Presentation is loading. Please wait.

Presentation is loading. Please wait.

Sharing Cloud Networks Lucian Popa, Gautam Kumar, Mosharaf Chowdhury Arvind Krishnamurthy, Sylvia Ratnasamy, Ion Stoica UC Berkeley.

Published byEmmalee Fawcett Modified over 9 years ago

Similar presentations

Presentation on theme: "Sharing Cloud Networks Lucian Popa, Gautam Kumar, Mosharaf Chowdhury Arvind Krishnamurthy, Sylvia Ratnasamy, Ion Stoica UC Berkeley."— Presentation transcript:

1 Sharing Cloud Networks Lucian Popa, Gautam Kumar, Mosharaf Chowdhury Arvind Krishnamurthy, Sylvia Ratnasamy, Ion Stoica UC Berkeley

2 State of the Cloud Network??? 2

3 Guess the Share 3 B2B2 A1A1 B1B1 A2A2 A3A3 Link L Alice : Bob = ? : ? 3 : 11 : 12 : 13 : 2 TCP Per flowPer Source Per Destination Per-VM Proportional

4 Challenges Network share of a virtual machine (VM) V depends on »Collocated VMs, »Placement of destination VMs, and »Cross-traffic on each link used by V Network differs from CPU or RAM »Distributed resource »Usage attribution (source, destination, or both?) Traditional link sharing concepts needs rethinking 4

5 Requirements 5 High Utilization Min Bandwidth Guarantee Aggregate Proportionality Network shares proportional to the number of VMs Introduce performance predictability Do not leave bandwidth unused if there is demand

6 Requirement 1: Guaranteed Minimum B/W Provides a minimum b/w guarantee for each VM Captures the desire of tenants to get performance isolation for their applications 6 ABX B minA B minB B minX … All VMs

7 Requirement 2: Aggregate Proportionality Shares network resources across tenants in proportion to the number of their VMs Captures payment-proportionality »Similar to other resources like CPU, RAM etc. Desirable properties »Strategy-proofness: Allocations cannot be gamed »Symmetry: Reversing directions of flows does not change allocation 7

8 Design Space 8 High Utilization Min Bandwidth Guarantee Aggregate Proportionality Strategy- Proofness Symmetry

9 Requirement 3: High Utilization Provides incentives such that throughput is only constrained by the network capacity »Not by the inefficiency of the allocation or by disincentivizing users to send traffic Desirable properties »Work Conservation: Full utilization of bottleneck links »Independence: Independent allocation of one VM’s traffic across independent paths 9

10 Tradeoff 1 Design Space 10 High Utilization Min Bandwidth Guarantee Aggregate Proportionality Work Conservation Strategy- Proofness IndependenceSymmetry

11 Tradeoff 1: Min B/W vs. Proportionality 11 Share of Tenant A can decrease arbitrarily! B2 A1 B1 A2 Link L with Capacity C C/2 B3 C/3 2C/3

12 Tradeoff 1 Tradeoff 2 Design Space 12 High Utilization Min Bandwidth Guarantee Aggregate Proportionality Work Conservation Strategy- Proofness IndependenceSymmetry

13 Tradeoff 2: Proportionality vs. Utilization 13 A1A1 A3A3 B1B1 B3B3 A2A2 A4A4 B2B2 B4B4 Link L with Capacity C C/4 To maintain proportionality, equal amount of traffic must be moved from A1-A2 to A1-A3 => Underutilization of A1-A3

14 Per-link Proportionality Restrict to congested links only Share of a tenant on a congested link is proportional to the number of its VMs sending traffic on that link 14

15 Per Endpoint Sharing (PES) Five identical VMs (with unit weights) sharing a Link L 15 B2B2 A1A1 B1B1 A2A2 A3A3 Link L

16 Per Endpoint Sharing (PES) Resulting weights of the three flows: 16 Link L 3/2 2 To generalize, weight of a flow A-B on link L is.. WAWA W A-B = NANA NBNB WBWB + N A = 2 B2B2 A1A1 B1B1 A2A2 A3A3

