cloud computing, virtualisation and the future
DESCRIPTION
Given at the International Summer School of Grid Computing 09 at Sophia Antipolis, France, in July 17, 2009TRANSCRIPT
AakeEdlund,PhDKTH,Sweden
Bal1cGrid(ProjectDirector)
Bal1cGridInnova1onLabandBal1cCloud(Ini1ator,Manager)
PawnPromo1on(CEO),Numeri(Co‐founder)
Cloudcompu0ng,Virtualisa0on&theFuture
Cloudcompu0ng,Virtualisa0on&theFuture
Paradigmchange
“Cloudcompu1ngistheabilitytomigratethecomputa1onthatusedtohappenattheedgesintothenetwork..”
RandyHKatz,BerkeleyRADLabs Seeberkeleyclouds.blogspot.com
July17,2009 2ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Paradigmchange
“Free up the relationship between software and hardware and
a whole new industry is born”
AlanWilliamson www.aw20.co.uk
July17,2009 3ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
CloudvsGrid
Google Trends, 1st of July 2009 July17,2009 4ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
CloudCompu0ng–shortversion
Cloud = Virtualization + Automation 1. The cloud is IT infrastructure as a service. 2. The IT infrastructure is delivered as virtual machines. 3. Automation moves those VMs around, thereby providing the delivery mechanism for the service.
Rachel Chalmers, 451 group, October 20th, 2008 July17,2009 5ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Virtualisa0on‐abstrac0onofcomputerresources
“Notphysicallyexis1ngassuchbutmadebyso^waretoappeartodoso”
July17,2009 6ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
PlaBormvirtualiza0on–Separatesanopera1ngsystemfromtheunderlyingplaaormresources
Encapsula0on‐Thehidingofresourcecomplexitybythecrea1onofasimplifiedinterface
Virtualisa0on‐abstrac0onofcomputerresources
July17,2009 7ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Storagevirtualiza0on‐Theprocessofcompletelyabstrac1nglogicalstoragefromphysicalstorage
Networkvirtualiza0on‐Crea1onofavirtualisednetworkaddressingspacewithinoracrossnetworksubnets
Virtualisa0on‐abstrac0onofcomputerresources
July17,2009 8ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Computerclusters,gridcompu1ng,andcloudcompu1ng‐thecombina1onofmul1plediscretecomputersintolargermetacomputers
Applica0onvirtualiza0on‐Thehos1ngofindividualapplica1onsonalienhardware/so^ware
Virtualisa0on‐abstrac0onofcomputerresources
July17,2009 9ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Virtualisa0on–why?
Hardwareisunderu1lized–Moore’s lawDatacentersrunoutofspaceEconomyandEnvironment–EnergycostexplosionSystemadministra1oncost…
July17,2009 10ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Servervirtualisa0onisgeHngcri0cal–we’rerunningoutofroominourdatacenters
July17,2009 11ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Threemainstrategies
OSvirtualisa1onHardwareemula1onParavirtualisa1on
July17,2009 12ISSGC09‐SophiaAn1polis
OSvirtualisa0on
July17,2009 13ISSGC09‐SophiaAn1polis
Virtualenvironment1
Applica1onso^ware
Systemso^ware
Systemlibraries
Virtualenvironment2
Applica1onso^ware
Systemso^ware
Systemlibraries
Virtualenvironment3
Applica1onso^ware
Systemso^ware
Systemlibraries
Hardware
Network
OS|ContainerVirtualisa1onLayer
HostOpera1ngSystem
Installedandrunsontopofanexis1nghostOS,andprovidesasetoflibrariesthatapplica1onsinteractwith.
Efficient,imposesliileoverhead,manycontainersononepieceofHWpossible.
OnlyoneOS,sameashostOS.
GoodforlargenumberofhomogeneousOS,e.g.webhos1ng.
July17,2009 14ISSGC09‐SophiaAn1polis
x86Architecture
Applica1on
OS
Applica1on
OS
CPU,Memory,NIC,Disk
CPU,Memory,NIC,Disk
VMwareVirtualisa1onLayer
OSinstalledonavirtualmachinethatemulatesthehardwarethattheOSusuallyinteractwith.VVMcoordinatesaccessbetweenguestVMsandtheactualunderlyinghardware–inrun1me.
