Simulating Energy Aware Networks in Large Scale Distributed Systems - - PowerPoint PPT Presentation

simulating energy aware networks in large scale
SMART_READER_LITE
LIVE PREVIEW

Simulating Energy Aware Networks in Large Scale Distributed Systems - - PowerPoint PPT Presentation

Simulating Energy Aware Networks in Large Scale Distributed Systems Betsegaw Lemma Amersho Supervised by: Prof. Martin Quinson, Dr. Anne-C ecile Orgerie Master Thesis Defence, June 30, 2017 Outline Recent Trends in Large-Scale Networks 1


slide-1
SLIDE 1

Simulating Energy Aware Networks in Large Scale Distributed Systems

Betsegaw Lemma Amersho

Supervised by: Prof. Martin Quinson, Dr. Anne-C´ ecile Orgerie

Master Thesis Defence, June 30, 2017

slide-2
SLIDE 2

Outline

1

Recent Trends in Large-Scale Networks

2

Energy Proportionality

3

Common Research Methods

4

Network Simulation

5

Our Approach

6

Proposed Solution

7

Validation Experiments Accuracy Validation Result Scalability Validation Result

8

Conclusion

slide-3
SLIDE 3

Recent Trends in Large-Scale Networks

Networked Devices are Increasing 17.1 billion in 2016 to 27.1 billion in 2021 IP Traffic is Growing Global IP traffic will increase threefold over the next 5 years Cloud Infrastructure Dependency is Increasing By 2020, 92% of workloads will be processed by cloud data centers Data Center are Expanding In response to the added components and the increased service demand This Expansion Led to Energy Consumption Concern

slide-4
SLIDE 4

Recent Trends in Large-Scale Networks

Networked Devices are Increasing 17.1 billion in 2016 to 27.1 billion in 2021 IP Traffic is Growing Global IP traffic will increase threefold over the next 5 years Cloud Infrastructure Dependency is Increasing By 2020, 92% of workloads will be processed by cloud data centers Data Center are Expanding In response to the added components and the increased service demand This Expansion Led to Energy Consumption Concern

slide-5
SLIDE 5

Recent Trends in Large-Scale Networks

Networked Devices are Increasing 17.1 billion in 2016 to 27.1 billion in 2021 IP Traffic is Growing Global IP traffic will increase threefold over the next 5 years Cloud Infrastructure Dependency is Increasing By 2020, 92% of workloads will be processed by cloud data centers Data Center are Expanding In response to the added components and the increased service demand This Expansion Led to Energy Consumption Concern

slide-6
SLIDE 6

Recent Trends in Large-Scale Networks

Networked Devices are Increasing 17.1 billion in 2016 to 27.1 billion in 2021 IP Traffic is Growing Global IP traffic will increase threefold over the next 5 years Cloud Infrastructure Dependency is Increasing By 2020, 92% of workloads will be processed by cloud data centers Data Center are Expanding In response to the added components and the increased service demand This Expansion Led to Energy Consumption Concern

slide-7
SLIDE 7

Energy Proportionality

Ideal Energy Proportional Network Equipment Consume no power when idle (Static Power) Consume power in proportion to their work load (Dynamic Power) Energy Inefficiency of Current Network Equipment Current Network Equipment have narrow dynamic power range

slide-8
SLIDE 8

Energy Proportionality

Ideal Energy Proportional Network Equipment Consume no power when idle (Static Power) Consume power in proportion to their work load (Dynamic Power) Energy Inefficiency of Current Network Equipment Current Network Equipment have narrow dynamic power range

slide-9
SLIDE 9

Common Research Methods

Experimenting on Real Production Environment (in vivo) Can give real picture of the problem being studied Difficult to repeat experiments Might not be available for experimentation Experimenting on Experimental Test-Bed (in vivo) Offers full control over the experiment Limited in scalablility and for testing different scenarios Experimenting using Simulation Software (in silico) Gives full control, more flexible, less expensive and less time consuming Might fail to correctly model the real situation

slide-10
SLIDE 10

Common Research Methods

Experimenting on Real Production Environment (in vivo) Can give real picture of the problem being studied Difficult to repeat experiments Might not be available for experimentation Experimenting on Experimental Test-Bed (in vivo) Offers full control over the experiment Limited in scalablility and for testing different scenarios Experimenting using Simulation Software (in silico) Gives full control, more flexible, less expensive and less time consuming Might fail to correctly model the real situation

