QoS-aware Energy-Efficient Algorithms for Ethernet Link Aggregates - - PowerPoint PPT Presentation

QoS-aware Energy-Efficient Algorithms for Ethernet Link Aggregates in Software-Defined Networks

Pablo Fondo Ferreiro Miguel Rodríguez Pérez Manuel Fernández Veiga September 15, 2018

1

Context

Context

Previous work on Aggregates of Energy Effjcient Ethernet Links Straightforward Solution Power ofg unused links

• Slow response time
• What about half used links?

2

EEE Links

• Formally IEEE 802.3az.
• Low Power Idle (LPI) state.
• Sleeping and waking up is not

instantaneous.

20 40 60 80 100 0.001 0.01 0.1 1 Normalized Energy Usage (%) Load EEE Link

........

Active Sleeping 𝑢s Quiet Refreshing Quiet Low Power Mode Quiet Waking up Active 𝑢w 𝑢r 𝑢r Refreshing

Figure 1: Energy-Effjcient Ethernet model. Retrieved from [1].

3

Problem statement

Goal Minimize energy consumption in bundles of EEE links leveraging SDN.

λ λ2 λ1 λ3 λ4

20 40 60 80 100 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Normalized Energy Usage (%) Load Equitable share

𝜇𝑗 = 𝜇/4

4

Problem statement

Goal Minimize energy consumption in bundles of EEE links leveraging SDN.

λ λ2 λ1 λ3 λ4

20 40 60 80 100 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Normalized Energy Usage (%) Load 1 Link Bundle Ideal Behavior

1-link bundle

𝜇𝑗 = min {𝐷, 𝜇 −

𝑗−1

∑

𝑙=1

𝜇𝑗}

5

Problem statement

Goal Minimize energy consumption in bundles of EEE links leveraging SDN.

λ λ2 λ1 λ3 λ4

20 40 60 80 100 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Normalized Energy Usage (%) Load 1 Link Bundle 2 Link Bundle Ideal Behavior

2-link bundle

𝜇𝑗 = min {𝐷, 𝜇 −

𝑗−1

∑

𝑙=1

𝜇𝑗}

5

Problem statement

Goal Minimize energy consumption in bundles of EEE links leveraging SDN.

λ λ2 λ1 λ3 λ4

20 40 60 80 100 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Normalized Energy Usage (%) Load 1 Link Bundle 2 Link Bundle 4 Link Bundle Ideal Behavior

4-link bundle

𝜇𝑗 = min {𝐷, 𝜇 −

𝑗−1

∑

𝑙=1

𝜇𝑗}

5

Problem statement

Goal Minimize energy consumption in bundles of EEE links leveraging SDN. Theoritical solution Presented in [2], provides a

• Packet level algorithm
• Assumes real time access to individual occupation of each output port

SDN Solution

• Needs flow level operation
• Cannot take real-time decisions based on queue occupation
• Will use ONOS for portability

6

SDN Algorithm

SDN Application

Main Tasks

• Flow identification
• Flow characterization
• Port allocation

7

Flow definition

Challenge Which fields of the packets will identify our flows?

• We need:
• Enough flows to distribute them along the bundle.
• Few flows to keep flow tables small.
• Flows with predictable demand.
• Two alternatives: Flow tagging vs field matching.
• We will aggregate IP flows:
• MAC flows can be insuffjcient (e.g., transit networks).
• Transport flows would be excessive.

8

Flow rate estimation

10 20 30 40 50 60 70 80 90 100 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 estimation error (Mbps) sampling period (seconds) EWMA α = 0.2 EWMA α = 0.4 EWMA α = 0.6 EWMA α = 0.8 previous

Figure 2: Average error in the estimation of the flow rate for difgerent periods.

Use rate of previous interval with sampling rate around 0.2 s

9

Port Allocation

In essence, a bin packing problem. Heuristics Greedy Corresponds to first fit decreasing. A flow level water-filling. Bounded Greedy Variation to reduce loses: Maximum usable capacity of a link: 1 − 𝑐𝑝𝑣𝑜𝑒

|𝑔𝑚𝑝𝑥𝑡|

Conservative

• Balanced distribution among needed ports.
• Safety margin to further avoid losses.
• Note: Energy consumption raises very rapidly with traffjc load.

10

Conservative Algorithm

Behavior

• Determines the number of needed links
• Distributed flows evenly among the links

Basis EEE energy usage rises rapidly with load.

20 40 60 80 100 5 10 15 20 Normalized Energy Usage (%) Incoming traffic load (Gb/s) 2-bundle link Ideal share Conservative share

11

Conservative Algorithm

Behavior

• Determines the number of needed links
• Distributed flows evenly among the links

Basis EEE energy usage rises rapidly with load.

20 40 60 80 100 5 10 15 20 25 30 35 40 Normalized Energy Usage (%) Incoming traffic load (Gb/s) 4-bundle link Ideal share Conservative share

20 40 60 80 100 5 10 15 20 Normalized Energy Usage (%) Incoming traffic load (Gb/s) 2-bundle link Ideal share Conservative share

11

Conservative Algorithm

Behavior

• Determines the number of needed links
• Distributed flows evenly among the links

Basis EEE energy usage rises rapidly with load.

