Remote Peering: More Peering without Internet Flattening Ignacio - - PowerPoint PPT Presentation

remote peering
SMART_READER_LITE
LIVE PREVIEW

Remote Peering: More Peering without Internet Flattening Ignacio - - PowerPoint PPT Presentation

Remote Peering: More Peering without Internet Flattening Ignacio Castro Juan Camilo Cardona * Sergey Gorinsky Pierre Francois IMDEA Networks Institute Open University of Catalonia * Carlos III University of Madrid Madrid,


slide-1
SLIDE 1

Remote Peering:

More Peering without Internet Flattening

Ignacio Castro†‡ Juan Camilo Cardona†* Sergey Gorinsky† Pierre Francois† WIE, La Jolla, 10 December 2014

† IMDEA Networks Institute Madrid, Spain ‡ Open University of Catalonia Barcelona, Spain * Carlos III University of Madrid Madrid, Spain

slide-2
SLIDE 2

“Everything depends on the color of the crystal that one looks through”

slide-3
SLIDE 3

“Everything depends on the color of the crystal that one looks through”

Looking at the Internet through layer-3 glasses

slide-4
SLIDE 4

Transit

4

Internet Layer-3 model

Peering

Customer Provider Peer Peer

Network Autonomous System (AS) Interconnections between networks

Modeling of Internet Economics

slide-5
SLIDE 5

Is there anything else?

slide-6
SLIDE 6
  • Service components

– Layer-2 connectivity of the AS to the IXP – Peering equipment at the IXP

6

IXP Remote-peering provider Remotely peering AS IP router Layer-2 switch

Remote-Peering Providers

slide-7
SLIDE 7

AMS-IX Hong Kong

Connecting two IXPs

7

Usage of Remote Peering

Dutch AS Remote peering

Trial peering

Dutch AS

?

Reducing costs

  • ver short

distances

AMS-IX

Reaching a distant IXP

African AS

slide-8
SLIDE 8

Dutch AS

8

Usage of Remote Peering

Remote peering

Reaching a distant IXP

African AS

Trial peering

Dutch AS

?

Reducing costs

  • ver short

distances

AMS-IX AMS-IX Hong Kong

Connecting two IXPs

slide-9
SLIDE 9

9

AMS-IX Hong Kong

Connecting two IXPs Reaching a distant IXP

African AS

Usage of Remote Reering

Remote peering

Trial peering

Dutch AS

?

Reducing costs

  • ver short

distances

Dutch AS AMS-IX

slide-10
SLIDE 10

10

AMS-IX Hong Kong

Reducing costs

  • ver short

distances Connecting two IXPs Reaching a distant IXP

African AS

Usage of Remote Peering

Trial peering

Dutch AS

?

Dutch AS AMS-IX Remote peering Peering

slide-11
SLIDE 11

11

Our Contributions

  • Measurement-based studies

– Spread of remote peering – Impact of remote peering on Internet traffic

  • Modeling of economic viability

– Remote peering vs. transit and direct peering

slide-12
SLIDE 12

12

Estimating the Spread

  • Studied questions

– How many IXPs have remote peering? – How many IXP members are remote peers?

  • Approach

– Conservative estimate – RTT (Round-Trip Time) as a metric of peer remoteness – 22 IXPs with colocated Looking Glass servers

slide-13
SLIDE 13
  • IP address from PCH, PeeringDB, and IXPs websites
  • Ping reply within one IP hop if its TTL = maximum TTL
  • 4 months and 6 filters to get minimum RTT reliably
  • If RTT > threshold, classify the peer as remote

– Empirical threshold of 10 ms

13

Classification of Peers as Remote

IXP Looking Glass Peering AS IP address ping request ping reply IP router

slide-14
SLIDE 14

Remote-peering provider IXP IXP Looking Glass ping request ping reply

14

IP router Layer-2 switch Remotely peering AS IP address

Validation

  • Public IXP information on remote peers
  • Ground truth from TorIX

– RTT measurements – Remotely peering ASes

slide-15
SLIDE 15

15

Spread across IXPs

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ X ✓ X ✓

91% of the IXPs have remote peering

slide-16
SLIDE 16

Around 20% of AMS-IX peers are remote

16

Spread within IXPs

slide-17
SLIDE 17

17

Our Contributions

  • Measurement-based studies

– Spread of remote peering – Impact of remote peering on Internet traffic

  • Modeling of economic viability

– Remote peering vs. transit and direct peering

slide-18
SLIDE 18

Estimating the Offload Potential

18

  • Studied questions

– How can an AS benefit from remote peering? – How much traffic can the AS offload from its transit-provider links?

  • Evaluated AS

– RedIRIS, the Spanish national academic network – 1 month of NetFlow traffic data – Routing tables

slide-19
SLIDE 19

Transit-Provider Traffic of RedIRIS

19

Transit providers (2) RedIRIS Transit Traffic

  • 2 transit providers
  • 29,570 ASes contribute transit traffic
slide-20
SLIDE 20

Choice of Reached IXPs

20

Remote peering Peering Transit Traffic Transit providers (2) RedIRIS IXPs IXP members Customer cones

  • f IXP members
  • Up to 65 IXPs from Euro-IX
  • Reaching up to 12,238 ASes

− Out of 29,570 ASes with RedIRIS transit traffic

slide-21
SLIDE 21

Choice of Peers for RedIRIS

21

Remote peering Peering Transit Traffic Transit providers (2) RedIRIS IXPs

Peering policies from Peering DB

  • 1. all open, lower bound
  • 2. all open and top 10 selective,
  • 3. all open and selective,
  • 4. all policies upper bound

IXP members Customer cones

  • f IXP members
slide-22
SLIDE 22

Between 8% and 25% of reduction in transit traffic

22

How Much Traffic can RedIRIS Offload?

8% 25%

slide-23
SLIDE 23

23

Utility of Reaching an Additional IXP

Reaching only 5 IXPs realizes most of the offload potential

slide-24
SLIDE 24

24

Is the RedIRIS Case Representative?

Decreasing marginal utility of reaching an additional IXP is a general property

slide-25
SLIDE 25

Conclusions

  • Remote peering, a new common interconnection

– AS reaches and peers at IXP via a layer-2 provider

  • Potential impact on Internet traffic is substantial

– Reaching only 5 IXPs realizes most of the potential

  • Internet economic structure needs refined models

– Layer-2 entities need to be represented

25 Contact: Ignacio Castro, ignacio.decastro@imdea.org