Introduction to ‘Cloud Computing’

చక్ర�వర్తి�9/19 - 2009

Expertise level• Using Computer for email• Use computer for programming since

– 0 to 2 yrs– 2 to 4 yrs– 4 yrs and Above

• Will learn later• I’m here because of my friend / philosopher /

blah .. blah .. • ఏదో.. టైమ్పాస్

What, How and Why

• Do I care• Why do I care• I don’t care• ..

• కెరీర్ ఇం�కామొదలవ�లేద�, ఇంప్పు�డెం�ద�క్ర�?

Agenda

• The hype – stats• Industry definitions• History – a review of computing• Issues that matter• Origin• Grid vs Cloud• Details

The hype

Cluster ComputingCloud ComputingGrid Computing 

The hype

Some Definitions

definitions

.. Moving computing and data away from the desktop and the portable PC and simply displaying the results of computing that takes place in a centralized location and is then transmitted via the internet on the user's screen ..

- John Makroff

.. a computing paradigm shift where computing is moved away from personal computers or an individual application server to a "cloud" of computers ..

- Wikipedia

.. the idea of relayin on Web-based applications and storing data in the "cloud" of the internet

- MIT Technology Review

.. it starts with the premise that the data services and architecture should be on servers, We call it cloud computing - they should be in a 'cloud' somewhere..

- Eric Schmidt

Some other ..• Common, Location-independent, Online Utility on Demand*

– Common implies multi-tenancy, not single or isolated tenancy – Utility implies pay-for-use pricing– on Demand implies ~infinite, ~immediate, ~invisible scalability

• Alternatively, a “Zero-One-Infinity” definition:** 0 On-premise infrastructure

Acquisition costAdoption costSupport cost

1 Coherent and resilient environment – not a brittle “software stack”

Scalability in response to changing needIntegratability/Interoperability with legacy assets and other servicesCustomizability/Programmability from data, through logic, up into theuser interface without compromising robust multi-tenancy

* Joe Weinman, Vice President of Solutions Sales, AT&T, 3 Nov. 2008** From The Jargon File: “Allow none of foo, one of foo, or any number of foo”

.. the trend towards online services that run in a web browser and store users information in a provider's data center ..

- చక్ర�వర్తి�

location

location

computations & data are not in the place they used to be in..

location

why does it matter? a bunch of geeky stuff..

possession implies control

control implies power

history

of computing, how is changed

distributed central

history of evolution of computing

SUN 3 workstation

Centralized

Low CostExpert Management

Distributed

Rich user InterfaceUser Autonomy

Server

Client

Client

Client

Client

Client Server – 80s & 90s

ServerClient

ServerClient

Data Store

S Q L

C #

JavaScript

HTML

XML

ASP.NET

Intermediary Application – Cloud Implementation Strategy

enterprise impact

Cultural Challenges

Integration is hard

ExpensiveInfrastructure

Data

Data Integration

Adds meaning

Data Integration is hard

InvisibleHard to quantify Discounted

Financial & Logistics Challenges

Cost v. BenefitTime (Cost v. Benefit)

Traditional Software Purchase

Purchase softwarePurchase hardwareFind space in data centerCreate migration planSetup development & testConfigure systemsConfigure databasesConfigure networksMore…

The Easy Way

Get a login

Data Integration is hard

Costs don’t match benefits

How do we make integration easier, and deliver benefits more

quickly?

Define Cloud Computing

How do we make integration easier, and deliver benefits more quickly?Cloud computing is a way to make integration easier, and deliver benefits more quickly

Turn a commodity into a utility

integration^

Necessary, but not

differentiating

Not everything’s a commodity

Web Server Farms. Commodity.

Much Enterprise Software. Commodity

Email. A commodity.

Messaging. Maybe not a commodity.

What if… we just gave IT a platform to create their own data models, interfaces, and processes on a dynamic infrastructure [that met corporate requirements] & simply existed as needed?

Results in elevated IT relevance

Other Results…

Focus on integration will evolve to a more disciplined approach

Match expenses to benefits

Enable new classes of applications

Best Practices

Mediation. A secret weapon.

Service Level Management. Don’t even start with “my piece is working fine!”

Security. It’s not (only) what you think it needs to be.

Build a culture of collaboration.

