Journey to Cloud: Fast Track to Azure

© Capgemini 2019. All rights reserved |Journey to Cloud: Fast Track | Fausto Pasqualetti | May 2019
Journey to Cloud: Fast
Track to Azure
Fausto Pasqualetti
Harness the cloud to accelerate
transformation, safe

2© Capgemini 2019. All rights reserved |Journey to Cloud: Fast Track | Fausto Pasqualetti | May 2019
Agenda
Journey to Cloud
Cloud Assessment
High Level Architecture
Cloud Migration

A
G
E
N
D
A
Journey to Cloud: Fast Track | Fausto Pasqualetti | May 2019
Journey to Cloud
Business Matter
Yes, money is a factor, in several ways.
ROI is easier to forecast, and implementation costs are minimal.
Services can be easier and less expensive.
It is scalable without breaking the budget, enabling both online and geographic
expansion.
It lets an organization do more with less downtime, cost, and
It reduces infrastructure overhead

A
G
E
N
D
A
Journey to Cloud: Fast Track | Fausto Pasqualetti | May 2019
Let’s get started
Move your application to
Azure is like to move your
House in the Cloud.
Easy only if you know what
you want to do.

Infrastructure as a Code DevOps
XaaS Cloud Brokering
Hybrid IT

We are living in a Complex IT Universe…
CONTAINERS BATCHES EVENT-DRIVEN
FUNCTIONS
MICROSERVICES DATA SERVICES MONOLITHIC
APPLICATIONS

…With different apps and workloads
to deploy…
CONTAINERS BATCHES
MICROSERVICES DATA SERVICES MONOLITHIC
APPLICATIONS
EVENT-DRIVEN
FUNCTIONS
IaaS
Application Platform
(PaaS)
Serverless Functions
(FaaS)
Containers
(CaaS)

XaaS Evolution
IaaS
CaaS
PaaS
FaaS
 Higher flexibility
 Higher Operational
Effort
 Lower Velocity
 Lower Abstraction
 Lower development
effort
 Higher Operational
Efficiency
 Higher Velocity
 Higher Abstraction

6Rs Cloud Journey Business View
9© Capgemini 2019. All rights reserved |
Time to migrate
Repurchase
Rehost
Replatform
Refactor
Cost,Effort,Skills
Cloud Migration Statergies Comparison
By 2011, Gartner stated 5 R’s for migration to cloud.
By 2017, a new R was added.

Workload Placement
Cloud Hosting Ready - Gen 2
 Self-contained, virtualized, multi-
tiered, web/thin client enabled
 Limited file system requirements/
leverage of cloud architecture &
features (e.g. typical vertical scaling)
 Very aware of required OS/runtime
environment and availability
Cloud Aware – Gen 2.5
 Modern applications, more
componentized and more API centric
 App design can tolerate infra failures
 Takes advantage of some cloud
features such as horizontal scaling or
cloud based DR
 Micro-service _ 12 Factor App
Architecture & Principles
 API first design
 Design for Failure
 Metrics & monitoring baked in
 Horizontal & massive scalability
 Cloud agnostic runtime
implementation
Cloud Native – Gen 3Traditional - Gen 1
 Legacy, monolithic applications
 Deployed on Physical Infrastructure
and on Composable
Re-host Re-Architect Re-factorRetire Re-purchase Retain
Automated provisioning Workload portability
No vendor
lock-in
Performance and agility
Low total cost of ownership (TCO) Availability Scalability Security and compliance
Criteria to consider
 New Development
 Web Application
 OpenSource RDBMS
 Containerized Apps
 Functions
 Application Servers
 OpenSource RDBMS
 Containerized Apps
 Scale Out Workload
 Very Large RDBMS
 Legacy COTS
 RDBMS requiring DR
 Application requiring DR
 COTS with specific certification
 Matrix (e.g. SAP/Oracle)

The Phoenix Project
A novel about IT, DevOps and helping your business win

Cloud native means…
Use 12-factor app principles to create cloud native applications (12factor.net)
A set of best practices for
developing and deploying
cloud-native software
Practices translate into
platform features and
workflow requirements

Benefits of Infrastructure as a Code
Capgemini suggest to use Infrastructure as a Code methodology in order to provide high level of
automation for the processes of design, build and deployment, all the infrastructure components
whatever is possible will be represented using a modeling language (e.g. XML, Jason files etc etc)
 Build services around the business capabilities
you require
 Scale up and down as required with virtually
no notice
 Make configuration code changes repeatable
and standardized
 API-driven model enables management of infrastructure
with language typically used in application code
 Free developers from manually configuring operating
systems, system applications, and server software

