Basics Of Cloud Computing - Kursused

Basics of Cloud ComputingMTAT.08.027 Basics of Cloud Computing (3 ECTS)MTAT.08.011 Basics of Grid and Cloud ComputingSatish Sriramasatish.srirama@ut.ee

Course Purpose Introduce cloud computing concepts Introduce cloud providers Introduction to distributed computingalgorithms like MapReduce Glance of research at Mobile Cloud Lab incloud computing domain http://courses.cs.ut.ee/2013/cloud/4/3/2013Satish Srirama2/33

Questions Is everyone comfortable with data structures? How comfortable you are with algorithms? How comfortable you are with programming?– Java ? External APIs?– Python – I assume you are– Web programming Were you able to submit all exercises in Gridpart?4/3/2013Satish Srirama3/33

Outline Cloud computingCloud providersMapReduceMapReduce in different domainsLarge scale data processing on cloud4/3/2013Satish Srirama4/33

Grading Written exam – 50% Labs – 45%– 6 lab exercises Active participation in the lectures (Max 5%) To pass the course– You need to score at least 50% in each of thesubsections– You need to score at least 50% in the total4/3/2013Satish Srirama5/33

Course schedule 03.04 Basics of Cloud computing10.04 Cloud Providers17.04 MapReduce24.04 MapReduce algorithms01.05 No lecture – May day08.05 MapReduce in Information Retrieval15.05 Cloud scale distributed data storage22.05 No lecture– I will be on a project meeting in Paris4/3/2013Satish Srirama6/33

Course schedule - continued Labs03-09.04 Starting with a cloud10-16.04 Working with SciCloud17-23.04 MapReduce - Basics24-30.04 Data analysis with MapReduce08-14.05 MapReduce in information retrieval15-21.05 NoSQL4/3/2013Satish Srirama7/33

Course schedule - continued 28.05 - Examination 1 29.05 - Examination 2 Students who are defending their thesesshould pass the exam in the first attempt Examination for second attempt students 12thJune4/3/2013Satish Srirama8/33

Lecture 1CLOUD COMPUTING4/3/2013Satish Srirama9

“It’s nothing new”“It’s a trap”“.we’ve redefined Cloud“It’s worse than stupidity: it’sComputing to include everythingmarketing hype. Somebody isthat we already do. I don’tsaying this is inevitable—andunderstand what we would dowhenever you hear that, it’s verydifferently . other than changelikely to be a set of businessesthe wording of some of our ads.”campaigning to make it true.”WHAT IS CLOUD COMPUTING?Larry Ellison, CEO, Oracle (WallRichard Stallman, Founder, FreeStreet Journal, Sept. 26, 2008)Software Foundation (TheGuardian, Sept. 29, 2008)No consistent answer!Everyone thinks it is something else 4/3/2013Satish SriramaSlide taken from Professor Anthony D. Joseph’s lecture at RWTH Aachen10

What is Cloud Computing? Computing as a utility– Utility services e.g. water, electricity, gas etc– Consumers pay based on their usage Cloud Computing characteristics– Illusion of infinite resources– No up-front cost– Fine-grained billing (e.g. hourly) Gartner: “Cloud computing is a style ofcomputing where massively scalable IT-relatedcapabilities are provided ‘as a service’ across theInternet to multiple external customers”4/3/2013Satish Srirama11/33

Timeline4/3/2013Satish Srirama12/33

How Cloud & Grid are related Share a lot commonality– Intention, architecture and technology Differences– Programming model, business model, computemodel, applications, and Virtualization. The problems are mostly the same– Manage large facilities;– Define methods by which consumers discover, requestand use resources provided by the central facilities;– Implement the often highly parallel computations thatexecute on those resources.4/3/2013Satish Srirama13/33

Virtualization Virtualization techniques are the basis of the cloudcomputing Virtualization technologies partition hardware and thusprovide flexible and scalable computing platforms Virtual machine techniquesApp App App– VMware and Xen– OpenNebula– Amazon EC2OS GridOSOSHypervisorHardwareVirtualized Stack– do not rely on virtualization as much as Clouds do, eachindividual organization maintain full control of their resources For cloud, virtualization is almost an indispensableingredient4/3/2013Satish Srirama14/33

Enabling Grids for E-sciencE - EGEE4/3/2013Satish Srirama15/33

Clouds - Why Now (not then)? Commoditization of HW– x86 as universal ISA, plus fast virtualization– Bet: Can statistically multiplex multiple instances onto a single boxwithout interference between instances Web 2.0– Standard software stack, largely open source (LAMP)– Asynchronous JavaScript and XML (AJAX) Novel economic model: fine grain billing– Earlier examples: Sun, Intel Computing Services—longer commitment,more /hour Infrastructure software: e.g. Google FileSystem, HDFS Operational expertise: failover, DDoS, firewalls. More pervasive broadband InternetISA - Instruction Set ArchitectureLAMP – Linux, Apache Http Server, MySQL, PHP, perl or python4/3/2013Satish Srirama16/33

Cloud Computing - Services Software as a Service – SaaSLevel ofAbstraction– A way to access applicationshosted on the web through yourweb browserSaaS Platform as a Service – PaaSFacebook, Flikr, Myspace.com,Google maps API, Gmail– A pay-as-you-go model for ITresources accessed over theInternetPaaS Infrastructure as a Service –IaaS– Use of commodity computers,distributed across Internet, toperform parallel processing,distributed storage, indexing andmining of data– Virtualization4/3/2013Google App Engine,Force.com, Hadoop, Azure,Amazon S3, etcIaaSAmazon EC2, SciCloud,Joyent Accelerators, NirvanixStorage Delivery Network, etc.Satish Srirama17/33