UnmodifiedguestOSes.SupportsdissimilarOSes.
Performancepenalty.
Devicedriverissues/inflexibility:couldlimittheusability.
Hardwareemula0on
Paravirtualisa0on
July17,2009 15ISSGC09‐SophiaAn1polis
Hardware
XenHypervisor
XenHypercallAPI
XenToolStack
DeviceDrivers
GuestOS
GuestOS
Does not create an entire VM to host the guest OS, rather enables the guest OS to interact directly with the hypervisor.
Less performance overhead.
Uses the device drivers contained in one of the guest operating systems – i.e. not limited to the drivers (as in hardware emulation) contained in the virtualisation software.
Cloudcompu0ng,Virtualisa0on&theFuture
CloudCompu0ng–shortversion
Cloud = Virtualization + Automation 1. The cloud is IT infrastructure as a service. 2. The IT infrastructure is delivered as virtual machines. 3. Automation moves those VMs around, thereby providing the delivery mechanism for the service.
Rachel Chalmers, 451 group, October 20th, 2008
Again:
July17,2009 16ISSGC09‐SophiaAn1polis
“Cloud Computing refers to both the applications delivered as services over the Internet and the hardware and systems software in the datacenters that provide those services…
The datacenter hardware and software is what we will call a Cloud…
Cloud computing has the following characteristics
1. The illusion of infinite computing resources..
2. The elimination of an up-front commitment by Cloud users..
3. The ability to pay for use … as needed…”
UC Berkeley RAD Labs July17,2009 17ISSGC09‐SophiaAn1polis
CloudCompu0ng–longerversion
Cloudcompu0ng–oncemore…
1. The illusion of infinite computing resources..
2. The elimination of an up-front commitment by Cloud users..
3. The ability to pay for use … as needed…”
July17,2009 18ISSGC09‐SophiaAn1polis
a) Signup for an AWS account
Use your existing Amazon.com account if you want b) Register a credit card
Billed on the 1st of every month for usage prior
c) http://aws.amazon.com/ec2
SpectrumofClouds
• Instruction Set VM (Amazon EC2) • Bytecode VM (Microsoft Azure) • Framework VM (Google App Engine)
EC2 Azure AppEngine
Lower-level, Less management
Higher-level, More management
Automatic scalability and failover July17,2009 19ISSGC09‐SophiaAn1polis
Twoviews‐ofthesamething
On the right, Sam Johnston’s 6 layer Cloud Computing Stack http://samj.net/2008/09/taxonomy-6-layer-cloud-computing-stack.html July17,2009 20ISSGC09‐SophiaAn1polis
‐Provisioningofcompu0ngresources:CPU,Memory,Processing‐Basicallyan“Opera0ngSystem”ondemand‐Usuallybilledonaper‐hourusagemodel
PlayersinthisspaceAmazonEC2,Flexiscale,GoGrid,Joyent,AppNexeusManagementproviders:RightScale,Elas0cServer
Opera0onalDifferences‐WildlydifferentCPU/Memoryspecs‐Opera0ngSystemdifferences‐Difficult/Impossibletomovewholeimagesbetweenproviders
Layer1:CloudInfrastructure
July17,2009 21ISSGC09‐SophiaAn1polis
- Provisioning of data storage: Either file/object based or Database like functionality. - Billed on bandwidth and storage consumed
Players in the space: Amazon S3, Nirvanix, Mosso, Amazon’s SimpleDB, Google’s BigTable, Azure Storage Management Providers: Jungle Disk, Elephant Disk, PutPlace.com
Issues - Different types of data storage models - Limitations on the size of individual data units - Different billing models makes it hard to do a straight comparison - Access to the data generally uses non-standard query syntax - No common API - Performance issues
Layer2:CloudStorage
July17,2009 22ISSGC09‐SophiaAn1polis
Provides a complete software stack - An IDE for the cloud Takes care of: Runtimes, Load balancing, Resource provisioning
