Simplifying Middlebox Policy Enforcement Using SDN

SIMPLE-fying Middlebox Policy
Enforcement Using SDN
Zafar Ayyub Qazi
Cheng-Chun Tu 
Luis Chiang
Vyas Sekar
Rui Miao
Minlan Yu
Middleboxes management is hard!
2
Critical for security, performance, compliance
But expensive, complex and difficult to manage
Survey across 57 network operators  
(J. Sherry et al. SIGCOMM 2012)
Can SDN simplify middlebox management?
Centralized Controller 
OpenFlow
3
 
Proxy
 
IDS
Necessity + Opportunity:
Incorporate functions markets views as important
Scope
: Enforce middlebox-specific steering policies
What makes this problem challenging?
Centralized Controller 
OpenFlow
4
Proxy
IDS
Middleboxes introduce new dimensions beyond L2/L3 tasks.
Achieve this with 
unmodified
 
middleboxes and 
existing
 
SDN APIs
Our Work: SIMPLE
5
Policy enforcement layer for
middlebox-specific “traffic steering”
Outline
Motivation
Challenges
SIMPLE Design
Evaluation
Conclusions
6
6
Challenge: Policy Composition
S1
S2
7
Firewall
Proxy
IDS
Firewall
IDS
Proxy
*
Policy Chain:
Oops!
Forward Pkt
to IDS or Dst?
Dst
“Loops”
Traditional flow rules may not suffice!
Challenge: Resource Constraints
S1
S2
S4
S3
Proxy
Firewall
Can we set up “feasible” forwarding rules?
8
9
 
S1
Proxy
S2
User 1
User 2
Proxy may
modify
flows
Are forwarding rules at S2 correct?
Challenge: Dynamic Modifications
Firewall
User1: Proxy 
 Firewall
User2: Proxy
New dimensions beyond Layer 2-3 tasks
1) Policy Composition 
 Potential loops
3) Dynamic Modifications 
 Correctness?
2) Resource Constraints 
 Switch + Middlebox
10
Can we address these with 
unmodified
middleboxes and 
existing
 
SDN APIs?
Outline
Motivation + Context for the Work
Challenges
SIMPLE Design
Evaluation
Conclusion
11
SIMPLE System Overview
12
Composition 
 
 Tag Processing State
13
Firewall
IDS
Proxy
*
Policy Chain:
S1
S2
Firewall
Proxy
IDS
Dst
ORIGINAL
Post-Firewall
Post-IDS
Post-Proxy
Fwd to
Dst
Insight: Distinguish different instances of the same packet
SIMPLE System Overview
14
Resource Constraints
 
Joint Optimization
Resource Manager
Optimal & Feasible
load balancing
Theoretically hard!
Not obvious if some configuration is feasible!
15
Offline + Online Decomposition
16
Offline Stage
Online Step
Deals with Switch constraints
Deals with only load balancing
Resource Manager
Network 
Topology
Switch 
TCAM
Policy 
Spec
Traffic
Matrix
Mbox Capacity 
+ Footprints
Offline Stage: ILP based pruning
17
 
Set of all possible middlebox
load distributions
 
Pruned Set
 
 
Balance the middlebox load
Feasible
Sufficient freedom
SIMPLE System Overview
18
Modifications 
 Infer flow correlations
19
 
Correlate
flows
Install
rules
S1
Proxy
S2
User 1
User 2
Firewall
User1: Proxy 
 Firewall
User2: Proxy
Payload
Similarity
SIMPLE Implementation
OpenFlow 1.0
POX
extensions
20
CPLEX
Outline
Motivation + Context for the Work
Challenges
SIMPLE Design
Evaluation
Conclusion
21
Evaluation and Methodology
What benefits SIMPLE offers? load balancing?
How scalable is the SIMPLE optimizer?
How close is the SIMPLE optimizer to the optimal?
How accurate is the dynamic inference?
Methodology
Small-scale real test bed experiments (Emulab)
Evaluation over Mininet (with up to 60 nodes)
Large-scale trace driven simulations (for convergence times)
22
Benefits: Load balancing
4-7X better load balancing and near optimal
23
Optimal
Overhead: Reconfiguration Time
Around 125 ms to reconfigure, most time spent in pushing rules
24
33 node topology
including 11 switches
Other Key Results
LP solving takes 1s for a 252 node topology
4-5 orders of magnitude faster than strawman
95 % accuracy in inferring flow correlations
Scalability of pruning: 1800s 
 
