Federated clouds and cloud brokering allow migration of virtual machines across clouds and even deployment of cooperating VMs in different cloud data centers. In order to fully benefit from these new opportunities, we propose a heuristic that outputs a matching between virtual machine and cloud data centers by taking resource capacities, VM topologies, performance and resource costs into account. Results of our initial evaluation using the CloudSim Framework indicate that, proposed heuristic is promising for a better optimized placement of networked VM groups onto the federated cloud topology.

1. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Subgraph Matching for Resource Allocation in the
Federated Cloud Environment
Atakan Aral, Tolga Ovatman
Istanbul Technical University – Department of Computer Engineering
June 27, 2015
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

2. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Outline
1 Introduction
2 Problem Modeling
Topology Modeling
Bandwidth Modeling
Cost Modeling
3 TBM Algorithm
4 Evaluation
Experimental Setup
Results
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

3. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Outline
1 Introduction
2 Problem Modeling
Topology Modeling
Bandwidth Modeling
Cost Modeling
3 TBM Algorithm
4 Evaluation
Experimental Setup
Results
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

4. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Geo-Distributed Clusters
Opportunities:
Available mechanisms and policies such as Federated Cloud;
Very high speed inter-DC communication technologies such as optical fiber;
Programming models that minimize size of data flow between nodes such as
MapReduce
Advantages:
fault tolerance
vendor independence
closer proximity to user base
cost benefits
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

5. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Geo-Distributed Clusters
Risks (regarding VM placement):
Cooperating VMs on distant DCs;
Clusters far away from their user base;
VMs placed without considering different pricing strategies of vendors
Our Objectives:
To decrease communication delay (by placing connected VMs to the
neighbour data centers)
To decrease deployment delay (by placing VMs close to the broker)
To reduce resource costs (by balancing load and avoiding overload in any DC)
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

6. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Topology Modeling
Bandwidth Modeling
Cost Modeling
Outline
1 Introduction
2 Problem Modeling
Topology Modeling
Bandwidth Modeling
Cost Modeling
3 TBM Algorithm
4 Evaluation
Experimental Setup
Results
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

7. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Topology Modeling
Bandwidth Modeling
Cost Modeling
Topology Modeling
Weighted, undirected, simple graphs
Vertice represent cloud data centers / requested VMs.
CPU, Memory, Storage
Edges represent the network connections between them.
Bandwidth, Latency
Brokers represent the user base at each node
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

8. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Topology Modeling
Bandwidth Modeling
Cost Modeling
Bandwidth Modeling
VM1 VM3VM2
VM1
VM2VM3
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

9. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Topology Modeling
Bandwidth Modeling
Cost Modeling
Cost Modeling
1 Fixed pricing based on memory, bandwidth and duration.
2 Dynamic pricing via Yield management
Increase the price of the resource that is running low in a DC
Cost = minCost + (maxCost − minCost) ∗ Util
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

10. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Outline
1 Introduction
2 Problem Modeling
Topology Modeling
Bandwidth Modeling
Cost Modeling
3 TBM Algorithm
4 Evaluation
Experimental Setup
Results
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

11. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Topology Based Matching
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

12. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Experimental Setup
Results
Outline
1 Introduction
2 Problem Modeling
Topology Modeling
Bandwidth Modeling
Cost Modeling
3 TBM Algorithm
4 Evaluation
Experimental Setup
Results
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

13. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Experimental Setup
Results
Experimental Setup
Number of Clusters Based on the population density
around each location.
Number of VMs Based on Poisson distribution: λ = 3
Cluster Topologies Either linear or complete
Arrival Times Uniform random in the range [0, 50)
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

14. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Experimental Setup
Results
Latencies
0,00
50,00
100,00
150,00
200,00
250,00
300,00
1 2 3 4 5 6 7 8
VMDeploymentLatency(Seconds)
VM RAM
ANF LBG RAN TBF LNF
0,0
0,5
1,0
1,5
2,0
2,5
3,0
1 2 3 4 5 6 7 8
VMCommunicationLatency(Seconds)
VM RAM
ANF LBG RAN TBF LNF
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

15. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Experimental Setup
Results
Duration and Throughput
0
50
100
150
200
250
300
350
400
1 2 3 4 5 6 7 8
TaskCompletionTime(Hours)
VM RAM
ANF LBG RAN TBF LNF
0
500
1000
1500
2000
2500
3000
1 2 3 4 5 6 7 8
Throughput(MIPS)
VM RAM
ANF LBG RAN TBF LNF
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

16. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Experimental Setup
Results
Rejection Rate and Cost
0
10
20
30
40
50
60
70
80
90
100
1 2 3 4 5 6 7 8
RejectionRate(%)
VM RAM
ANF LBG RAN TBF LNF
0
10000
20000
30000
40000
50000
60000
1 2 3 4 5 6 7 8
Cost($)
VM RAM
ANF LBG RAN TBF LNF
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

17. Introduction
Problem Modeling
TBM Algorithm
Evaluation
Experimental Setup
Results
Thank you for your attention.
Atakan Aral
Istanbul Technical University
Department of Computer Engineering
aralat@itu.edu.tr
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

18. Appendix
Subgraph Matching
Future Work
More Results
Outline
5 Appendix
Subgraph Matching
Future Work
More Results
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

19. Appendix
Subgraph Matching
Future Work
More Results
Subgraph Matching
Search space is all possible injective matchings from the set of pattern nodes
to the set of target nodes.
Systematically explore the search space:
Start from an empty matching
Extend the partial matching by matching a non matched pattern node to a non
matched target node
Backtrack if some edges are not matched
Repeat until all pattern nodes are matched (success) or all matchings are
already explored (fail).
Filters are necessary to reduce the search space by pruning branches that do
not contain solutions.
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

20. Appendix
Subgraph Matching
Future Work
More Results
LAD Filtering
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

21. Appendix
Subgraph Matching
Future Work
More Results
LAD Filtering
D1 = D3 = D5 = D6 = A, B, C, D, E, F, G
D2 = D4 = A, B, D
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

22. Appendix
Subgraph Matching
Future Work
More Results
Future Work
Algorithm
Additional constraints (jurisdiction, partially known topology)
Vertical scaling support
Hybrid cloud support
Homeomorphism
Connected components
Evaluation
Significance study
Evaluation with topology improvements
Multi-objective optimization
Dynamic heuristic selection, meta-heuristics
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

23. Appendix
Subgraph Matching
Future Work
More Results
Link bandwidth request
0,0
0,5
1,0
1,5
2,0
2,5
1 2 3 4 5 6 7 8
VMCommunicationLatency(Seconds)
Link BW
ANF LBG RAN TBF LNF
0
10000
20000
30000
40000
50000
60000
1 2 3 4 5 6 7 8
Cost($)
Link BW
ANF LBG RAN TBF LNF
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

24. Appendix
Subgraph Matching
Future Work
More Results
Minimum number of requests
0
500
1000
1500
2000
2500
1 2 3 4 5 6 7 8
Throughput(MIPS)
Number of Requests
ANF LBG RAN TBF LNF
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment

25. Appendix
Subgraph Matching
Future Work
More Results
VM network intensity
0
50
100
150
200
250
300
350
1 2 3 4 5 6 7 8
TaskCompletionTime(Hours)
Network intensity
ANF LBG RAN TBF LNF
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
1 2 3 4 5 6 7 8
Cost($)
Network intensity
ANF LBG RAN TBF LNF
Atakan Aral, Tolga Ovatman Subgraph Matching for Resource Allocation in the Federated Cloud Environment