Players in the space - Google App Engine - Python (initially, now also Java) - Force.com (SalesForce) - Microsoft’s Azure - .NET - Heroku.com (RubyOnRails)
Issues Different languages -- Most platforms are unary Different operational philosophies - Google App Engine for example doesn’t permit files
Lots of limitations in terms of deployment Completely reliant on the provider for complete uptime and operation Widely different billing models
Layer3:CloudPlaBorm
July17,2009 23ISSGC09‐SophiaAn1polis
- Applications that are completely ‘online’ - Operate on data that is stored in the ‘cloud’ or ‘ether’ - No client software generally required - Billing: Ad. Revenue, Premium Services
Players in this space - Google Apps - Gmail / Google Docs - Apple’s MobileMe - Microsoft’s Live - Hotmail, Live Spaces - SalesForce.com
Issues - Near on impossible to move between providers - GMail to Hotmail requires major disruption - End user focused o Consumer side of cloud computing
- Completely reliant on the provider for complete uptime and operation
Layer4:CloudApplica0ons
July17,2009 24ISSGC09‐SophiaAn1polis
- Provides services to which other applications can utilise - Specific to vertical markets where most “Web2.0” standards live - Usually free for non-commercial use
Players in the space - Google/Yahoo Maps - PayPal / Google Checkout - Google / Yahoo WebServices - Amazon Merchant Services - Amazon Simple Queue Service
Issues/Comments - Some “Web2.0” services have attempted a standardization path - Most however are complete vendor lock-in - Mashup applications utilize Cloud Service
Layer5:CloudServices
July17,2009 25ISSGC09‐SophiaAn1polis
- Accessing the cloud - Clients utilize standard access protocols o XML o JSON o REST / SOAP
Browsers o FireFox / IE / Chrome Mobile clients o Google Android / Symbian / iPhone / J2ME Desktop Apps o Google Gears / Adobe AIR / Microsoft Azure
Layer6:CloudClients
July17,2009 26ISSGC09‐SophiaAn1polis
Infrastructure
Storage
Platform
Applications
Services
Clients
Increasing level of
complexity and
vendor lock-in
ComplexityandVendorlock‐in
July17,2009 27ISSGC09‐SophiaAn1polis
Vendorlock‐in,ortherisk/benefitdance
There are alternatives to Amazon, and the list is growing
Many pricing models are confusing and unclear, and hard to compare
Many pricing models will change (and are already)
Important to get the big picture of what you need, today and later on
Vendor locking is definitely a problem to address. Before you jump into it. But this is always true
July17,2009 28ISSGC09‐SophiaAn1polis
CloudStorageCalculator
29
hip://www.aw20.co.uk/tools/storagecosts.cfm
July17,2009 ISSGC09‐SophiaAn1polis
Butisthatthefullstory?
• Likeanygoodmodelitisonlyvalidforabroadrangeofservices
• Manyservicesspandifferentlayersatonce– Forexample
• Amazon/Google/Microso^• Facebook/MySpace• YouTube/Flickr• Ebay
...makingstandardiza1onevenharder30July17,2009 ISSGC09‐SophiaAn1polis
Whynow?Experiencewithverylargedatacenters
Unprecedentedeconomiesofscale
Resource Cost in
Medium DC ≈ 1000 servers
Cost in Very Large DC ≈ 50,000 servers
Ratio
Network $95 / Mbps / month $13 / Mbps / month 7.1x Storage $2.20 / GB / month $0.40 / GB / month 5.7x Administration ≈140 servers/admin >1000 servers/admin 7.1x
Price per KWH Where Possible Reasons Why
3.6¢ Idaho Hydroelectric power; not sent long distance 10.0¢ California Electricity transmitted long distance over the
grid; limited transmission lines in Bay Area; no coal fired electricity allowed in California.
