cloud computing - making it simple

<Insert Picture Here>

OTN Architect Day Cloud Computing - Making IT Simple Dr. James Baty VP, Oracle Global Enterprise Arch. Pgm.

© 2011 Oracle Cloud Computing – Making IT Simple 2

The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle’s products remain at the sole discretion of Oracle.


How key is standardization? What are Deployable Entities? ‘Refactoring’ Dev / Ops Roles Building a Roadmap to Cloud


Moving to Cloud First Inventory Applications & Workloads

Suitable for cloud now Time based

Very parallel (i.e. batch) Spiky traffic

Capital intensive (especially startup)

Proof of Concept Low utilization

Less deployment costs High bandwidth costs / high real

estate

Not as suitable for cloud Vertically scaled applications

Consistent load levels Latency sensitive applications

Insecure applications Hardware device dependent (e.g.

fax server, SNA gateway) ISV unsupported

Per CPU licensed applications


Development and Test

Resource sharing (consolidation)

Shared Services

Augmentation (Elastic scaling)

What Do You Want the Cloud to Do? Start with Common Use Cases

Most enterprises are trying • Shared development and test environments • Hardware & Services consolidation


Shared Services – Many Possibilities

X X X

X X X X

Java PaaS

Shared SOA

Shared Security

• Build & deploy to common platform • Enterprise Private Cloud

• Application services integration

• Sharing Applications across org • Enabled by SOA, BPM

• Centralized authorization for all apps

Shared Functions

DBaaS • Rapid access to all enterprise data • Parallel Processing of Transactions


Private Database Cloud Architectures Common Building Blocks - Shared Server & Storage Pools

Server Deploy in dedicated VMs

Server virtualization

Hypervisor

CRM DW ERP

OS

DB

OS

DB

OS

DB

OS

ERP DW CRM

DB

OS

DB

DB

Operating System

Share server pool

Real Application Clusters

OS

ERP DW CRM

OS

DB

Database

Share database instance

Real Application Clusters

Hypervisor


Which Apps for Which DB Cloud? Each Architecture Serves Different Workloads

Workload Type Optimal Cloud Architecture

Mission or Business Critical Deployment Operating System

Packaged Applications Operating System

Data Warehouse Applications Operating System

Standardized environment Operating System or Database

Internal Applications Database

Rapid provisioning (i.e. Test and Dev) Database or Server

Mixed workload consolidation Server

As-Is consolidation Server


Database Cloud Planning Identification of Applications to Migrate

-  New applications are deployed to the Cloud -  Existing applications are migrated based on:

•  Difficulty •  ROI •  Suitability

-  The benefits and difficulties of consolidating existing applications in the Cloud will vary

•  Applications with highly varying peaks will show greatest benefit

-  The “lowest hanging fruit” should be migrated to the Cloud first


Roadmap to Cloud Multi-Dimensional Journey

Optimize

Automate

Consolidate

Standardize

Individual enterprises or applications may join the roadmap at different points

Define a single solution

for a given problem

Reduce the footprint of deployed

applications

Reduce the manual tasks for managing

IT

Achieve new operational models & greatest

efficiency


Engineered Systems in the Cloud – extreme performance PaaS services

•  Reference configuration •  Known sizing •  Order as ‘part number’ •  Unified support •  Simplified deployment •  Run existing apps •  Enterprise scale •  High performance


Build from Scratch vs. Exadata Commonwealth Bank of Australia

• DB deployment time reduced from 3 months to < 1 week

Build From Scratch with Components

Reference Configurations

Take delivery of Oracle Database Machine

Weeks to Months

Acquisition of components

Installation and configuration

Acquisition of components

Installation and configuration

Testing and Validation

Testing and Validation

Weeks to Months

Oracle Exadata Database Machine

 Server Pool pre-configured

 Faster deployment

 Lower Risk

< 1 Week after Delivery

Testing and Validation Configuration

Pre-implementation System sizing


How key is standardization?

What are Deployable Entities? ‘Refactoring’ Dev / Ops Roles Building a Roadmap to Cloud


vDCs

Serv

ices

PaaS Container

IaaS Storage Network Server

SaaS Application

Queue

Business Process Business Service

Interfaces Portals Native Protocols Custom UIs Self-Service APIs

Acc

ess

Facilities Perimeter Security Proxy Naming Balancing

Reso

urc

es

Physical Pools Networks Compute Storage

Logical Pools Networks Compute Storage External Clouds

Legacy

Partners

Other

Data

Pool Managers

Security / Policy Mgmt

Mediation, Policy

enforcement

Service Mgmt Monitoring

Capacity mgmt. Metering & Billing Resource mgmt.

