Can Cloud Computing Can Cloud Computing and Optical Networking and - PowerPoint PPT Presentation

Can Cloud Computing Can Cloud Computing and Optical Networking and Optical Networking S S Save the Planet? Save the Planet? th th Pl Pl t? t? Rod Tucker, Jayant Baliga, Robert Ayre, Kerry Hinton ARC Special Research Centre for Ultra-Broadband Information Networks (CUBIN) University of Melbourne r.tucker@ee.unimelb.edu.au

Cloud Computing Cloud Computing What do the following have in common? • Happiness • Love • Marketing Hype • Recession • Cloud Computing p g They all mean different things to different people

Summary Summary Cloud Computing as a Logistics Problem Cloud Computing as a Logistics Problem Key elements of logistics • Processing • Storage Storage Energy costs Energy costs • Transport

Cloud Computing Energy Model Cloud Computing Energy Model Computer, Enterprise Data Center Storage Storage Servers, PON Access Network PON Access Network Processors Fiber OLT ONU Splitter Core Network Core Router Broadband Network Total Energy Total Energy Gateways G OXC Consumption = Edge Routers Energy/bit used in • Storage Ethernet Switch • Transport • Processing Servers, M t Metro/Edge Network /Ed N t k Processors X Total number of bits Storage Cloud Public Data Center Public Data Center

Network Energy Model Network Energy Model 1000 2009 2009 100 nJ) per bit (n 10 Energy 1 0.1 0 1 0.01 0 01 Optical PoS Ethernet Core PON Server WDM Amp Switch Router ONU Tx/Rx (10 Mb/s)

Data Centre Energy Model Data Centre Energy Model Edge Router Servers 26 nJ/b 26 nJ/b 430 nJ/b 430 nJ/b Storage 8 pW/b p [29 nJ/hr/b] LAN – 2 stages of aggregation Watts not Joules using Ethernet switches g Baliga et al., OThT6

User Hardware Energy Model User Hardware Energy Model • End user computer: – High Energy Use Desktop 140 Watts • Eg. Intel Quad-core, or older P4, mid-range graphics – Modern mid-range Desktop 70 Watts • Eg. Intel Core Duo 2.5 GHz, 2 GB RAM, mid-range graphics – Low-end Desktop/Thin Client 18 Watts • Eg. Intel Atom 1.6 GHz, 2 GB RAM, chipset-based graphics g , , p g p – Ignore monitor (LCD/CRT) energy consumption • This will be similar in all our comparisons

Outsource to Cloud or Oversize Computer? Outsource to Cloud or Oversize Computer? Should I use a low-end Computer & Outsource Jobs Service Computing Outsource jobs Center Resource Low energy Access Metro/Edge Network consumption. consumption. Network - OR - OR Use a high-end computer that g p can handle any of my jobs But high energy consumption even for light jobs for light jobs

Consider Three Scenarios Consider Three Scenarios 1. Software-as-a-Service – Stored on user’s computer with updates downloaded regularly • Eg current anti-virus software model extended to mainstream applications, MS Update (weekly patch) model 2. Service Bureau 2 S i B – Most tasks done on lower end user machine, outsource the “big” jobs 3. Computing-as-a-Service – Hosted and run on provider’s computer “farm” with data initially Hosted and run on provider s computer farm with data initially uploaded from user • User data may be hosted or locally stored • Like enterprise “Thin Client” model Like enterprise Thin Client model Can cloud computing save energy ?

