1copyright © 2013, oracle and/or its affiliates. all rights reserved


Using Cloud Computing to operate aSpatial Data Infrastructure efficiently

Hans ViehmannProduct Manager EMEA


NIST Definition of Cloud Computing

Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.

Essential characterictics: On-demand self-service, Resource pooling, Rapid elasticity, Measured service, Broad network access


Spatial Data Infrastructure (SDI)

• Distributed Web Services– Highly standardized– Unpredictable load requirements– eg. for INSPIRE – 727 services operational in Germany (May 2012)

• Heterogeneous, distributed source systems

• New requirements– Business specific: 3D data support, crowd sourcing, ...– Technical: data volumes, access rights, ...

• Limited resources

Distributed Responsibility for Data Provisioning


Cloud Computing for SDIs

• more efficient management of data and services by using a central hosted platform• Economy of scale

• reduce CapEx by using hosted services

• achieve elasticity to address variable load

• reduce time-to-market through self-service and higher degree of automation


Private

Deployment Model

Public

HybridProvider Owns

Provider Operates

Customer OwnsProvider Operates

Customer OwnsCustomer Operates

Operating Model

Applications

Platform

Service Model

Infrastructure

(SaaS)

(PaaS)

(IaaS)

Running SDIs on a Cloud Infrastructure


Customers Have a Choice of CloudsPrivate, Public, Hybrid

Private Cloud Public CloudHybrid Cloud

• OpEx• Fast & inexpensive to start• Outsourced services• Multiple tenants sharing

resources

• CapEx & OpEx• Lower total costs• Control & visibility• Multiple apps sharing

resources

• Cloudbursting – overdraft for peak loads

• Dev/Test & production• B2B integration

Exclusive Shared by multiple organizations


ApplicationApplication

PlatformPlatform

CustomizationsCustomizations

ApplicationApplication

CustomizationsCustomizations CustomizationsCustomizations

Co

nsu

me

r Co

nsu

me

r

Se

rvic

eP

rovi

de

r

Se

rvic

e P

rovi

de

r

Customers Have a Choice of CloudsIaaS, PaaS, SaaS

IaaS Cloud

SaaS CloudPaaS Cloud

IT Professional Developer Business End User

Different Users


Flexible Adoption – Roadmap to Cloud

Private Cloud

• Self-service• Auto-scaling• Metering &

chargeback• Capacity planning

Public Cloud

• Specialized• Shared• Standardized

Hybrid Cloud

• Federation across public & private clouds

• Interoperability• Cloudbursting

Traditional Silos Consolidated

• Physical• Dedicated• Static• Heterogeneous• Disparate

Spatial Data

• Virtual• Shared platform• Dynamic• Standardized

platform & infrastructure

• Integrated Spatial Data

Start with consolidation • Extend to private cloud • Use public cloud where appropriate


Consolidation at PaaS and IaaS Layers

App

IaaS

PaaS

Consolidate onto shared IaaS without standardization

Consolidate onto standard, shared and elastically scalable PaaS

vs.

App App

App App App

App App App

• Cost savings from less hardware, power and data center space

• Software stack heterogeneity, cost and complexity persists

• No administration (O&M) cost savings

• Standardized PaaS for all applications reduces heterogeneity, cost and complexity

• Accelerated new application development• Cost savings from less hardware, power and

data center space


Consolidatation of data and ServicesDatabase functionality such as

• Spatial queries

• Semantic queries

• Versioning/Long Transactions

SOA Plattform

• Application Server

• Service Bus/Orchestration

Required platform capabilities

PaaS

WebGIS Srv. App

f1

f2n1n2

e1

e2e3

e40D (points) 1D (lines) 2D (areas) Geo-Raster Network Topology 3D, LIDAR

WFS

CSW WCS

WMS

OpenLS


Exadata, Exalogic, SPARC SuperClusterExtreme Performance, Engineered Systems

• Building blocks for consolidation and cloud computing• Unmatched performance, simplified deployment, lower total cost


Complete Cloud Lifecycle Management

1. Plan & Setup the Cloud•Capacity & consolidation planning•Asset discovery•Bare-metal provisioning•Policy setup

2. Build, Test & Deploy Appson the Cloud•Packaging apps as assemblies•Testing applications•Self-service provisioning

3. Manage & Monitor the Cloud•Auto-scaling•Full stack management•End-user, business-level, app monitoring

4. Meter, Charge, Optimize•Metering resource utilization•Chargeback/Showback•Optimize performance, capacity, QoS


Cloud Computing – Security Aspects

Aud

iting a

nd Com

pliance

Security G

overnance

Identity and A

ccess Mana

gement

Measured S

ervice

Rapid E

lasticity

Resource P

ooling

Broa

d Netw

ork Access

On-D

emand S

elf-Service

Cloud Security Issues Traditional IT Security

Secure Development

Security Engineering

Security Operations

Incident Response

IT Service Mgmt.

Risk and Trust Management

Cloud ServiceModels

Cloud DeploymentModels

Cloud ManagementModels

Cloud InteractionModels

Cloud TrustModels


Recommendations

• develop a vision to move to cloud computing• save operational cost, improve time-to-market

• start with file and database consolidation• reduce complexity, improve security & availability

• consider engineered systems for consolidation• save operational cost, reduce deployment time, improve scalability

• integrate geospatial data in all layers of the stack• simplify SW development, improve security & availability

• use standards wherever possible• protect investments, improve interoperability


ITSO – IT Strategies from ORACLEMore best practices for Cloud Computing


Upcoming Events

• Oracle Spatial and Graph Users Conference 2013• Washington, DC, May 22• in conjunction with Location Intelligence 2013, May 21

• new: attend by Webcast, „live“ or „on-demand“• see http://www.locationintelligence.net/dc/registration/


Oracle Strategy

Complete Stack

• Best-of-breed

• Open

• Vertical Integration

• Extreme Performance

• Engineered Systems

Complete Customer Choice

• On-premise

• Private Cloud

• Public Cloud

• Hybrid Cloud

1copyright © 2013, oracle and/or its affiliates. all rights reserved

Documents

cloud infrastructure

oracle andor

cloud auto

data provisioning slide

cloud packaging apps

access rights

appropriate slide

qos slide