RIPE NCC Charging
Scheme 2005
|
 |
Date: 29 September 2004
Document ID: ripe-330
Updates: ripe-293.
For 2005, the changes made for the Charging Scheme 2004 are refined,
while the overall principles remain the same. Several modifications have
been made to better match services rendered and to align them with membership
developments. Furthermore based on feedback received from members over
the last year, the Billing Score Algorithm for defining billing categories
is improved.
In the Charging Scheme 2005, billing categories are based on Internet
resources allocated or assigned over time. These resources consist of
IPv4 (PA), IPv6 (PA), IPv4 (PI) and AS Numbers.
The changes in the Charging Scheme 2005 are the following:
- The average service fee for existing members decreases by
10%
- The sign-up fee for new members decreases by 20%
- New members will start in the ‘Extra Small’ billing category
- The ‘Takeover fee’ is renamed ‘Administration fee’.
The administration fee is charged when resources are transferred from
one member to another or when members request the RIPE NCC to change
their RegID
- A revised algorithm to determine the billing category for
members
- IPv4 PI allocations are taken into account in the Billing
Score Algorithm
The service fees for 2005 are fixed annual charges for the RIPE NCC
membership and are based on the billing category of a member. For the
2005 service fees, and for a comparison with the service fees since 2001,
see the following table:
|
Annual Service Fees
|
2001
|
2002
|
2003
|
2004
|
2005
|
|
Extra Small
|
-
|
-
|
-
|
2,000
|
1,750
|
|
Small
|
2,100
|
1,800
|
2,750
|
2,500
|
2,250
|
|
Medium
|
2,950
|
2,500
|
3,750
|
3,500
|
3,150
|
|
Large
|
3,900
|
3,400
|
5,250
|
5,000
|
4,750
|
|
Extra Large
|
-
|
-
|
-
|
6,750
|
6,500
|
|
Sign-up fee
|
2,100
|
2,100
|
2,500
|
2,500
|
2,000
|
|
Administration fee
|
-
|
-
|
-
|
1,250
|
1,250
|
Membership Growth Projections
To improve the accuracy of growth forecasts the RIPE NCC conducted a
membership growth survey to gather input from the members regarding membership
developments in their country. Following the membership growth survey
and analysing the growth for the current year, a net growth rate of approximately
9% is expected over the year 2004. For 2005, taking into account expected
member closures, a net growth rate of 11% is expected.
The membership figures for 2001 - 2003, the projections for the remainder
of 2004 and the estimation for the budget for 2005 are as follows:
|
Number of LIRs
|
2001
|
2002
|
2003
|
Projection 2004
|
Budget 2005
|
|
Extra Small
|
-
|
-
|
-
|
729
|
1,156
|
|
Small
|
2,536
|
2,503
|
2,664
|
2,247
|
2,082
|
|
Medium
|
441
|
614
|
659
|
647
|
757
|
|
Large
|
145
|
152
|
165
|
130
|
151
|
|
Extra Large
|
-
|
-
|
-
|
33
|
38
|
|
Total membership
|
3,122
|
3,269
|
3,488
|
3,786
|
4,184
|
|
Net Growth %
|
22 %
|
5 %
|
7 %
|
9 %
|
11 %
|
Note: New members that are expected after the billing scores for 2005
have been determined are included in the Extra Small billing category.
Each member is allocated a score according to the Billing Score Algorithm
(see Appendix I). Starting from the lowest score, all members are ranked
in order depending on their score. Members with the same score get identical
rankings. The billing categories are defined using the following cumulative
boundaries:
- Up to 20% of the members will make up the Extra Small billing
category
- Up to 75% of the members will make up the Extra Small and
Small billing categories
- Up to 95% of the members will make up the Medium billing
category and all smaller billing categories
- Up to 99% of the members will make up the Large billing
category and all smaller billing categories
- The remaining members will make up the Extra Large billing
category
Note: These percentages may deviate slightly. If a set of members with
the same score fall across the boundary between billing categories, these
members will be part of the next higher billing category.