20 40 60 80 100 10 20 30 40 50 60 70 80 Normalized Energy Usage (%) Incoming traffic load (Gb/s) 8-bundle link Ideal share Conservative share

20 40 60 80 100 5 10 15 20 Normalized Energy Usage (%) Incoming traffic load (Gb/s) 2-bundle link Ideal share Conservative share 20 40 60 80 100 5 10 15 20 25 30 35 40 Normalized Energy Usage (%) Incoming traffic load (Gb/s) 4-bundle link Ideal share Conservative share

11

Experimental setup

• Topology: Two switches connected by 5 EEE interfaces 10 GBASE-T.
• We have used real traffjc traces retrieved from CAIDA [3].
• Baseline: Equitable algorithm.
• Metrics:
• Energy consumption
• Packet losses
• Packet delay

12

Results: Energy consumption

78 80 82 84 86 88 90 92 94 96 98 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 real energy consumption (%) sampling period (seconds) Greedy Bounded-Greedy Conservative Equitable

(a) 32.5 Gbit/s trace.

20 40 60 80 100 6.5 13.0 19.5 26.0 32.5 energy consumption (%) rate (Gbps) Greedy Bounded-Greedy Conservative Equitable

(b) sampling period = 0.5 seconds. Figure 3: Normalized energy consumption (bufger = 10000 packets).

• Theoretical bound for the consumption of the 32.5 Gbit/s: 78.5 %.

13

Results: Packet losses

2 4 6 8 10 12 14 16 18 10 100 1000 10000 100000 packet loss (%) buffer size(packets) Greedy Bounded-Greedy Conservative Equitable

(a) 32.5 Gbit/s trace.

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 6.5 13.0 19.5 26.0 32.5 packet loss (%) rate (Gbps) Greedy Bounded-Greedy Conservative Equitable

(b) bufger = 10000 packets. Figure 4: Packet loss percentage (sampling period = 0.5 seconds).

14

Results: Packet delay

500 1000 1500 2000 2500 3000 3500 4000 4500 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 average delay (microseconds) sampling period (seconds) Greedy Bounded-Greedy Conservative Equitable

(a) 32.5 Gbit/s trace.

1 10 100 1000 10000 6.5 13.0 19.5 26.0 32.5 average delay (microseconds) rate (Gbps) Greedy B-Greedy Conservative Equitable

(b) sampling period = 0.5 seconds. Figure 5: Average per packet delay (bufger = 10000 packets).

15

QoS-aware algorithms

Problem statement

Goal Provide low-latency service while reducing energy consumption.

• The previous algorithms manage to reduce energy consumption.
• However, they increase the delay of the packets.
• We consider now the QoS latency requirements of the flows.
• Two types of traffjc:
• Best-efgort.
• Low-latency.
• Modifications to the previous algorithms.

16

Solutions

Spare Port

• 1. Apply energy-effjcient algorithm to

best-efgort flows.

• 2. Low-latency flows are allocated to the most

empty port.

Port 1 Port 2 Port 3 Port 4 best−effort flows low−latency flows

Figure 6: Spare Port.

Two Queues

• 1. Apply energy-effjcient algorithm to all the

flows.

• 2. Low-latency flows are allocated to the

high-priority queue of the assigned ports.

Port 1 Port 2 Port 3 Port 4

best−effort flows low−latency flows

Figure 7: Two Queues.

17

Simulations

• Same topology: 5-link bundle of 10 GBASE-T EEE interfaces.
• Real traces for best-efgort traffjc.
• Synthetic traffjc for low-latency packets.
• Baseline: Conservative algorithm.
• Parameters:
• Bufger = 10 000 packets.
• Sampling period = 0.5 seconds.
• Metrics:
• Delay of low-latency packets.
• Delay of best-efgort packets.
• Energy consumption.

18

Results: Delay of low-latency packets

1 10 100 1000 0.1 1 10 100 1000 average delay (microseconds) low-latency rate (Mbps) Conservative Spare Port T wo Queues

(a) 32.5 Gbit/s trace.

1 10 100 1000 0.1 1 10 100 1000 average delay (microseconds) low-latency rate (Mbps) Conservative Spare port T wo queues

(b) 45.5 Gbit/s trace. Figure 8: Average delay of low-latency packets.

19

Results: Delay of best-effort packets

200 300 400 500 600 700 800 900 1000 1100 0.1 1 10 100 1000 average delay (microseconds) low-latency rate (Mbps) Conservative Spare Port T wo Queues

Figure 9: Average delay of best-efgort packets (32.5 Gbit/s trace).

20

Results: Energy consumption

20 40 60 80 100 0.1 1.0 10.0 100.0 1000.0 energy consumption (%) low-latency rate (Mbps) Conservative Spare Port T wo Queues

Figure 10: Normalized energy consumption (32.5 Gbit/s trace).

21

Conclusions

Conclusions

• SDN can be leveraged to implement energy saving algorithms
• Results match theoretical model
• Provided low latency service based on QoS requirements

Future work

• Reuse edge allocations for inner switches.
• Reduce control plane traffjc (e.g., minimize flow re-allocations).

22

Thank you for listening!

Email: miguel@det.uvigo.es

References i

• S. Herrería-Alonso, M. Rodríguez-Pérez, M. Fernández-Veiga, and C. López-García, “How

effjcient is energy-effjcient ethernet?” in Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), 2011 3rd International Congress on. IEEE, 2011, pp. 1–7.

• M. Rodríguez Pérez, M. Fernández Veiga, S. Herrería Alonso, M. Hmila, and
• C. López García, “Optimum Traffjc Allocation in Bundled Energy-Effjcient Ethernet

Links,” IEEE Syst. J., vol. 12, no. 1, pp. 593–603, Mar. 2018. “The CAIDA UCSD Anonymized Internet Traces 2016 — 2016/04/06 13:03:00 UTC.” [Online]. Available: http://www.caida.org/data/passive/passive_2016_dataset.xml