Joint Virtual Machine Placement and Migration in Dynamic - - PowerPoint PPT Presentation

▶
joint virtual machine placement and migration in dynamic
SMART_READER_LITE
LIVE PREVIEW

Joint Virtual Machine Placement and Migration in Dynamic - - PowerPoint PPT Presentation

Sep 02, 2023 351 likes •856 views

Joint Virtual Machine Placement and Migration in Dynamic Policy-Driven Data Centers Hugo Flores J Lucas California State University Dominguez Hills Department of Computer Science 1 Presentation Overview 1. Introduction 2. Related Works

slide-1
SLIDE 1

Joint Virtual Machine Placement and Migration in Dynamic Policy-Driven Data Centers

Hugo Flores J Lucas California State University Dominguez Hills Department of Computer Science

1

slide-2
SLIDE 2

Presentation Overview

1. Introduction 2. Related Works 3. System Model 4. Virtual Machine Migration 5. Virtual Machine Placement 6. Performance Evaluation 7. Conclusion

2

slide-3
SLIDE 3

Introduction

3

slide-4
SLIDE 4

What is a Dynamic Policy Driven Data Center (PDDC)?

  • Data Center

○ Physical Machines (PMs) ○ Switches ○ Virtual Machines (VMs)

  • Policy Driven

○ Middleboxes (MBs) ○ Policy Chains (Ordered or Unordered)

  • Dynamic

○ Communication Frequencies

4

slide-5
SLIDE 5

What is VM Placement?

5

slide-6
SLIDE 6

What is VM Migration?

6

slide-7
SLIDE 7

Goals

  • Given:

○ An empty PDDC ○ Policies (Ordered or Unordered) ○ Unplaced VM Pairs with Comm. Frequency

  • Output:

○ VM Placement with minimum Comm. cost

  • How:

○ Optimal Algorithm ○ Placement Approximation Algorithm

  • Given:

○ A PDDC ○ Policies (Ordered or Unordered) ○ Placed VM Pairs with new Comm. Frequency

  • Output:

○ VM Migration with minimum Comm. & Migration cost

  • How:

○ MCF Algorithm ○ Migration Approximation Algorithm

Virtual Machine Placement Virtual Machine Migration

7

slide-8
SLIDE 8

Related Works

8

slide-9
SLIDE 9

Virtual Machine Placement or Migration

  • Improving the Scalability of Data Center Networks with Traffic-aware Virtual Machine

Placement

○ 2010 Proceedings IEEE INFOCOM ○

  • X. Meng, V. Pappas, & L. Zhang

○ TrafficAware Algorithm

  • PACE: Policy-Aware Application Cloud Embedding

○ 2013 Proceedings IEEE INFOCOM ○

  • L. E. Li et al.
  • PLAN: Joint Policy- and Network-Aware VM Management for Cloud Data Centers

○ 2016 IEEE Transactions on Parallel and Distributed Systems ○

  • L. Cui et al.

○ PLAN Algorithm

9

slide-10
SLIDE 10

Virtual Machine Placement and Migration

  • Joint Virtual Machine Placement and Migration Scheme for Data Centers

○ 2014 IEEE Global Communications Conference ○

  • T. Duong-Ba, T. Nguyen, B. Bose, & T. Tran
  • Traffic-Aware Virtual Machine Migration in Topology-Adaptive DCN

○ 2017 IEEE/ACM Transactions on Networking ○

  • Y. Cui et al.

10

slide-11
SLIDE 11

System Model

11

slide-12
SLIDE 12

Datacenter

  • Fat Tree Topology

○ K-parameter determines number of PMs & switches

  • PDDC:

○ Undirected Graph G(V, E) ○ V = VP ∪ VS ○ E is the set all edges

  • Physical Machines:

○ i-th PM has m(i ) resource slots ○ Each VM requires 1 slot

12

slide-13
SLIDE 13

Middleboxes

  • Set of Middleboxes:

○ M = { mb1 , mb2 , … , mbm }

  • MB Switch:

○ mbj → sw (j ) ∈ VS

  • Bump Off the Wire Design

13

slide-14
SLIDE 14

VM Pairs

  • VM Pairs:

○ P = { (v1 , v’1 ) , (v2 , v’2 ) , … , (vL , v’L ) } ○ vi = Source VM ○ v’i = Destination VM

  • Communication Frequency:

