the base station at the moment t, the queue is used to buffer the service packets, T is the transmission time interval, ζ( ) is a Riemann function, E[S] is an expected value of a quantity of bits that arrive within one transmission time interval, and p 0 is a probability that a quantity of service packets that arrive is zero; and
the capacity planning apparatus determines the lower limit value of the user experienced rate according to the following formula:
(
1
-
p
0
)
ζ
(
s
-
1
)
m
α
(
R
τ
-
E
[
S
]
)
ζ
(
s
)
(
α
-
1
)
(
τ
R
min
-
ζ
(
s
-
1
)
ζ
(
s
)
m
+
m
)
-
α
+
1
=
η
,
wherein R min is the lower limit value of the user experienced rate, and η is the preset user experienced rate satisfaction degree.
8. The method according to claim 6 , wherein the performing, by the capacity planning apparatus, the bandwidth control based on the lower limit value of the user experienced rate comprises:
increasing bandwidth if a difference between the lower limit value of the user experienced rate and the downlink transmission rate of the base station is greater than a third difference threshold; or
decreasing bandwidth if a difference between a busy-hour average user experienced rate value and the downlink transmission rate of the base station is less than a fourth difference threshold.
9. The method according to claim 1 , wherein the method further comprises:
receiving, by the capacity planning apparatus, identification information of the first distribution model and identification information of the second distribution model from the service measurement apparatus, wherein the identification information of the first distribution model is used to identify the first distribution model, and wherein the identification information of the second distribution model is used to identify the second distribution model.
10. The method according to claim 1 , wherein the method further comprises:
receiving, by the capacity planning apparatus, identification information of a service arrival model from the service measurement apparatus, wherein the identification information of the service arrival model is used to identify a service arrival model corresponding to the first distribution model and the second distribution model; and
determining, by the capacity planning apparatus, the first distribution model and the second distribution model based on the identification information of the service arrival model.
11. A capacity planning apparatus, comprising:
a transceiver, the transceiver configured to receive a distribution parameter value of a first distribution model and a distribution parameter value of a second distribution model from a service measurement apparatus, wherein the first distribution model and the second distribution model are respectively a distribution model matching a quantity of service packets in each transmission time interval within specified duration and a distribution model matching a length of the service packets that are obtained by the service measurement apparatus;
at least one processor; and
one or more memories coupled to the at least one processor and storing programming instructions for execution by the at least one processor to perform bandwidth control based on the first distribution model, the second distribution model, the distribution parameter value of the first distribution model, and the distribution parameter value of the second distribution model.
12. The apparatus according to claim 11 , wherein the programming instructions are for execution by the at least one processor to determine a user experienced rate distribution model based on the first distribution model, the second distribution model, a downlink transmission rate of a base station, and the transmission time interval; and perform the bandwidth control based on the user experienced rate distribution model, the distribution parameter value of the first distribution model, the distribution parameter value of the second distribution model, and a quality of service requirement parameter value.