Model Mgmt Provisioning

Customer info model Service catalog

User Interaction / self service

Clo

ud

Man

ag

em

en

t

Other

© Oracle, 2010 (GEAP)

Cloud Architecture - Logical View


Compatibility & Portability Implications

Deployable Entities

Serv

ices

Platform as a Service

Infrastructure as a Service

Software as a Service

Acc

ess

R

eso

urc

es


Service Mgmt

Model Mgmt

Clo

ud

Man

ag

em

en

t

Consumers & Delivery Channels

Cloud Service Provider

Cloud Service Consumer

Cloud Service Developer

Access Points Portals Native Protocols Custom UIs Self-Service APIs

Storage Networks Servers Legacy Systems

Other Clouds

Partner Systems


API Code deployment,

developer discovery and application control

Images Application and data

deployment

Data Access and

formatting of data may differ between

clouds

Model Architectural models of cloud applications may be external (deployment code), embedded (e.g.

OVF) or internal

Services Run time services within a particular

cloud may be absent or

significantly different


Deployable Entities

Serv

ices

Platform as a Service

Infrastructure as a Service

Software as a Service

Acc

ess

R

eso

urc

es


Service Mgmt

Model Mgmt

Clo

ud

Man

ag

em

en

t

Consumers & Delivery Channels


Cloud Service Consumer


Access Points Portals Native Protocols Custom UIs Self-Service APIs

Storage Networks Servers Legacy Systems

Other Clouds

Partner Systems


Cloud Infrastructure – Key Abstractions – making it simple

•  Separation of roles (e.g., Cloud Provider vs Service Developer)

•  Model Management incl. Services, Consumers, etc.

•  Deployable Entities (aka VDCs) include both Service Templates and Service Context (e.g. – VAB ‘Assemblies’)

•  Separation of control plane ‘cloud’ mgmt vs ‘pool’ resource mgmt

•  Resources abstracted as logical resource “pools” which are addressable.


Service Catalogue

• Repository of models • Models contain

templates and service contracts -  Payloads like virtual

server images -  Application metadata

such as configurations and policies


Model / Service Management •  Model management supports

build-time functions in cloud -  Exposes resources to developers -  Stores developer’s models for

deployment -  Validates cloud models

•  Developer’s main point of interaction with the cloud

•  Model management also contains the solution catalogue, facilitating discovery, storage, use and re-use of cloud services

•  Service management controls the run-time aspects of the cloud

-  Capacity management -  Service management

•  Operator’s main point of contact for the cloud

•  Provisions / monitors resource tier •  Contains configuration

management repository which is the current state of all cloud vDCs


Oracle Virtual Assembly Builder – automating deployment of ‘assemblies)

• Package up complex structure from dev/test and reconstitute in production • Minimize setup time and risk of hard-to-debug configuration errors • Easily replicate in production with minor variations • Each production instance has well-contained configuration parameters for flexibility

config1

Dev/Test Environment

Production Environments

config2

Assembly = Appliances (VM Templates + configuration Metadata) + relationships & start order Metadata


How Key is Standardization? What are Deployable Entities?

‘Refactoring’ Dev / Ops Roles Building a Roadmap to Cloud


vDCs

Serv

ices

PaaS Container

IaaS Storage Network Server

SaaS Application

Queue

Business Process Business Service

Interfaces Portals Native Protocols Custom UIs Self-Service APIs

Acc

ess

Facilities Perimeter Security Proxy Naming Balancing

Reso

urc

es

Physical Pools Networks Compute Storage

Logical Pools Networks Compute Storage External Clouds

Legacy

Partners

Other

Data

Pool Managers


Mediation, Policy

enforcement

Service Mgmt Monitoring

Capacity mgmt. Metering & Billing Resource mgmt.

