Greening the Internet with Nano Data Centers V. Valancius (Georgia - - PowerPoint PPT Presentation

Greening the Internet with Nano Data Centers

• V. Valancius (Georgia Institute of Technology), N. Laoutaris

(Telefonica Research), L. Massoulie, C. Diot (Thomson), P. Rodriguez (Telefonica Research)

Presented by Sean Barker

2

Gateway vs. Set-top-box

Internet

Home gateway

• Internet connection
• Traffjc management

Set-top-boxes

• Content processing

3

Gateway as new center of media experience

The Internet

Content & Control Servers

Internet services Extended home network Home network

4

The current model: Data centers

The Internet

Data Centers . . . DSLAM Home boxes

5

Limitations

 Expensive

– High capital investment – Customer generally pays per byte

 Location constraints in order to be “central”  Requires a lot of redundancy to be robust

– Electricity shortage – Content availability

 Power, power, power  New service deployment is slow

– ISPs not encouraged to take risks, nor to deploy new services

6

The nano data center

 Take advantage of always-on gateways  Add memory and stronger CPU to home gateways  Push content to gateways when bandwidth is cheap  Manage millions of gateways as a logical ‘single

server’ using P2P infrastructure

7

The nano data center model

The Internet

Video server Control server . . . DSLAM Home boxes

8

The nano data center

 PROS

–

Multiple applications can take advantage of the model (VoD, gaming, catchTV, UGC)

–

ISP friendly

–

Reduces traffic volumes and variability on backbones.

–

Highly scalable and robust by design

–

Cheap for ISPs

–

Flexible for users

–

Localized & personalized services

 CONS

–

Uplink bandwidth often limited

–

Millions of boxes to manage using P2P

–

Cost of gateway

–

Incentive?

–

Privacy?

–

Always on?

9

Gateway uptime (*)

(*) Courtesy of Krishna Gummadi

More than 60% of gateways are up more than 80% of the time

10

Push phase

Video content server Control server . . . DSLAM RHGs

11

Pull phase

Control server . . . DSLAM RHGs Video content server

12

Placement strategy

… … 1 Data window

…

2 2 1 M

… …

2 M 1 M 3 3 3

… … Movie

11… 1 22 2 33 3 M M…M 1 2 2 1 K

…

2 K 1 K 3 3 3 …

… … … Erasure codes (e.g., LDPC) gateways

13

Placement strategy

 Replicate content according to popularity  Popular content served by gateways

– slack bandwidth from original content servers

 Number of replicas determined by solving

• ptimization problem

– Constraints on available upload and storage, number of

clients, request rates, etc.

14

Popularity aware placement

 Partition content into hot / warm / cold categories

• Hot: replicate on all gateways
• Warm: use code-based placement
• Cold: no proactive placement (stays on servers)

hot warm cold b/w memory popularity movies

15

Where should NaDa save energy?

The Internet

Data Centers . . . DSLAM Home boxes

Data Centers: capital cost, redundancy, cooling Internet: less hops, less bandwidth Free ride on always-on Gateways

16

Energy issues

 Variables

–

Network topology

–

Hardware power consumption

–

Placement algorithms

–

Content popularity

–

User behavior

 Data available

–

DSL gateways and VoD servers power (Thomson)

–

Routers power (Cisco data)

–

Telefonica Spain and Peru network topologies

–

Imagenio VoD platform (Telefonica Spain)

–

Telefonica IPTV

–

Netflix movie popularity

αs = 4%

17

VoD server power

αg = 1%

18

Gateway power

19

When would NaDa not work ?

# bytes transferred power Server Gateways

Baseline power

20

When does NaDa work ?

# bytes transferred power Server Gateway

Baseline power

s N

21

Trace driven simulations

 Traces from

– Netflix, IPTV (Telefonica), YouTube

 Content popularity from Netflix  Topologies and workload from Telefonica  Power numbers from Thomson’s gateway and IPTV

servers

 Popularity aware placement

22

Simulation parameters

Gateway Storage 100MB-10GB Gateway Upstream 0.1-2Mbps Content characteristics from data set Users 10k-30k Content window 10s-120s Replicas for warm content 1 (20s windows) Simulation duration 1 day - 86400 s Router energy/bit 150 10−9 Server energy/bit 40 10−9 Gateway energy/bit 18 10−9 Power Usage Effectiveness (PUE) 1.7 Home electricity cost factor 1.1 Hops to server 4 Hops to peer 2

23

Total energy use (YouTube)

10k 20k 30k 40k 50k 60k

24

Gateway storage

2500 3000 3500 4000 4500 5000 500 1,000 1,500 2,000 2,500 3,000

25

Number of users

2500 3000 3500 4000 4500 5000 5500 15,000 20,000 25,000 30,000

Number of users

26

Upstream bandwidth

3000 3500 4000 4500 5000 0.2 0.4 0.6 0.8 1.0 1.2

Video streaming > uplink rate Video streaming < uplink rate

27

Conclusions

 Free-riding on existing infrastructure can significantly

reduce load on conventional servers

 Simulations demonstrated energy savings ranging

from 20% to 60% versus data centers

 Gateways can accomplish this with only modest

resources (a few GB of storage, limited upload)

28

Questions?

 Potential QoS issues moving control from content

providers (YouTube) to ISPs

 Effects of consumer line overprovisioning  Security of content serving from home gateways