Allocation Considering Dependability Issues Victor Lira Orientador: - - PowerPoint PPT Presentation

Victor Lira Orientador: Eduardo Tavares

Virtual Network Resource Allocation Considering Dependability Issues

Introduction

• Internet notably has a vital role in society;

– Entertainment; – Education; – Health; – So on;

Internet’s Ossification

Speed, Capacity, New Applications Architecture Innovations (e.g., for better mobility support)

Network Virtualization

• Promising approach to deal with Internet’s
• ssification problem;
• Coexistence of multiple instances of virtual

networks on a single shared physical infrastructure;

• Flexibility in the topology, manageability,

scalability and traffic isolation;

Network Virtualization

Dependability

• Ability of a system to deliver a particular

service in a reliable way;

• Metric/attribute of interest:

– Availability;

• Probability of a system being in a functioning condition.

It considers the alternation of operational and nonoperating states;

Proposed Method

PROBLEM FORMULATION

Substrate/Virtual Network

• The physical network is represented by an

undirected weighted graph GS = (NS, ES);

• 𝑜𝑇 ∈ 𝑂𝑇 → Nodes;
• 𝑓𝑇 𝑗, 𝑘 ∈ 𝐹𝑇 → Links;
• A VN request is denoted by 𝐻𝑊 = (𝑂𝑊, 𝐹𝑊);
• 𝐸(𝐻𝑊) → Availability Constraint;

Substrate Network Resources

• The remaining or available capacity of a

physical node, 𝑆𝑂 𝑜𝑇 , 𝑜𝑇 ∈ 𝑂𝑇, is defined by: 𝑆𝑂 𝑜𝑇 = 𝑑 𝑜𝑇 − 𝑑(𝑜𝑊)

∀𝑜𝑊↑𝑜𝑇

in which 𝑦 ↑ 𝑧 means that the virtual node 𝑦 is mapped on the physical node 𝑧

Substrate Network Resources

• Also, the available bandwidth of a path

𝑄 ∈ 𝑄𝑇 is given by: 𝑆𝐹 𝑄 = 𝑛𝑗𝑜

𝑓𝑇∈𝑄 𝑆𝐹 𝑓𝑇

Virtual Network Allocation

• For each VN request received, the VNP

accepts or rejects the request, according to the available resources and constraints;

• In case of acceptance, a mapping for the VN
• n the physical network is accomplished,

reserving the required network resources;

Virtual Network Allocation

• VN mapping is split into activities: (i) node

mapping and (ii) link mapping.

• Besides, all requests are subject to:

𝐵𝑤 𝐻𝑊 ≥ 𝐸 𝐻𝑊

Node Mapping

• Each virtual node is mapped into a physical

node using 𝑁𝑂 ∶ 𝑂𝑊 → 𝑂𝑇, so that, ∀𝑜𝑊 ∈ 𝑂𝑊: 𝑑 𝑜𝑊 ≤ 𝑆𝑂 𝑁𝑂(𝑜𝑊)

Node Mapping

• If redundancy is adopted, an additional

mapping 𝑁𝑇𝑂 ∶ 𝑂𝑊 → 𝑂𝑇 is considered, such that, ∀𝑜𝑊 ∈ 𝑂𝑊, 𝑁𝑇𝑂(𝑜𝑊) ≠ 𝑁𝑂(𝑜𝑊) subject to: 𝑑 𝑜𝑊 ≤ 𝑆𝑂 𝑁𝑇𝑂(𝑜𝑊)

Node Mapping

• In addition, considering cold standby

redundancy, ∀𝑜𝑊 ∈ 𝑂𝑊: 𝑁𝑇𝑂(𝑜𝑊) ≠ 𝑁𝑂(𝑛𝑊) 𝑁𝑂 𝑜𝑊 = 𝑁𝑂 𝑛𝑊 , 𝑗𝑔𝑔(𝑜𝑊=𝑛𝑊) 𝑁𝑇𝑂 𝑜𝑊 = 𝑁𝑇𝑂 𝑛𝑊 , 𝑗𝑔𝑔(𝑜𝑊=𝑛𝑊)