Percentage of Total Members per Billing Category
|
Billing Category
|
2001
|
2002
|
2003
|
July 2004
|
Cumulative % 2005
|
|
Extra Small
|
-
|
-
|
-
|
20 %
|
20 %
|
|
Small
|
81 %
|
77 %
|
78 %
|
58 %
|
75 %
|
|
Medium
|
14 %
|
19 %
|
18 %
|
17 %
|
95 %
|
|
Large
|
5 %
|
4 %
|
4 %
|
4 %
|
99 %
|
|
Extra large
|
-
|
-
|
-
|
1 %
|
100 %
|
The billing scores for members will be determined on 30 September 2004.
The billing category and score range will be published and every member
will be notified of their billing score and billing category by e-mail.
Change Matrix - Expected Changes of Members Between the Billing Categories
The Change Matrix indicates the percentage of members currently in a
certain billing category that are expected to move to a different billing
category for 2005.
For example: the matrix shows that, for 2005, 53% of the members currently
in the Extra Small billing category are expected to remain in that category,
41% of the members currently in the Extra Small billing category are
expected to move to the Small category and 6% of the members currently
in the Extra Small billing category are expected to move to the Medium
category.
|
|
Extra Small
|
Small
|
Medium
|
Large
|
Extra Large
|
|
Extra Small
|
53%
|
41%
|
6%
|
-
|
-
|
|
Small
|
29%
|
62%
|
9%
|
0%
|
-
|
|
Medium
|
1%
|
23%
|
68%
|
8%
|
0%
|
|
Large
|
-
|
1%
|
27%
|
64%
|
8%
|
|
Extra Large
|
-
|
-
|
-
|
29%
|
71%
|
Appendix I: Billing Score Algorithm
To determine the billing category for a member, a score is calculated
on the basis of the resources allocated or assigned at the request of
the member. The scoring system is based on Internet resource allocations
or assignments made to the member over time taking into account IPv4
(PA and PI), IPv6 and AS Numbers.
For the purpose of this scoring algorithm, an IPv4 /20 is equivalent
to one IPv6 /32, to one AS Number or to one PI IPv4 /24. The following
table shows how scoring units are determined based on resource usage,
where
the relative weights of the various resource
types
are established based on work load caused by the respective resource
requests. To establish
scoring units based on larger or smaller resource usage, the same principles
apply.
|
Prefix IPv4
|
Prefix IPv6
|
AS Numbers
2004
|
Prefix PI IPv4 2004
|
Scoring Unit
|
|
IPv4 / 21 ≙
|
IPv6 / 33 ≙
|
|
PI IPv4 / 25 ≙
|
0.5
|
|
IPv4 / 20 ≙
|
IPv6 / 32 ≙
|
1 ASN ≙
|
PI IPv4 / 24 ≙
|
1
|
|
IPv4 / 19 ≙
|
IPv6 / 31 ≙
|
2 ASN ≙
|
PI IPv4 / 23 ≙
|
2
|
|
IPv4 / 18 ≙
|
IPv6 / 30 ≙
|
4 ASN ≙
|
PI IPv4 / 22 ≙
|
4
|
Note: For AS Numbers and PI IPv4 allocations,
only the allocations from the past 12 months dating back from the 30
September 2004 are taken into account as these resources are allocated
or assigned on behalf of third parties.
Using this matching system, the following algorithm is run to determine
the total score per member:
ai = Scoring unit
ti = Time function of allocation/assignment i ( year of allocation – 1992
)
N = Number of allocations/assignments
The total score per member is the sum of all allocation scores for that
member with a time factor applied to give more weight to recent allocations.
Thus, the relative weight of a given allocation decreases over time.
Note: PI stands for Provider Independent and PA stands for Provider Aggregatable. |