○ ƛ = 〈 ƛ1 , ƛ2 , … , ƛL 〉 ○ Non-constant vector

14

slide-15
SLIDE 15

Policies

  • Ordered Policies

○ ( mb1 , mb2 , … , mbm ) ○ Ingress Switch = First MB visited ○ Egress Switch = Last MB visited ○ Sequential MB Dependencies

  • Unordered Policies

○ { mb1 , mb2 , … , mbm } ○ Independant MBs

15

slide-16
SLIDE 16

Costs

  • Distance Cost

○ c ( i , j )

  • VM Pair Communication Cost

○ ( frequency ) * ( number of hops )

  • VM Pair Migration Cost

○ μ * c ( i , j )

16

slide-17
SLIDE 17

Virtual Machine Migration

17

slide-18
SLIDE 18

Ordered Policy Goal

18

slide-19
SLIDE 19

Ordered Policy Goal

19

MB Traversal Cost Migration and Ingress Cost Migration and Egress Cost

slide-20
SLIDE 20

Ordered Policy Solution - MCF Algorithm

1. Add Source & Sink Node: 2. Connect Source/Sink to VMs/PMs: 3. Source to VM: capacity 1, cost 0 & PM to Sink: capacity mj , cost 0

4.

Source VM to PM edges: capacity 1, cost: Destination VM to PM edges: capacity 1, cost:

5. Supply = 2L, Demand = 2L

20

slide-21
SLIDE 21

Ordered Policy Solution - MCF Algorithm

21

slide-22
SLIDE 22

Ordered Policy Solution - MCF Algorithm

22

slide-23
SLIDE 23

Unordered Policy Goal

23

slide-24
SLIDE 24

Unordered Policy Goal

24

Migration Cost Variable MB Cost Cost to First MB Cost to Last MB

slide-25
SLIDE 25

Unordered Policy Solution - Approximation

25

slide-26
SLIDE 26

Unordered Policy Solution - Approximation

26

slide-27
SLIDE 27

Sketch of Optimal Proof

27

slide-28
SLIDE 28

Virtual Machine Placement

28

slide-29
SLIDE 29

Ordered Policy Goal

29

slide-30
SLIDE 30

Ordered Policy Goal

30

MB Traversal Cost Ingress and Egress Cost

slide-31
SLIDE 31

Ordered Policy Solution - Optimal

31

slide-32
SLIDE 32

Ordered Policy Solution - Optimal

32

slide-33
SLIDE 33

Unordered Policy Goal

33

Cost to First MB Variable MB Traversal Cost Cost to Last MB

slide-34
SLIDE 34

Unordered Policy Solution - Approximation

34

slide-35
SLIDE 35

Unordered Policy Solution - Approximation

35

slide-36
SLIDE 36

Unordered Policy Solution - Approximation

36

slide-37
SLIDE 37

Performance Evaluation

37

slide-38
SLIDE 38

Common Simulation Parameters

  • Fat Tree Topology (k = 8)

○ 128 Physical Machines ○ Frequency Range [1, 1000]

  • Varying One of the Following (Placement):

○ Number of VM Pairs ○ Number of MBs ○ Number of Resource Slots

  • Varying Mu Parameter (Migration)

38

slide-39
SLIDE 39

Ordered Placement - VM Simulation (rc = 40, mb = 3)

39

slide-40
SLIDE 40

Ordered Placement - MB Simulation (rc = 40, l = 1000)

40

slide-41
SLIDE 41

Ordered Placement - RC Simulation (l = 1000, mb = 3)

41

slide-42
SLIDE 42

Unordered Placement - VM Simulation (rc = 40, mb = 3)

42

slide-43
SLIDE 43

Unordered Placement - MB Simulation (rc = 40, l = 1000)

43

slide-44
SLIDE 44

Unordered Placement - RC Simulation (l = 1000, mb = 3)

44

slide-45
SLIDE 45

Ordered Migration - (l = 1000, mb = 3, rc=40)

45

slide-46
SLIDE 46

Unordered Migration - (l = 1000, mb = 3, rc=40)

46

slide-47
SLIDE 47

Conclusion

47

slide-48
SLIDE 48

Conclusion

  • Placement Special Case of Migration
  • Ignoring PDDC constraints leads to Inefficiencies
  • Future Work:

○ Testing in Real Networks ○ Variable ‘sized’ VMs ○ Network Function Virtualization (NFVs)

48