Model Mgmt Provisioning

Customer info model Service catalog

User Interaction / self service

Clo

ud

Man

ag

em

en

t

Other


Identify Roles and Interactions – Cloud implies changes in IT roles

User

uses service

Cloud Operator

creates resources

Monitor/manages cloud

Cloud Builder

App Owner DevOps Developer

Monitors & Approves Services

Creates Services Packages & Deploys Services

Models Service


Build time vs run time perspectives – need to separate development / operations horizons

Process Modeling, Simulation and Documentation

Deployment Engineering

Cloud Monitoring and Metrics

Service Identification & Discovery

BUILD TIME

RUN TIME


Cloud Roles / Actors – each role category has possible sub-roles

Category Actor DMTF Roles


Operator Service Operations Mgr

Builder Service Business Mgr

Service Transition Mgr


Service Developer Service Developer

Deployer / DevOps

Cloud Service Consumer Application Owner

Consumer Business Mgr Consumer Service Admin

User Service User


Set up Cloud Platform

Set up self-service portal

Set up shared

components

Dept App

Build app using shared

components IT

App Developer

Deploy using self service

App Users

1. Set Up Cloud 2. Build App

3. Use App

App Owner

4. Manage App

Manage

Adjust Capacity

Review Charge-back

Use app

Self-Service Interface Shared Components

Application Server

Integration / SOA BPM Portal Security &

Identity

System Manager

Database

Operating System,Virtualization,Server,Storage

Example of Build vs Run Time – Oracle Virtual Assembly Builder roles

Oracle Cloud Platform

BUILD TIME

RUN TIME


How Key is Standardization? What are Deployable Entities? ‘Refactoring’ Dev / Ops Roles

Building a Roadmap to Cloud


Cloud Computing Readiness May Require Diverse Business Changes

 Consider, for example, IT governance & risk management, information modeling & ownership, operations & service management.

 How are these areas managed today? •  identified responsibilities, documented processes, etc.

 Do you have a mechanism for assessing capabilities in each area?

 How will you identify needs for changes or improvements to support cloud computing?


Example Cloud Capabilities by Domain Areas Important to Cloud Readiness

•  To succeed at Cloud services adoption, an organization must adequately progress in all the appropriate domains.

 Business & IT drives  Costs & Benefits

 Executive sponsorship  Roles & responsibilities

 Risk management  Cloud change management

 Services portfolio management  Services engineering approach

 Capacity management  Operational tools & processes

 Model packaging  Service monitoring

 Model templates  Data ownership

 Reference architecture  Standards


Cloud Computing Roadmap How Will Your Cloud Be Introduced?

Cloud computing efforts range from small ‘experiment’ projects to major strategic initiatives.

Most companies have multiple projects underway or anticipated. •  Separate from the new architecture, is there a plan for

how the new model will be rolled out? •  E.g., by application, by business unit, by geography…

•  Is cloud viewed as a limited tactical deployment, major strategic initiative, or both?


Focused Implementation Vs. Wide Diffusion

•  Strategic – Complete migration for a given architecture/application, often focusing on revenue enhancement

•  Tactical – Wide deployment of a limited technology (e.g. virtualization), often focusing on cost reduction

Clo

ud

M

atu

rity

Cloud Adoption

Ad Hoc Cloud

Opportunistic Cloud

Systematic Cloud

Optimized Cloud

Enterprise Level

Region Level

Suite Level

Data Center Level

Exploring

Expanding

Exploiting

Tactical

Strategic Managed Cloud

Application Level


Key Business/IT Transformations

IT Architecture designed up front (early binding)

IT operations developed and performed by the IT department

Systems and application management was specific to select systems and applications

Build infrastructure up-front, deploy later (late binding)

IT will move to building up-front operational functions for a self-service model.

The cloud ‘control plane’ has to be architected as a general service

Cloud Approach Current Approach


‘Enterprise’ Private Clouds are Different

•  NIST identifies 5 essential cloud characteristics -  On-demand self-service, Resource pooling, Rapid

elasticity, Measured service, Broad network access

•  But private clouds are different from public clouds… -  E.g., On-demand self service

•  Public developer cloud – unrestricted resources provisioned in minutes, but with no controls or corporate governance

•  Enterprise private cloud – need provisioning controls, standards enforcement, prioritization, approvals, etc.

-  I.e., Enterprise cloud faster to deploy than traditional IT, but probably slower that public cloud

•  There are other criteria with similar differences -  Security, governance, high availability, global access ….

Iaa

Paa

Saa

IaaS

PaaS

SaaS I N T R A N E T


For More Information….

oracle.com/cloud

cloud computing - making it simple

Technology

deployable

applications

service

cloud

front

oracle

build

application