17 Per Endpoint Sharing (PES) Symmetric Proportional »sum of weights of flows of a tenant on a link L = sum of weights of its VMs communicating on that link Work Conserving Independent Strategy-proof on congested links 17 WAWA W A-B = NANA NBNB WBWB + = W B-A

18 Generalized PES 18 AB α > β if L is more important to A than to B (e.g., A’s access link) WAWA W A-B = W B-A = α NANA NBNB WBWB + β+ β L Scale weight of A by α Scale weight of B by β

19 One-Sided PES (OSPES) 19 AB α = 1, β = 0 if A is closer to L α = 0, β = 1 if B is closer to L WAWA W A-B = W B-A = α NANA NBNB WBWB + β+ β L Scale weight of A by α Scale weight of B by β Per Source closer to source Per Destination closer to destination Highest B/W Guarantee *In the Hose Model

20 Comparison 20 Per Flow Per Source Static Reservation Link PESOSPES Link Proportionality   Symmetry ✓✗✓✓✓  Strategy-Proofness ✗✓✓✓✓ Utilization   Independence ✓✓✓✓✓  Work Conservation ✓✓✗✓✓ B/W Guarantee 

21 Full-bisection B/W Network 21 Tenant 1 has one-to-one communication pattern Tenant 2 has all-to-all communication pattern OSPES

22 MapReduce Workload 22 W1:W2:W3:W4:W5 = 1:2:3:4:5 OSPES

23 Summary Sharing cloud networks is all about making tradeoffs »Min b/w guarantee VS Proportionality »Proportionality VS Utilization Desired solution is not obvious »Depends on several conflicting requirements and properties »Influenced by the end goal 23

Download ppt "Sharing Cloud Networks Lucian Popa, Gautam Kumar, Mosharaf Chowdhury Arvind Krishnamurthy, Sylvia Ratnasamy, Ion Stoica UC Berkeley."

Similar presentations

Capacity of wireless ad-hoc networks By Kumar Manvendra October 31,2002.

Capacity of wireless ad-hoc networks By Kumar Manvendra October 31,2002.
Towards Predictable Datacenter Networks

Towards Predictable Datacenter Networks
Internet Protocols Steven Low CS/EE netlab.CALTECH.edu October 2004 with J. Doyle, L. Li, A. Tang, J. Wang.

Internet Protocols Steven Low CS/EE netlab.CALTECH.edu October 2004 with J. Doyle, L. Li, A. Tang, J. Wang.
1 EP2210 Fairness Lecture material: –Bertsekas, Gallager, Data networks, 6.5 –L. Massoulie, J. Roberts, Bandwidth sharing: objectives and algorithms,“

1 EP2210 Fairness Lecture material: –Bertsekas, Gallager, Data networks, 6.5 –L. Massoulie, J. Roberts, "Bandwidth sharing: objectives and algorithms,“
Resource Management §A resource can be a logical, such as a shared file, or physical, such as a CPU (a node of the distributed system). One of the functions.

Resource Management §A resource can be a logical, such as a shared file, or physical, such as a CPU (a node of the distributed system). One of the functions.
Max Flow Problem Given network N=(V,A), two nodes s,t of V, and capacities on the arcs: uij is the capacity on arc (i,j). Find non-negative flow fij for.

Max Flow Problem Given network N=(V,A), two nodes s,t of V, and capacities on the arcs: uij is the capacity on arc (i,j). Find non-negative flow fij for.
Lucian Popa Minlan Yu Steven Y. Ko UC Berkeley/ICSI Princeton Univ. Princeton Univ. Sylvia Ratnasamy Ion Stoica Intel Labs Berkeley UC Berkeley CloudPolice:

Lucian Popa Minlan Yu Steven Y. Ko UC Berkeley/ICSI Princeton Univ. Princeton Univ. Sylvia Ratnasamy Ion Stoica Intel Labs Berkeley UC Berkeley CloudPolice:
Stratos: A Network-Aware Orchestration Layer for Middleboxes in the Cloud Aditya Akella, Aaron Gember, Anand Krishnamurthy, Saul St. John University of.