Link Mapping

• The mapping of virtual links to physical paths is

defined by 𝑁𝑁𝐹 ∶ 𝐹𝑊 → 𝑄𝑇(𝑁𝑂 𝑛𝑊 , 𝑁𝑂 𝑜𝑊 ), such that, for any 𝑓𝑊 = (𝑛𝑊, 𝑜𝑊) ∈ 𝐹𝑊: 𝑆𝐹 𝑞 ≥ 𝑐 𝑓𝑊 , ∀𝑞 ∈ 𝑁𝑁𝐹 𝑓𝑊

Link Mapping

• In VN requests with redundancy, three additional

virtual links are required due to redundant nodes:

• 1. Spare-primary:

𝑁𝑇𝑄: 𝐹𝑊 → 𝑄𝑇(𝑁𝑇𝑂 𝑛𝑊 , 𝑁𝑂 𝑜𝑊 );

• 2. Primary-spare:

𝑁𝑄𝑇 ∶ 𝐹𝑊 → 𝑄𝑇(𝑁𝑂 𝑛𝑊 , 𝑁𝑇𝑂 𝑜𝑊 );

• 3. Spare-spare:

𝑁𝑇𝑇 ∶ 𝐹𝑊 → 𝑄𝑇(𝑁𝑇𝑂 𝑛𝑊 , 𝑁𝑇𝑂 𝑜𝑊 );

Link Mapping

• They are mappings from virtual links to physical

paths, such that, for any 𝑓𝑊 = (𝑛𝑊, 𝑜𝑊) ∈ 𝐹𝑊, 𝑆𝐹 𝑞 ≥ 𝑐 𝑓𝑊 , ∀𝑞 ∈ 𝑁𝑇𝑄 𝑓𝑊 𝑆𝐹 𝑞 ≥ 𝑐 𝑓𝑊 , ∀𝑞 ∈ 𝑁𝑄𝑇 𝑓𝑊 𝑆𝐹 𝑞 ≥ 𝑐 𝑓𝑊 , ∀𝑞 ∈ 𝑁𝑇𝑇 𝑓𝑊

Objective

• Allocating VN requests to meet specified constraints

(e.g., availability), minimizing the cost resulting from allocations: 𝑔

𝑓𝑇 𝑓𝑊 𝑓𝑇∈ 𝐹𝑇 𝑓𝑊∈ 𝐹𝑊

+ 𝑑(𝑜𝑊)

𝑜𝑊∈ 𝑂𝑊

∗ 𝑦

in which 𝑔

𝑓𝑇 𝑓𝑊 represents the total bandwidth allocated on link 𝑓𝑇 to the

virtual link 𝑓𝑊. 𝑦 is an integer variable , which is equal to ‘2’ whenever redundancy is considered on VN request. Otherwise, the value is equal to ‘1’.

DEPENDABILITY MODELING

No redundancy

• 𝑁𝑂 𝑇1 = 𝐵;
• 𝑁𝑂 𝑇2 = 𝐶;
• 𝑁𝑁𝐹 𝑇1, 𝑇2 = (𝐵, 𝐶);

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

No redundancy

A B LINK A-B

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

Hot Standby

• 𝑁𝑂 𝑇1 = 𝐷;
• 𝑁𝑇𝑂 𝑇1 = 𝐹;
• 𝑁𝑂 𝑇2 = 𝐸;
• 𝑁𝑇𝑂 𝑇2 = 𝐹;
• 𝑁𝑁𝐹 𝑇1, 𝑇2 = { 𝐷, 𝐸 };
• ...