Example: Google Apps

Cloud Application(Google Docs)

Cloud Client(Mozilla Firefox)

Cloud hardware(Apps Server)

Cloud Platform(Apps Engine)

Cloud Services(Word Processing)

BigTable

Cloud storage(BigTable DB)

issues that matter

Implications

possession of data

privacy

our data is hold by 3rd party

security

portability and preservation

“possession” of computation

market power

how big is the problem?

location

Understanding Cloud Computing

Origin

• “.. Comes from the early days of the Internet where we drew the network as a cloud… we didn’t care where the messages went… the cloud hid it from us” – Kevin Marks, Google

• First cloud around networking (TCP/IP abstraction)• Second cloud around documents (WWW data

abstraction)• The emerging cloud abstracts infrastructure complexities

of servers, applications, data, and heterogeneous platforms– (“muck” as Amazon’s CEO Jeff Bezos calls it)

3 Cloud Service Models

• Cloud Software as a Service (SaaS)– Use provider’s applications over a network

• Cloud Platform as a Service (PaaS)– Deploy customer-created applications to a cloud

• Cloud Infrastructure as a Service (IaaS)– Rent processing, storage, network capacity, and other

fundamental computing resources

• To be considered “cloud” they must be deployed on top of cloud infrastructure that has the key characteristics

Service Model ArchitecturesCloud Infrastructure

IaaS

PaaS

SaaS

Infrastructure as a Service (IaaS) Architectures

Platform as a Service (PaaS)Architectures

Software as a Service (SaaS)

Architectures

Cloud Infrastructure

SaaS

Cloud Infrastructure

PaaS

SaaS

Cloud Infrastructure

IaaS

PaaS

Cloud Infrastructure

PaaS

Cloud Infrastructure

IaaS

4 Cloud Deployment Models

• Private cloud – enterprise owned or leased

• Community cloud– shared infrastructure for specific community

• Public cloud– Sold to the public, mega-scale infrastructure

• Hybrid cloud– composition of two or more clouds

Common Cloud Characteristics

• Cloud computing often leverages:– Massive scale– Virtualization– Non-stop computing– Free software– Geographic distribution– Service oriented software – Autonomic computing– Advanced security technologies

is this GRID Computing?

Clouds aren’t all the same

• Not every cloud is a “grid”– Grids imply dynamic arrival/departure– Electrical analogy has limits: CPU cycles aren’t substitutable

• Most clouds are not “compute clusters”– Clusters are typically monocultures: just one type of node– Applications may require tuning to a particular cluster size

• Some clouds are servers in virtual slices– Virtualized servers can be quickly provisioned– Spin-up of instances = new management task– Hardware gets cheaper, management…not so much

• Enterprise cloud computing implies API leverage

– Immediate focus on function; immediate delivery of value

– Using appropriate frameworks enables a huge head start

Single tenancy gives each customer a dedicated software stack – and each layer in each stack still requires configuration, monitoring, upgrades, security updates, patches, tuning and disaster recovery.

On a multi-tenant platform, all applications run in a single logical environment: faster, more secure, more available, automatically upgraded and maintained. Any improvement appears to all customers at once.

Shared infrastructure

Other apps

ServerOS

DatabaseApp Server

StorageNetwork

App 1

ServerOS

DatabaseApp Server

StorageNetwork

App 2

ServerOS

DatabaseApp Server

StorageNetwork

App 3

Single-Tenant (vs) Multi-Tenant Architecture

Coherent Code Base and Managed InfrastructureCoherent Code Base and Managed Infrastructure

Your Clicks

Your Code

User Interface

Logic

Database

Metadata representations:Partitioned data, logic and customizations for multiple customers

Build strategic applications

Customize any aspect

Upgrade when convenient

Retain IP ownership

The technical part : Why multi-tenant matters

Virtual ServersVirtual Servers

Virtual ServersVirtual Servers

Virtual Servers

PaaS for theInquiring Developer

“Servers as a Service” PaaS as anApplication Framework

Infrastructure as a Service

Database as a Service

Integration as a Service

Logic as a Service

UI as a Service

Infrastructure as a Service Infrastructure as a Service

Database as a Service

Python App Server

~Familiar Developer ModelRapid Scalability

OfferingInnovative Technology

Supports Large-Scale SaaSDeep-Dyed Multitenancy

Database as a Service

PaaS Taxonamy : Proliferating Platforms

Thank you