slide-11
SLIDE 11

Common Research Methods

Experimenting on Real Production Environment (in vivo) Can give real picture of the problem being studied Difficult to repeat experiments Might not be available for experimentation Experimenting on Experimental Test-Bed (in vivo) Offers full control over the experiment Limited in scalablility and for testing different scenarios Experimenting using Simulation Software (in silico) Gives full control, more flexible, less expensive and less time consuming Might fail to correctly model the real situation

slide-12
SLIDE 12

Network Simulation

Packet Level Simulators Modeling at a packet-level (capture low-level details) Are close to the real network phenomenon being modeld Fail to scale well when simulating large-scale networks Flow Level Simulators Abstract away low-level details and use analytical equations Suitable for simulating large-scale networks (Scalable) Loss of low-level details/accuracy The Goal of this Study To investigate the level of energy estimation accuracy that can be

  • btained from flow-level energy consumption models.
slide-13
SLIDE 13

Network Simulation

Packet Level Simulators Modeling at a packet-level (capture low-level details) Are close to the real network phenomenon being modeld Fail to scale well when simulating large-scale networks Flow Level Simulators Abstract away low-level details and use analytical equations Suitable for simulating large-scale networks (Scalable) Loss of low-level details/accuracy The Goal of this Study To investigate the level of energy estimation accuracy that can be

  • btained from flow-level energy consumption models.
slide-14
SLIDE 14

Network Simulation

Packet Level Simulators Modeling at a packet-level (capture low-level details) Are close to the real network phenomenon being modeld Fail to scale well when simulating large-scale networks Flow Level Simulators Abstract away low-level details and use analytical equations Suitable for simulating large-scale networks (Scalable) Loss of low-level details/accuracy The Goal of this Study To investigate the level of energy estimation accuracy that can be

  • btained from flow-level energy consumption models.
slide-15
SLIDE 15

Our Approach

Energy Consumption Model Literature Search Model Implementation Model Validation Simulating Energy Consumption Simulating Energy Consumption M

  • d

e l V a l i d a t i

  • n

Comparing Scalability

Ⓐ Ⓑ Ⓒ Ⓓ Ⓔ Ⓕ Ⓖ

slide-16
SLIDE 16

Proposed Flow Level Model

Energy Consumption E(T) = T P(t)dt Power Consumption Ptotal = Pstatic + Pdynamic Flow-Level Model Implemented in SimGrid E(T) = T (Pstatic + Pdynamic)(t)dt where, Pdynamic = (Pbusy − Pidle) ∗ u

slide-17
SLIDE 17

Proposed Flow Level Model

Energy Consumption E(T) = T P(t)dt Power Consumption Ptotal = Pstatic + Pdynamic Flow-Level Model Implemented in SimGrid E(T) = T (Pstatic + Pdynamic)(t)dt where, Pdynamic = (Pbusy − Pidle) ∗ u

slide-18
SLIDE 18

Proposed Flow Level Model

Energy Consumption E(T) = T P(t)dt Power Consumption Ptotal = Pstatic + Pdynamic Flow-Level Model Implemented in SimGrid E(T) = T (Pstatic + Pdynamic)(t)dt where, Pdynamic = (Pbusy − Pidle) ∗ u

slide-19
SLIDE 19

Validation Experiments

Accuracy Scenarios Data Size: [10,500] MB Traffic Flow: [1,4] Network Path Length: 1 and 3 Scalability Scenarios Simulation Time and Memory Usage Traffic Flow: 2 Data Size: [50,500] MB Network Path Length: 1, 2, 4, 6, 8, & 10

slide-20
SLIDE 20

Validation Experiments

Accuracy Scenarios Data Size: [10,500] MB Traffic Flow: [1,4] Network Path Length: 1 and 3 Scalability Scenarios Simulation Time and Memory Usage Traffic Flow: 2 Data Size: [50,500] MB Network Path Length: 1, 2, 4, 6, 8, & 10

slide-21
SLIDE 21

Accuracy Validation Result

ECOFEN vs SimGrid Comparison for all scenarios the accuracy estimation error is < 0.3% Flows 2, Path-Length 1 Flows 2, Path-Length 3

slide-22
SLIDE 22

Scalability Validation Result

Simulation Time and Peak Memory Usage Comparison SimGrid is 243 to 2723 times faster than ECOFEN SimGrid is 2 to 15 times more memory efficent than ECOFEN Simulation Time Peak Memory Usage

slide-23
SLIDE 23

Conclusion

Flow-Level energy consumption models can give energy estimation with very good accuracy without lossing their scalability.