ipv4 address lifetime presented by paul wilson, apnic research activity conducted by geoff huston...

IPv4 Address Lifetime

Presented by Paul Wilson, APNIC

Research activityconducted by Geoff Huston

and supported by APNIC

IPv4 Address Lifetime Expectancy

• IETF activity within the Routing and Addressing (ROAD) group in the early 1990’s– The objective was to understand the rate of

allocation of IPv4 addresses and predict the date of eventual exhaustion of the unallocated pool

– At the time the prediction was that the pool of IPv4 addresses would be exhausted around 2008-2011

• This is a re-visiting of this activity considering latest data– IETF, IANA and RIR delegations– Also, ISP announcements to the BGP routing table

Modeling the Process

1. IETF definition of IPv4– Source: IETF standards (RFCs)

• Delegation of address space for IANA administration

2. IANA allocations to RIRs– Source: IANA IPv4 Address Registry

• Allocation of /8 blocks to RIRs and others

3. RIR allocations to ISPs– Source: RIR Stats files

• Allocation of blocks to LIRs

4. ISP announcements – Source: BGP routing table

• Amount of address space advertised

Modeling the Process

IANA

RIR

Allo

catio

n

RIR

ISPA

lloca

tion

ISP

BGP

Ann

ounc

emen

t

IETF

IANA

Del

egat

ion

1. IETF Delegations

IPv4 Address Space

• Defined by the IETF– 32 bits providing 4B addresses

• The IETF has defined space for global unicast (administered by the IANA) and for other purposes

• IANA allocates space to the RIRs for further allocation and assignment

IETF Reserved, 20.1, 8%

Multicast, 16, 6%

Unicast, 219.9, 86%

IPv4 Address Space

Breakdown of IPv4 address Space by /8 block equivalents

2. IANA Allocations

IANA Allocations

• IANA allocates address space to RIRs

• The IANA IPv4 address registry records the date of each /8 allocation undertaken by the IANA

• This data has some inconsistencies– Changing IANA administration and

practices over many years

• However data is stable enough to allow some form of projection

IETF Reserved, 20.1, 8%

Multicast, 16, 6%

IANA Pool, 89, 35%

IANA Allocated, 130.9, 51%

IANA Allocations - Current

IANA Allocations - HistoricalIANA Allocated IPv4 /8 Address Blocks

0

20

40

60

80

100

120

140

Jan-91 Jan-93 Jan-95 Jan-97 Jan-99 Jan-01 Jan-03

IANA Allocations - Projection

IANA Allocated IPv4 /8 Address Blocks

100

120

140

160

180

200

220

Jan-96 Jan-98 Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-12 Jan-14 Jan-16 Jan-18 Jan-20

IANA Allocations - Projection

• Projected date of IANA address pool exhaustion: 2020

• This projection is very uncertain because of:– Sensitivity of allocation rate to prevailing

RIR assignment policies– Sensitivity to any significant uptake up of

new applications that require end-to-end IPv4 addressing vs use of NATs

3. RIR Allocations

RIR Allocations

• RIRs allocate address space to LIRs (ISPs)

• RIR stats files records the date of each allocation to an LIR, together with the allocation details

RIR Allocations - Current

IETF Reserved, 20.1, 8%

Multicast, 16, 6%

IANA Pool, 89, 35%

RIR Pool, 14, 5%

Assigned, 116.9, 46%Allocated

RIR Allocations - Current

Address Allocation Status - by /8

0

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000

Reserved

IANA

Unallocated

Allocated

RIR Allocations - Historical

RIR Assigned IPv4 /8 Address Blocks

0

20

40

60

80

100

120

Jan-83 Jan-85 Jan-87 Jan-89 Jan-91 Jan-93 Jan-95 Jan-97 Jan-99 Jan-01 Jan-03

RIR Allocations - Projection

RIR Assigned IPv4 /8 Address Blocks - Projection

80

100

120

140

160

180

200

220

Jan-96 Jan-98 Jan-00 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-12 Jan-14 Jan-16 Jan-18 Jan-20 Jan-22 Jan-24 Jan-26