Cloud Computing - Themes Massively scalable On-demand & dynamic Only use what you need - Elastic– No upfront commitments, use on short term basis Accessible via Internet, location independent Transparent– Complexity concealed from users, virtualized,abstracted Service oriented– Easy to use SLAsSLA – Service Level Agreement4/3/2013Satish Srirama18/33

Cloud Models Internal (private) cloud– Cloud with in an organization Community cloud– Cloud infrastructure jointlyowned by several organizations Public cloud– Cloud infrastructure owned byan organization, provided togeneral public as service Hybrid cloud– Composition of two or morecloud models4/3/2013Satish Srirama19/33

Short Term Implications of Clouds Startups and prototyping– Minimize infrastructure risk– Lower cost of entry Batch jobs One-off tasks– Washington post, NY Times Cost associatively for scientific applications Research at scale4/3/2013Satish Srirama20/33

Cloud Application Demand Many cloud applications have cyclical demand curvesResources– Daily, weekly, monthly, DemandTime Workload spikes are more frequent and significant– When some event happens like a pop star has expired: More # tweets, Wikipedia traffic increases 22% of tweets, 20% of Wikipedia traffic when Michael Jackson expired in 2009– Google thought they are under attack4/3/2013Satish Srirama21/33

Economics of Cloud UsersResourcesCapacityDemandResources Pay by use instead of provisioning for peakCapacityDemandTimeTimeStatic data centerData center in the cloudUnused resources4/3/2013Satish Srirama22/33

Economics of Cloud Users - continued Risk of over-provisioning: underutilization– Huge sunk cost in infrastructureCapacityResourcesUnused resourcesDemandTimeStatic data center4/3/2013Satish Srirama23/33

Economics of Cloud Users - continuedResourcesResources Heavy penalty for under-provisioning23Time (days)Lost me (days)CapacityDemand213Time (days)4/3/2013Satish SriramaLost users24/33

Economics of Cloud Providers Building a very large-scale datacenter is very expensive– 100 Million (Minimum) Large Internet Companies Already Building Huge DCs– Google, Amazon, Microsoft 5-7x economies of scale [Hamilton 2008]ResourceCost inMedium DCCost inVery Large DCRatioNetwork 95 / Mbps / month 13 / Mbps / month7.3xStorage 2.20 / GB / month 0.40 / GB / month5.5xAdministration 140 servers/admin 1000 servers/admin4/3/2013Satish Srirama7.1x25/33

Economics of Cloud Providers continued PowerPrice perKWHWherePossible Reasons Why3.6 IdahoHydroelectric power; not sent long distance10.0 CaliforniaElectricity transmitted long distance over thegrid; limited transmission lines in Bay Area; nocoal fired electricity allowed in California.18.0 HawaiiMust ship fuel to generate electricity Cooling is also expensive– Build data centers near rivers Extra benefits– Amazon: utilize off-peak capacity– Microsoft: sell .NET tools– Google: reuse existing infrastructure4/3/2013Satish Srirama26/33

Economics of Cloud Providers - Failures Cloud Computing providers bring a shift from highreliability/availability servers to commodity servers– At least one failure per day in large datacenter Why?– Significant economic incentives – much lower per-servercost Caveat: User software has to adapt to failures– Very hard problem! Solution: Replicate data and computation– MapReduce & Distributed File System (Will discuss later inLecture 3)4/3/2013Satish Srirama27/33

Adoption ChallengesChallengeAvailabilityData lock-inOpportunityMultiple providers & Useelasticity to prevent DDoSattacksStandardizationData Confidentiality and Encryption, VLANs,AuditabilityFirewalls; GeographicalData Storage4/3/2013Satish Srirama28/33

Growth ChallengesChallengeData able storageOpportunityFedEx-ing disks, DataBackup/ArchivalImproved VM support, flashmemory, scheduling VMsInvent scalable storeBugs in large distributed Invent Debugger that reliessystemson Distributed VMsScaling quicklyInvent Auto-Scaler;Snapshots for conservation4/3/2013Satish Srirama29/33

Policy and Business ChallengesChallengeOpportunityReputation Fate Sharing Offer reputation-guardingservices like those for emailSoftware LicensingPay-for-use licenses; Bulkuse sales4/3/2013Satish Srirama30/33

Long Term Implications of clouds Application software:– Cloud & client parts, disconnection tolerance Infrastructure software:– Resource accounting, VM awareness Hardware systems:– Containers, energy proportionality4/3/2013Satish Srirama31/33

Cloud Computing ProgressArmando Fox, 20104/3/2013Satish Srirama32/33

This week in Lab (Homework) Registration to the cloud & keys– Firefox plugin & working with Eucatools and API Study the following paper and write anexcerpt (At least 1 page)– M. Armbrust et al., “Above the Clouds, A BerkeleyView of Cloud Computing”, Technical Report,University of California, Feb, 2009. Write why and where you think cloud is usefulfor you?4/3/2013Satish Srirama33/33

Next lecture Cloud providers– Amazon EC2, S3, EBS Eucalyptus OpenStack4/3/2013Satish Srirama34/33

References Several of the slides are taken from Prof. AnthonyD. Joseph’s lecture at RWTH Aachen (March2010) Papers to read– M. Armbrust et al., “Above the Clouds, A BerkeleyView of Cloud Computing”, Technical Report,University of California, Feb, 2009. The Cloud: Battle of the Tech Titans– Cover story in Businessweek– http://www.businessweek.com/magazine/content/11 11/b4219052599182.htm4/3/2013Satish Srirama35/33

What is Cloud Computing? Computing as a utility -Utility services e.g. water, electricity, gas etc -Consumers pay based on their usage Cloud Computing characteristics -Illusion of infinite resources -No up-front cost -Fine-grained billing (e.g. hourly) Gartner: "Cloud computing is a style of