110s
25
Conclusions
 
Middleboxes: Necessity and opportunity for SDN
 
Goal: Simplify middlebox-specific policy enforcement
 
Challenges: Composition, resource constraints, modifications
 
SIMPLE: policy enforcement layer
Does not modify middleboxes
No changes to SDN APIs
No visibility required into the internal of middleboxes
 
Scalable and offers 4-7X improvement in load balancing
26
 
 
27
Decompose Optimization:
Slow Offline + Fast Online Steps
Policy 
Spec
Network 
Topology
Enumerate
Physical
Sequences
Prune for
Feasible
Configs
Rule
Model
 
Offline Pruning
28
Enumerating Physical Sequences
29
S1
S6
S2
S5
S3
 
Proxy
 
Policy Chains
Proxy1
IDS1
S4
29
Slide Note

Hello everyone, I am Zafar, Today I will be talking about how we can simplify middlebox management using Software Defined networking. This is a joint work with collaborators from Stony Brook university and university of southern California.

Embed
Share

Middlebox management is a complex task for network operators, with challenges in security, performance, and compliance. This research explores how Software-Defined Networking (SDN) can simplify middlebox management by enforcing middlebox-specific steering policies through a centralized controller. The study addresses the difficulties of managing a large number of middleboxes and introduces a solution called SIMPLE that incorporates various network functions and market views for efficient policy enforcement. The work focuses on designing a policy enforcement layer for middlebox-specific traffic steering using OpenFlow-capable legacy middleboxes.

  • SDN
  • Middlebox Management
  • Network Operators
  • Policy Enforcement
  • OpenFlow

Uploaded on Sep 22, 2024 | 0 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript


  1. SIMPLE-fying Middlebox Policy Enforcement Using SDN Zafar Ayyub Qazi Cheng-Chun Tu Luis Chiang Vyas Sekar Rui Miao Minlan Yu

  2. Middleboxes management is hard! Survey across 57 network operators (J. Sherry et al. SIGCOMM 2012) e.g., a network with ~2000 middleboxes required 500+ operators Critical for security, performance, compliance But expensive, complex and difficult to manage 2

  3. Can SDN simplify middlebox management? Centralized Controller Web Firewall IDS Proxy OpenFlow Flow FwdAction Flow FwdAction Proxy IDS Scope: Enforce middlebox-specific steering policies Necessity + Opportunity: Incorporate functions markets views as important 3

  4. What makes this problem challenging? Centralized Controller Web Firewall IDS Proxy OpenFlow Flow FwdAction Flow FwdAction Proxy IDS Middleboxes introduce new dimensions beyond L2/L3 tasks. Achieve this with unmodifiedmiddleboxes and existingSDN APIs 4

  5. Our Work: SIMPLE Web Firewall IDS Proxy Policy enforcement layer for middlebox-specific traffic steering Flow Action Flow Action OpenFlow capable Legacy Middleboxes 5

  6. Outline Motivation Challenges SIMPLE Design Evaluation Conclusions 6 6

  7. Challenge: Policy Composition Firewall IDS Proxy * Policy Chain: IDS Proxy Firewall Oops! Forward Pkt to IDS or Dst? S1 S2 Dst Loops Traditional flow rules may not suffice! 7

  8. Challenge: Resource Constraints Firewall Proxy Space for traffic split? S2 S4 S1 S3 IDS1 = 50% IDS2 = 50% Can we set up feasible forwarding rules? 8