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

Hot Standby

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

C LINK C-D D E LINK D-E D C LINK C-D LINK D-E E E C E D E

S1 node S2 node S1-S2 link

Cold Standby

• 𝑁𝑂 𝑇1 = 𝐵;
• 𝑁𝑇𝑂 𝑇1 = 𝐶;
• 𝑁𝑂 𝑇2 = 𝐷;
• 𝑁𝑇𝑂 𝑇2 = 𝐸;
• 𝑁𝑁𝐹 𝑇1, 𝑇2 = { 𝐵, 𝐷 };
• ...

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

Cold Standby

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

A_ON A_Repair A_Failure A_OFF B_ON B_Repair B_Failure B_OFF Wait_B Activate_B Deactivate_B C_ON C_Repair C_Failure C_OFF D_ON D_Repair D_Failure D_OFF Wait_D Activate_D Deactivate_D LINK_A_C_ON LINK_A-C_Repair LINK_A_C_OFF LINK_A-C_Failure LINK_B-D_ON LINK_B_D_Repair LINK_B_D_OFF LINK_B_D_Failure LINK_C_D_ON LINK_C_D_Repair LINK_C_D_OFF LINK_C_D_Failure

S1 node S2 node S1-S2 link

Cold Standby

S1 S2 A

10

C

12

B

15

D

10

E

20 22 25 25 30 23 11

VN Request Substrate Network Topology

10 10

P{( ((#A_ON + #B_ON>0)) AND ( )} ((#A_ON > 0)AND(#LINK_A_C_ON > 0)AND(#C_ON > 0)) OR ((#B_ON > 0)AND(#LINK_B_D_ON > 0)AND(#LINK_C_D_ON > 0)AND(#C_ON > 0)) OR ((#A_ON > 0)AND(#LINK_A_C_ON > 0)AND(#LINK_C_D_ON > 0)AND(#D_ON > 0)) OR ((#B_ON > 0)AND(#LINK_B_D_ON > 0)AND(#D_ON > 0)) ) AND ((#C_ON + #D_ON>0))

S1 node S2 node S1-S2 link

GRASP FOR VIRTUALIZED NETOWRK ALLOCATION

GRASP

• GRASP (Greedy Randomized Adaptive Search

Procedure);

• Two phases:

– Construction; – Local search;

GRASP – Construction Phase

GRASP – Local Search

EXPERIMENTAL RESULTS

Experiment Settings

• GT-ITM tool to generate the physical network

topology;

• Substrate network:

– 50 nodes randomly conected with probability 0.5; – Nodes capacities and link bandwidths are real numbers uniformly distributed between 50 and 100;

Experiment Settings

• 800 VN requests are considered over a period
• f 50,000 hours;
• 0.9 (90%) is the availability constraint for each

VN request;

Results - Cost

1000 2000 3000 4000 5000 6000 1 9 17 25 33 41 49

Average C

• st

Time (thousands of hours) No R edundancy Hot S tandby C

• ld S

tandby R

• ViNE

Results - Availability

0,7 0,75 0,8 0,85 0,9 0,95 1 1 9 17 25 33 41 49

Average Availability

Time (thousands of hours) No R edundancy Hot S tandby C

• ld S

tandby R

• ViNE

Results – Availability ECDF

1.00 0.95 0.90 0.85 0.80 1.0 0.8 0.6 0.4 0.2 0.0 Availability P{ X < = x}

no redundancy R-ViNE hot standby cold standby

Results – Acceptance Rate

0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1 9 17 25 33 41 49

Average Acceptance Rate

Time (thousands of hours) No R edundancy Hot S tandby C

• ld S

tandby

Conclusion

• Network Virtualization has received particular

attention from the scientific community, as several VNs can coexist in the same physical network;

• Many algorithms have been proposed to

allocate VNs considering performance metrics. However, dependability is usually neglected.

Conclusion

• This work proposes a GRASP-based algorithm

for allocating virtual networks taking into account dependability issues;

Thanks!