18.0¢ Hawaii Must ship fuel to generate electricity July17,2009 31ISSGC09‐SophiaAn1polis
– PervasivebroadbandInternet– Fastx86virtualiza0onVirtualiza1onmakesitallpossibleVirtualiza1onisnowbuilt‐intomicroprocessors
o Intel’sVirtualiza1on(IntelVT)o Hardwarebased,veryfast
– Pay‐as‐you‐gobillingmodel– Standardsogwarestack
July17,2009 32ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Why now? Technology is ready
Whynow?Businessdrive–cloudusers
• Flexibility• Eco‐efficiency• Creditcrunchbusinessimpera0ves
‐ CapExtoOpEx–pay‐as‐you‐go‐ Fixedcosttovariablecost
• Improved0metomarket
July17,2009 33ISSGC09‐SophiaAn1polis
Unused resources
Paybyuseinsteadofprovisioningforpeak
Static data center Data center in the cloud
Demand
Capacity
Time
Res
ourc
es
Demand
Capacity
Time R
esou
rces
July17,2009 34ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Heavy penalty for under-provisioning
Lost revenue
Lost users
Res
ourc
es
Demand
Capacity
Time (days) 1 2 3
Res
ourc
es
Demand
Capacity
Time (days) 1 2 3
Res
ourc
es
Demand
Capacity
Time (days) 1 2 3
July17,2009 35ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Trends:Howiscloudcompu0ngbeingused?
July17,2009 36ISSGC09‐SophiaAn1polis
Trends:Cloudproviders
July17,2009 37ISSGC09‐SophiaAn1polis
Thedevilisinthedetails…
Your data in the cloud
July17,2009 38ISSGC09‐SophiaAn1polis
Challenge Opportunity Availability Multiple providers & DCs
Data lock-in Standardization
Data Confidentiality and Auditability
Encryption, VLANs, Firewalls; Geographical Data Storage
July17,2009 39ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Adop0onChallenges
July17,2009 40ISSGC09‐SophiaAn1polis
GrowthChallenges
Challenge Opportunity Data transfer bottlenecks FedEx-ing disks, Data Backup/
Archival Performance unpredictability
Improved VM support, flash memory, scheduling VMs
Scalable storage Invent scalable store Bugs in large distributed systems
Invent Debugger that relies on Distributed VMs
Scaling quickly Invent Auto-Scaler that relies on ML; Snapshots
Challenge Opportunity Reputation Fate Sharing Offer reputation-guarding
services like those for email Software Licensing Pay-for-use licenses; Bulk
use sales
July17,2009 41ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
PolicyandBusinessChallenges
Startupsopportuni0es
• Startupsandprototyping• One‐offtasks• Researchatscale• CloudKillerApps:Mobileandwebapplica0ons
….moreaboutthisinthe‘Future’sec0on
July17,2009 42ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Bal0cCloudBringingcloudcompu.ngtotheBal.cStatesandBelarus
Bal1cGrid
Plannedac1vi1es
SA1‐3,NA1‐4,JRA
BGi‐Bal1cGridInnova1onLab
Bal1cCloud Courses SMEconnec1vity
BC
July17,2009 43ISSGC09‐SophiaAn1polis
Opensource“cloudinabox”
Our (main) choice:
- Great team! Rich Wolski, UCSB. - Integrated with - Integrated with
44
BC
July17,2009 ISSGC09‐SophiaAn1polis
Cloudpilot–spreading
Bal0cCloud+SE+NO+DK+FIN+Iceland+…
45
July17,2009 ISSGC09‐SophiaAn1polis
Almost‐there‐first‐summaryCloud computing is a new way of using existing technology, driven by business value for the whole chain, from the providers to their users and their
user’s users.
At this point in time, cloud computing is best used in startups and prototyping…
July17,2009 46ISSGC09‐SophiaAn1polis
from http://geekandpoke.typepad.com July17,2009 47ISSGC09‐SophiaAn1polis
• See more cloud computing being used by startups and in-house quick prototyping
• See more cloud computing being used in academia – esp. HPC flavours
• See more open source alternatives, and alternative cloud providers – also in academia (BalticCloud, SweCloud, NordicCloud)
• Improved solutions addressing the dangers in cloud computing. • Low expectations on standard APIs (takes time, not sure about the
interest from industry, looks like for grids). I.e. industry will not help here, this is the competition epicenter.
• Hype, problems, business issues: Successful usage will prove the value. If it is a paradigm shift, it will be clear this year.
• Plenty to opportunities for everyone! July17,2009 48ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Expecta0onson2009
Cloudcompu0ng,Virtualisa0on&theFuture
Startupscanberunsocheaplynow(withopen‐sourceso^ware,cloudcompu0ng,andvirtualteamsspreadacrosstheWeb)thatmanymorecanachieveprofitabilitywithoutanyVCcash.
