Refining Network Intents for Self-Driving Networks Arthur Selle - - PowerPoint PPT Presentation

▶
refining network intents for self driving networks
SMART_READER_LITE
LIVE PREVIEW

Refining Network Intents for Self-Driving Networks Arthur Selle - - PowerPoint PPT Presentation

Aug 05, 2023 552 likes •875 views

Refining Network Intents for Self-Driving Networks Arthur Selle Jacobs Ricardo Jos Pfitscher, Ronaldo Alves Ferreira, Lisandro Zambenedetti Granville UFRGS UFMS Budapest, Hungary August 24, 2018 Self-Driving Networks High-level

slide-1
SLIDE 1

Refining Network Intents for Self-Driving Networks

Arthur Selle Jacobs¹

Ricardo José Pfitscher¹, Ronaldo Alves Ferreira², Lisandro Zambenedetti Granville¹ ¹UFRGS ²UFMS Budapest, Hungary August 24, 2018

slide-2
SLIDE 2

High-level Architecture

Self-Driving Networks

2

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

Network Operator Intent Config Monitor Process Learn Adapt

slide-3
SLIDE 3

High-level Architecture

Self-Driving Networks

3

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

Network Operator Intent Config Monitor Process Learn Adapt

Focus of our work!

slide-4
SLIDE 4

Nowadays...

4

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-5
SLIDE 5

Nowadays...

5

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

Higher-level

slide-6
SLIDE 6

Nowadays...

6

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

NetKat

COCOON

PGA JANUS

Higher-level

slide-7
SLIDE 7

How to deploy intents expressed in natural language?

7

slide-8
SLIDE 8

A Network Intent Refinement using Nile

8

slide-9
SLIDE 9

A Network Intent Refinement using Nile

1. Receive intents expressed in natural language

9

slide-10
SLIDE 10

A Network Intent Refinement using Nile

1. Receive intents expressed in natural language 2. Use Nile to ask for operator feedback

10

slide-11
SLIDE 11

Intent Refinement By Example

11

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

Experimental Service Chaining scenario, using SONATA-NFV and Mininet Original scenario

slide-12
SLIDE 12

Intent Refinement By Example

12

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

“Please add a firewall and an IDS from Iperf client to server”

Original Intent

slide-13
SLIDE 13

Intent Refinement By Example

13

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

“Please add a firewall and an IDS from Iperf client to server”

Original Intent

slide-14
SLIDE 14

Intent Refinement By Example

14

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

“Please add a firewall and an IDS from Iperf client to server”

Original Intent

DialogFlow.com

slide-15
SLIDE 15

Intent Refinement By Example

15

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

“Please add a firewall and an IDS from Iperf client to server”

Extracted entities

slide-16
SLIDE 16

Intent Refinement By Example

16

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

“Please add a firewall and an IDS from Iperf client to server”

Extracted entities

slide-17
SLIDE 17

Intent Refinement By Example

17

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

“Please add a firewall and an IDS from Iperf client to server”

Extracted entities

Neural Sequence to Sequence learning model, using Recursive Neural Networks.

slide-18
SLIDE 18

Intent Refinement By Example

18

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

define intent testIntent: from endpoint('iperf client' ) to endpoint('iperf server' ) add middlebox('firewall'), middlebox( 'ids') Nile intent

slide-19
SLIDE 19

Intent Refinement By Example

19

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

define intent testIntent: from endpoint('iperf client' ) to endpoint('iperf server' ) add middlebox('firewall'), middlebox( 'ids') Nile intent Is this what you want? YES NO

slide-20
SLIDE 20

define intent testIntent: from endpoint('iperf client' ) to endpoint('iperf server' ) add middlebox('firewall'), middlebox( 'ids') Nile intent YES NO

Intent Refinement By Example

20

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

Is this what you want?

slide-21
SLIDE 21

Intent Refinement By Example

21

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

define intent testIntent: from endpoint('iperf client' ) to endpoint('iperf server' ) add middlebox('firewall'), middlebox( 'ids') Nile intent

slide-22
SLIDE 22

Intent Refinement By Example

22

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

define intent testIntent: from endpoint('iperf client' ) to endpoint('iperf server' ) add middlebox('firewall'), middlebox( 'ids') Nile intent

Nile compiler to SONATA-NFV commands

slide-23
SLIDE 23

Intent Refinement By Example

23

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018) # deploy vnfs vim-emu compute start -n fw <params> vim-emu compute start -n ids <params> # chain vnfs vim-emu network add -b -src iperf-c:c-eth0 -dst fw:in vim-emu network add -b -src fw:out -dst ids:in vim-emu network add -b -src ids:out -dst iperf-s:s-eth0

Compiled SONATA-NFV commands

slide-24
SLIDE 24

Intent Refinement By Example

24

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018) # deploy vnfs vim-emu compute start -n fw <params> vim-emu compute start -n ids <params> # chain vnfs vim-emu network add -b -src iperf-c:c-eth0 -dst fw:in vim-emu network add -b -src fw:out -dst ids:in vim-emu network add -b -src ids:out -dst iperf-s:s-eth0

Compiled SONATA-NFV commands

slide-25
SLIDE 25

Intent Refinement By Example

“Please add a firewall and an IDS from Iperf client to server” Resulting scenario

25

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-26
SLIDE 26

Evaluation

(i) The accuracy we can achieve with different sizes of training datasets, aiming to find the

  • ptimal ratio between dataset size and prediction accuracy.

(ii) The impact of the operator feedback on the accuracy of predictions over time to determine if it improves accuracy.

  • 5 dataset sizes:

○ 100, 500, 1000, 2000, 5000 entries. ○ 20% validation split.

  • We generated the datasets automatically with random sets of entities and Nile intent

pairs, combining a different number of middleboxes, endpoints, traffic matching rules, time, and QoS requirements in each intent.

26

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-27
SLIDE 27

Results

(i) The accuracy we can achieve with different sizes of training datasets, aiming to find the optimal ratio between dataset size and prediction accuracy.

27

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-28
SLIDE 28

Results

(ii) The impact of the operator feedback on the accuracy of predictions over time to determine if it improves accuracy.

28

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-29
SLIDE 29

Results

(ii) The impact of the operator feedback on the accuracy of predictions over time to determine if it improves accuracy.

29

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-30
SLIDE 30

Summary

“How to deploy network intents expressed as natural language?”

Using our refinement process + Nile Low-level of technical knowledge required Feedback from user allows to learn over time

“What’s next?”

Fully implement Nile compilation into OpenFlow and P4 backends. Further evaluate the end-to-end proposed solution.

30

ACM SIGCOMM 2018 Afternoon Workshop on Self-Driving Networks (SelfDN 2018)

slide-31
SLIDE 31

Thank you!

Arthur Jacobs asjacobs@inf.ufrgs.br

github.com/NetworkIntentAssistent

31