1. Software- 1. Software -as as- -a a- -Service Service Service I want to use Data Center the Word Center Processor Processor Mid-range Computer Access Metro/Edge Network Network Software as a Service Software-as-a-Service • Software maintenance

1. Software 1. Software- -as as- -a a- -Service Service Here is the latest Service version or patch Data Center Center Mid-range Computer Access Metro/Edge Network Network Software as a Service Software-as-a-Service • Software maintenance

2. 2. Service Bureau Service Bureau Here is my raw video footage Service and edit Computing Center commands Resource Mid-range Computer Access Metro/Edge Network Network Service Bureau Service Bureau – Hardware & Software Hardware & Software • Service Bureau for “Big” Jobs

2. 2. Service Bureau Service Bureau Service Here is your Computing Center DVD Image Resource Mid-range Computer Access Metro/Edge Network Network Service Bureau Service Bureau – Hardware & Software Hardware & Software • Service Bureau for “Big” Jobs

3. Computing 3. Computing- -as as- -a a- -Service Service Service Enterprise Center Server Access Here is the new Metro/Edge Network data for my Network spreadsheet p Computing (Hardware and Software-as- Computing (Hardware and Software-as- a-Service) Low-end or “Thin Client” • Centralised Storage and Processing C Computer t

3. 3. Computing Computing- -as as- -a a- -Service Service Service Enterprise Center Server Access Here is the Metro/Edge Network output file to Network print or e-mail p Computing (Hardware and Software-as- Computing (Hardware and Software-as- a-Service) Low-end or “Thin Client” • Centralised Storage and Processing C Computer t

1. Software 1. Software- -as as- -a a- -Service Example Service Example Store on Local Disk or in the Cloud? Store on Local Disk or in the Cloud? • Energy impact of the gy p “Software as a Service” & Service Data Centre Centre “Storage as a Service” models • Compare the energy cost to Access Metro store & transport 50 MB Network Network blocks to the user, bl k t th cf. store on local drive • U Users require 1GB and 5GB “cloud” storage i 1GB d 5GB “ l d” t

1. Software- 1. Software -as as- -a a- -Service Example Service Example Store on Local Disk or in the Cloud? Store on Local Disk or in the Cloud? Storage of application & data “in the cloud” compared with storing on a local disk. 50 MBytes per download. 50 MBytes per download. Modern laptop style HDD 20% read/write and 80% idle. Modern laptop-style HDD 20% read/write and 80% idle. 100 ) ption (W) Transport energy dominates 10 Consump Local HDD L l HDD 1 Power C 0.1 5 GB Cloud Storage 1 GB Cloud Storage 1 GB Cloud Storage 0.01 0.01 0.1 1 10 100 D Downloads per Hour l d H

2. Service Bureau Example 2. Service Bureau Example Computer used for 20 hrs/week, plus some video encoding of ½ hour videos 180 Older Computer on, W 160 Transport Transport nsumptio 140 energy dominates 120 Mid-range Computer Power Con 100 100 80 60 60 Average P 40 Low-end Computer & outsourcing 20 A 0 0 20 40 60 80 100 120 Number of Encodings per Week Number of Encodings per Week

3. Computing- 3. Computing -as as- -a a- -Service Example Service Example Thin Client or Small PC in Office? Thin Client or Small PC in Office? • • Office situation Office situation Service Enterprise – non-intensive tasks, Center Server • Should this office deploy: Should this office deploy: – a low-end machine per user OR “dumb, thin clients”? Access Metro/Edge Network Network Commonly, thin client would receive and display • HTML formatted data from the server • Thi Thin client capability may be limited to displaying li t bilit b li it d t di l i bitmaps developed from its server

3. Computing 3. Computing- -as as- -a a- -Service Example Service Example Thin Client compared to Small PC in Office Thin Client compared to Small PC in Office nt (W) Thin Clien 7.5 “local” server 5 g using T 2.5 Distant Server wer Saving 0 -2.5 Pow 1% 2% 5% 10% -5 % of screen 0 1 2 3 4 5 area updated area updated Average Bitrate for Screen Refresh Mb/s per second Thin client solution “green” only when the server is local Thin client solution green only when the server is local

Summary Summary • Cloud Storage and Cloud Computing can provide energy savings – More so for less-demanding users – For handling the occasional intensive task • But... – Advantages may be lost when data transfer is large – This is where optical networking needs to come to the rescue • Cloud Computing may not save the planet…. Cl d C ti t th l t – But if used appropriately, it could help to mitigate the growth in energy consumption in our industry g gy p y – A classical Logistics problem