PaulGraham,Ycombinator
Thestartupworldischanging
July17,2009 49ISSGC09‐SophiaAn1polis
“ ”
Cloudcompu0ng,Virtualisa0on&theFuture
ImaginewhatitwoulddototheVCbusinessifthenexthotcompanydidn’ttakeVCatall.Thelessventurecapitalthereisfornewstartups,thefasterthedecouplingwillbegin
PaulGraham,Ycombinator
Thestartupworldischanging
July17,2009 50ISSGC09‐SophiaAn1polis
“ ”
Cloudcompu0ng,Virtualisa0on&theFuture
RethinkVCs
IPOmarketisdead&M&Avalua1onsaredepressed
Fundamentalproblem:ITandmarke1nghasbecamesocheapthatVCsmaynotbeneededanymoreAngelstotakecareofearlierstage,VCsmovebackto
laterstagee.g.AllanMar1nson,MTVP
July17,2009 51ISSGC09‐SophiaAn1polis
“
”
Cloudcompu0ng,Virtualisa0on&theFuture
Cloudcompu0ng+Microstartups=Truelove!
Bootstrapping–morethaneverQuickstart–reallyquick!Newdirec1ons,prototyping,customerprojectsLoweryourowninternalITcostsQuicktostart,agile“VC=VISACard”
July17,2009 52ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
Future‐Fortune1M
Our vision is to enable one person to invent and run the next revolutionary IT service, operationally expressing a new business idea as a multi-million-user service over the course of a long weekend. By doing so we hope to enable an Internet "Fortune 1 million”.
RAD Lab, Berkeley
“ ”
July17,2009 53ISSGC09‐SophiaAn1polis
Cloudcompu0ng,Virtualisa0on&theFuture
EnterMicrostartups!
Thezerocoststartuphasledtotheageofthe“microstartup.”It’snolongertwofolksinagaragehopingtobuildaprototypeinordertolandahugeVCround,thengeAngmillionsofdollarstobuildoutanoffice.Microstartupsaresustainablefromprototypetolaunchandontoacoreuserbase,allforaround$5‐10,000incosts.
Jason(Calacanis.com)
“ ”
July17,2009 54ISSGC09‐SophiaAn1polis
STARTUP@BGi
Scalability–Howtobuild–technologyandbusinessaspects
Course‐Learninghowtoleverageoncloudcompu1ngCourse–“Startupschool”
Networking–alumni,investors,excellencecenters,industryMentors–findingandfollowup
Prototypingresources‐thewholeBal1cCloudto“playwith”
Bal0cGridInnova0onLab(BGi)–thestar0ngpointforearlystagestartupsintheBal0cs
July17,2009 55ISSGC09‐SophiaAn1polis
STARTUP@BGi
BGiisauniquestar0ngpointformicrostartups
Wanttoknowmore?
Tryit!Youdon’thavetobeintheBal0cs
July17,2009 56ISSGC09‐SophiaAn1polis
Acknowledgement and references
Alan Williamson, www.aw20.co.uk Berkeley RAD lab, berkeleyclouds.blogspot.com Baltic Cloud team, cloud.balticgrid.eu Rich Wolski
and his Eucalyptus team, eucalyptus.cs.ucsb.edu Dan Reed, Microsoft,
www.hpcdan.org/reeds_ruminations/microsoft 451 group, www.451group.com July17,2009 57ISSGC09‐SophiaAn1polis
STARTUPSCLOUDSBGi
Links
July17,2009 ISSGC09‐SophiaAn1polis 58
BGiandBal0cCloud:
[email protected]@gmail.comcloud.bal1cgrid.eu Startups, and Microstartups:
www.infochachkie.com blog.guykawasaki.com microstartups.blogspot.com
Clouds: berkeleyclouds.blogspot.com
Cloudcompu0ng,Virtualisa0on&theFuture
Finalslide–yes,youmadeit!
July17,2009 59ISSGC09‐SophiaAn1polis
July17,2009 ISSGC09‐SophiaAn1polis 60
Thanks!