RIR Allocations - Projection

• Projected date of RIR address pool exhaustion: 2027

• The projection has the same levels of uncertainty as noted for the IANA projections:– RIR management policies– Technological developments

4. BGP Routing Table

BGP Routing Table

• The BGP routing table spans a set of advertised addresses– Representing addresses in use by ISPs

• A similar analysis of usage and projection can be undertaken on this data

• Assumption: BGP routing table represents actual IP address usage– Therefore it “drives” the other trends

BGP Routing Table - Current

IETF Reserved, 20.1, 8%

Multicast, 16, 6%

IANA Pool, 89, 35%

RIR Pool, 14, 5%

Assigned, 42.4, 17%

Advertised, 74.5, 29%

BGP Announcements - Current

Address Allocation Status - by /8

0

2000000

4000000

6000000

8000000

10000000

12000000

14000000

16000000

Reserved

IANA

Unallocated

Unadvertised

Advertised

BGP Announcements - HistoricalBGP Table - Address Span

0

50

100

150

200

Nov

/99

Jan/

00

Mar

/00

May

/00

Jul/0

0

Sep

/00

Nov

/00

Jan/

01

Mar

/01

May

/01

Jul/0

1

Sep

/01

Nov

/01

Jan/

02

Mar

/02

May

/02

Jul/0

2

Sep

/02

Nov

/02

Jan/

03

Mar

/03

May

/03

BGP Announcements - ProjectionBGP Announced Address Space - Projection

0

50

100

150

200

Jan/

00

Jan/

01

Jan/

02

Jan/

03

Jan/

04

Jan/

05

Jan/

06

Jan/

07

Jan/

08

Jan/

09

Jan/

10

Jan/

11

Jan/

12

Jan/

13

Jan/

14

Jan/

15

Jan/

16

Jan/

17

Jan/

18

Jan/

19

Jan/

20

Jan/

21

Jan/

22

Jan/

23

Jan/

24

Jan/

25

Jan/

26

Jan/

27

Jan/

28

BGP Announcements - Projection

• Projected date of address pool exhaustion according to BGP: 2027

• This projection uses a 3 year data baseline to obtain the projection– This is much shorter baseline than the

IANA and RIR projections– There are, again, considerable

uncertainties associated with this projection

BGP Announcements - Projection

• Comments received about this projection have prompted a more detailed analysis of the BGP data

• It appears that there is a different view that can be formed from the data

• Firstly, here’s the raw data – hourly measurements over 3 years…

Another look at that BGP data…

Another look at that BGP data…

• The most obvious noise comes from flaps in /8 advertisements.

• The first step was to remove this noise by recalculating the address data using a fixed number of /8 advertisements

• The value of 19 was used to select one of the ‘tracks’ in the data

Another look at that BGP data…Filter to 19 /8s

980000000

1030000000

1080000000

1130000000

1180000000

1230000000

Nov-99 Mar-00 Jul-00 Nov-00 Mar-01 Jul-01 Nov-01 Mar-02 Jul-02 Nov-02 Mar-03 Jul-03

Another look at that BGP data…

• This is still noisy, but there is no systematic method of raw data grooming that can efficiently reduce this noise

• Now use gradient smoothing, limiting the absolute values of the first order differential of the data (gradient limiting) to smooth the data

Another look at that BGP data…Gradient Filtered Data

980000000

1030000000

1080000000

1130000000

1180000000

1230000000

Nov-99 Mar-00 Jul-00 Nov-00 Mar-01 Jul-01 Nov-01 Mar-02 Jul-02 Nov-02 Mar-03 Jul-03

Another look at that BGP data…

• Now, further smoothing is needed to reduce the data set to allow projection models to be generated