Stratos: A Network-Aware Orchestration Layer for Middleboxes in the Cloud Aditya Akella, Aaron Gember, Anand Krishnamurthy, Saul St. John University of.
1 Traffic Engineering (TE). 2 Network Congestion Causes of congestion –Lack of network resources –Uneven distribution of traffic caused by current dynamic.

1 Traffic Engineering (TE). 2 Network Congestion Causes of congestion –Lack of network resources –Uneven distribution of traffic caused by current dynamic.
Cloud Control with Distributed Rate Limiting Barath Raghavan, Kashi Vishwanath, Sriram Ramabhadran, Kenneth Yocum, and Alex C. Snoeren University of California,

Cloud Control with Distributed Rate Limiting Barath Raghavan, Kashi Vishwanath, Sriram Ramabhadran, Kenneth Yocum, and Alex C. Snoeren University of California,
A Flexible Model for Resource Management in Virtual Private Networks Presenter: Huang, Rigao Kang, Yuefang.

A Flexible Model for Resource Management in Virtual Private Networks Presenter: Huang, Rigao Kang, Yuefang.
Communication Networks A Second Course Jean Walrand Department of EECS University of California at Berkeley.

Communication Networks A Second Course Jean Walrand Department of EECS University of California at Berkeley.
1 EL736 Communications Networks II: Design and Algorithms Class8: Networks with Shortest-Path Routing Yong Liu 10/31/2007.

1 EL736 Communications Networks II: Design and Algorithms Class8: Networks with Shortest-Path Routing Yong Liu 10/31/2007.
Look Who’s Talking: Discovering Dependencies between Virtual Machines Using CPU Utilization HotCloud 10 Presented by Xin.

Look Who’s Talking: Discovering Dependencies between Virtual Machines Using CPU Utilization HotCloud 10 Presented by Xin.
FairCloud: Sharing the Network in Cloud Computing Computer Communication Review(2012) Arthur : Lucian Popa Arvind Krishnamurthy Sylvia Ratnasamy Ion Stoica.

FairCloud: Sharing the Network in Cloud Computing Computer Communication Review(2012) Arthur : Lucian Popa Arvind Krishnamurthy Sylvia Ratnasamy Ion Stoica.
Reciprocal Resource Fairness: Towards Cooperative Multiple-Resource Fair Sharing in IaaS Clouds School of Computer Engineering Nanyang Technological University,

Reciprocal Resource Fairness: Towards Cooperative Multiple-Resource Fair Sharing in IaaS Clouds School of Computer Engineering Nanyang Technological University,
My Point of View about Bandwidth Sharing Bin Wang.

My Point of View about Bandwidth Sharing Bin Wang.
A Game Theoretic Approach to Provide Incentive and Service Differentiation in P2P Networks John C.S. Lui The Chinese University of Hong Kong Joint work.

A Game Theoretic Approach to Provide Incentive and Service Differentiation in P2P Networks John C.S. Lui The Chinese University of Hong Kong Joint work.
EECC694 - Shaaban #1 lec #7 Spring2000 3-28-2000 The OSI Reference Model Network Layer.

EECC694 - Shaaban #1 lec #7 Spring The OSI Reference Model Network Layer.
High Throughput Route Selection in Multi-Rate Ad Hoc Wireless Networks Dr. Baruch Awerbuch, David Holmer, and Herbert Rubens Johns Hopkins University Department.

High Throughput Route Selection in Multi-Rate Ad Hoc Wireless Networks Dr. Baruch Awerbuch, David Holmer, and Herbert Rubens Johns Hopkins University Department.

Similar presentations

Ads by Google