  9. Challenge: Dynamic Modifications User1: Proxy Firewall User2: Proxy Proxy may modify flows User 1 Proxy S1 S2 User 2 Firewall Are forwarding rules at S2 correct? 9

  10. New dimensions beyond Layer 2-3 tasks 1) Policy Composition Potential loops 2) Resource Constraints Switch + Middlebox 3) Dynamic Modifications Correctness? Can we address these with unmodified middleboxes and existingSDN APIs? 10

  11. Outline Motivation + Context for the Work Challenges SIMPLE Design Evaluation Conclusion 11

  12. SIMPLE System Overview Web Firewall IDS Proxy Modifications Handler Resource Manager Rule Generator Flow Action Flow Action OpenFlow capable Legacy Middleboxes 12

  13. Composition Tag Processing State Firewall IDS Proxy * Policy Chain: IDS Proxy Firewall Fwd to Dst S1 S2 Dst Post-Firewall Post-Proxy ORIGINAL Post-IDS Insight: Distinguish different instances of the same packet 13

  14. SIMPLE System Overview Web Firewall IDS Proxy Modifications Handler Resource Manager Rule Generator Flow Action Flow Action OpenFlow capable Legacy Middleboxes 14

  15. Resource Constraints Joint Optimization Topology & Traffic Middlebox Capacity + Footprints Policy Spec Switch TCAM Resource Manager Optimal & Feasible load balancing Theoretically hard! Not obvious if some configuration is feasible! 15

  16. Offline + Online Decomposition Mbox Capacity + Footprints Switch TCAM Policy Spec Network Topology Traffic Matrix Resource Manager Offline Stage Online Step Deals with Switch constraints Deals with only load balancing 16

  17. Offline Stage: ILP based pruning Feasible Sufficient freedom Set of all possible middlebox load distributions Pruned Set Balance the middlebox load 17

  18. SIMPLE System Overview Web FW IDS Proxy Modifications Handler Resource Manager Rule Generator Flow Action Flow Action OpenFlow capable Legacy Middleboxes 18

  19. Modifications Infer flow correlations Correlate flows Install rules Payload Similarity User 1 Proxy S1 S2 User 2 Firewall User1: Proxy Firewall User2: Proxy 19

  20. SIMPLE Implementation Web FW IDS Proxy Modifications Handler (Dynamic modifications) Resource Manager (Resource Constraint) CPLEX Rule Generator (Policy Composition) POX extensions OpenFlow 1.0 Flow Tag/Tun nel Action Flow Tag/Tun nel Action 20

  21. Outline Motivation + Context for the Work Challenges SIMPLE Design Evaluation Conclusion 21

  22. Evaluation and Methodology What benefits SIMPLE offers? load balancing? How scalable is the SIMPLE optimizer? How close is the SIMPLE optimizer to the optimal? How accurate is the dynamic inference? Methodology Small-scale real test bed experiments (Emulab) Evaluation over Mininet (with up to 60 nodes) Large-scale trace driven simulations (for convergence times) 22

  23. Benefits: Load balancing Optimal 4-7X better load balancing and near optimal 23

  24. Overhead: Reconfiguration Time 33 node topology including 11 switches Around 125 ms to reconfigure, most time spent in pushing rules 24

  25. Other Key Results LP solving takes 1s for a 252 node topology 4-5 orders of magnitude faster than strawman 95 % accuracy in inferring flow correlations Scalability of pruning: 1800s 110s 25

  26. Conclusions Middleboxes: Necessity and opportunity for SDN Goal: Simplify middlebox-specific policy enforcement Challenges: Composition, resource constraints, modifications SIMPLE: policy enforcement layer Does not modify middleboxes No changes to SDN APIs No visibility required into the internal of middleboxes Scalable and offers 4-7X improvement in load balancing 26

  27. 27

Related


More Related Content

giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#