• The technique used is a sliding window average, with a window of 1501 entries

Another look at that BGP data…Smoothed Average

980000000

1030000000

1080000000

1130000000

1180000000

1230000000

Nov-99 Mar-00 Jul-00 Nov-00 Mar-01 Jul-01 Nov-01 Mar-02 Jul-02 Nov-02 Mar-03 Jul-03

Another look at that BGP data…Linear Squares Best Fit

58

60

62

64

66

68

70

72

74

Nov-99 Feb-00 May-00 Aug-00 Nov-00 Feb-01 May-01 Aug-01 Nov-01 Feb-02 May-02 Aug-02 Nov-02 Feb-03 May-03 Aug-03

Another look at that BGP data…

• First order differential of total BGP announcement– Until 2000, exponential (accelerating)

growth– Since 2000, oscillating differential and

overall deceleration– Last 6 months, differential approaching 0

(i.e. no growth)

• Linear fit seems most appropriate for this data

Another look at that BGP data…daily rate of change in address growth per month

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

Dec-99 Mar-00 Jun-00 Sep-00 Dec-00 Mar-01 Jun-01 Sep-01 Dec-01 Mar-02 Jun-02 Sep-02 Dec-02 Mar-03 Jun-03

Combining the Data

Combining the DataIPv4 Address Space

0

20

40

60

80

100

120

140

Jan-83 Jan-85 Jan-87 Jan-89 Jan-91 Jan-93 Jan-95 Jan-97 Jan-99 Jan-01 Jan-03

IANA

RIR

BGP

Recent DataIPv4 Address Space

0

20

40

60

80

100

120

140

Dec-99 Mar-00 Jun-00 Sep-00 Dec-00 Mar-01 Jun-01 Sep-01 Dec-01 Mar-02 Jun-02 Sep-02 Dec-02 Mar-03 Jun-03

IANA

RIR

BGP

Holding Pools

Holding Pools:RIR & Unannounced Space

Size of Holding Areas (/8)

0

5

10

15

20

25

30

35

40

45

50

Nov-99 Feb-00 May-00 Aug-00 Nov-00 Feb-01 May-01 Aug-01 Nov-01 Feb-02 May-02 Aug-02 Nov-02 Feb-03 May-03

UnAnn

RIR

Modeling the Process

• Assume that the RIR efficiency in allocation slowly declines, so that the amount of RIR-held space increases over time

• Assume that the Unannounced space shrinks at the same rate as shown over the past 3 years

• Assume an exponential best fit model to the announced address space projections and base RIR and IANA pools from the announced address space projections, using the above 2 assumptions

IPv4 Model

0

50

100

150

200

Jan-00 Jan-05 Jan-10 Jan-15 Jan-20 Jan-25

IANA

RIR

BGP

IANA-P

RIR-P

BGP-P

RIR

LIR

Modeling the Process - Exponential

Projections

IANA Pool Exhaustion 2022

RIR Pool Exhaustion 2024

Modeling the Process - LinearIPv4 Model

0

50

100

150

200

Jan-00 Jan-05 Jan-10 Jan-15 Jan-20 Jan-25 Jan-30 Jan-35 Jan-40 Jan-45

IANA

RIR

BGP

IANA-P

RIR-P

BGP-P

RIR

LIR

Projections

IANA Pool Exhaustion 2030

RIR Pool Exhaustion 2037

Observations

• Extrapolation of current allocation practices and BGP-based demand model

• Derived from 2000-2003 data• Considering

– IANA/RIR unallocated pool– Total address space including allocated but

unannounced

• Assuming exponential growth– Address space lasts until 2025, or up to 2029

• Assuming linear growth– Address space lasts until 2037 - 2047

Issues

• This is just a model - reality will be different!• Will the BGP routing table continue to reflect

allocation rates? • Is the model of the unadvertised pools and

RIR holding pools appropriate?• Externalities…

– What are the underlying growth drivers (applications) and how are these best modeled?

– What forms of disruptive events would alter this model, and to what extent?

Questions?

gih@telstra.net

http://www.potaroo.net