Code & Image Lifecycle with Infra as a Code
DEPENDENCY
ARTIFACTS AND
APPLICATION BUILD
CLONE
EDIT
ANALYZE
1
2
3
BUILD4
PACKAGE5
RUN6
DEBUG7
COMMIT8
PUSH9
CLONE
EDIT
1
2
BUILD
3
LABEL4
PUSH5
CODE LIFECYCLE IMAGE LIFECYCLE

Modernize existing .NET Applications
Author:
Cesar de la Torre, Sr. PM, .NET Product Team, Microsoft Corp.
To optimize the cost-effectiveness of your organization's cloud migration
strategy, it's important to consider the needs of your business and
requirements of your apps. You'll need to determine:
 Which apps require a transformation or re-architecting.
 Which apps need to be only partially modernized.
 Which apps you can "lift and shift" directly to the cloud
https://docs.microsoft.com/en-us/dotnet/standard/modernize-with-azure-and-containers/

XaaS – How to move your house?
Cloud Infrastructure-Ready
Lift and shift
Cloud-Aware
Re-platform
Cloud-Native
Re-platfom, Re-architect and re-write
Application's compute target
Applications deployed to VMs in Azure Monolithic or N-Tier apps deployed to Azure App Service,
Azure Container Instance (ACI), VMs with containers,
Azure Service Fabric, or AKS (Azure Kubernetes Service)
Containerized microservices on Azure Kubernetes Service
(AKS), Service Fabric and/or serverless microservices
based on Azure Functions.
Data target
SQL or any relational database on a VM Azure SQL Database Managed Instance or another
managed database in the cloud.
Fined-grain databases per microservice, based on Azure
SQL Database, Azure Cosmos DB, or another managed
database in the cloud
Advantages
No rearchitecting, no new code Least effort for quick
migration Least-common denominator supported in Azure
Basic availability guarantees After moving to the cloud, it's
easier to modernize even more
No rearchitecting Minimal code/config changes Improved
deployment and DevOps agility to release because of
containers Increased density and lower deployment costs
Portability of apps and dependencies Flexibility of host
targets: PaaS approaches or IaaS
Architect for the cloud, you get the best benefits from the
cloud but new code is needed Microservices cloud-native
approaches Modern mission-critical applications, cloud-
resilient hyper-scalable Fully managed services Optimized
for scale Optimized for autonomous agility by subsystem
Built on deployment and DevOps
Challenges
Smaller cloud value, other than shift in operating expense
or closing datacenters Little is managed: No OS or
middleware patching; might use infrastructure solutions,
like Terraform, Spinnaker, or Puppet
Containerizing is an additional step in the learning curve
for developers and IT Operations DevOps and CI/CD
pipelines is usually ‘a must’ for this approach. If not
currently present in the culture of the organization, it
might be an additional challenge
Requires rearchitecture for cloud native apps and
microservice architectures and usually requires significant
code refactoring or rewriting when modernizing
(increased time and budget) DevOps and CI/CD pipelines
is usually ‘a must’ for this approach. If not currently
present in the culture of the organization, it might be an
additional challenge

Migration Strategy to PaaS
CMO – Virtual Hosting FMO - Azure PaaS
Firewall – Load Balancer Application Gateway (Firewall- LB)
SQL Server on VMs Sql Azure
Web Applications Web apps
Network Virtual Network
Network Network Security Groups
DR/Backup Azure Site Recovery
Batch VMs Azure Web Jobs
SSiS VM Azure DataFactory v2
Monitoring Azure Monitor
Backup Azure Backup
SMTP SendGrid or Office 365

Maturity Model for .NET applications

Azure IaaS Reference Architecture

Azure PaaS Reference Architecture
Traffic
Manager
Internet
Azure Active
Directory
Azure DNS
Azure CDN
Production
SQL
Database
Redis cache
App Service plan
Web App
Static
content
Blob
Web App Web App Web App
SQL
Database
SQL
Database
SQL
Database
Pre-Production
SQL
Database
Redis cache
App Service plan
Web App
Static
content
Blob
Web App Web App Web App
SQL
Database
SQL
Database
SQL
Database

Migration Journey: Workplace Placement
Governance
Study Discovery & Assessment
Wave Migration
Planning
Migration Execution
Workload Wave
Evaluation Groups
Cloud Mapped
Workload
Cloud Suitable
Evaluation Groups
Migrated Workloads
Optimized &
Managed Workloads
Interview
Discovery
CMDB
Conversion
Bridge
Image
Container
Applications
Code / Services
Unix VM /
Partition
Physical
AS IS
IaaS / Virtual
Host
SaaS
CaaS / PaaS
Cloud Services
Bare metal
FaaS / Lambda
TO BE
Automated
& Manual
Migration
Capgemini Migration Factory
Capgemini
Cloud
Platform

High Level View on Execution
Migration Path
Wave Strategy
Technical & E2E testing
Requirements
Design & Build
deploy
Setup
Modifying
Split dei percorsi di lifecycle tra l’AS IS e TO BE con mantenimento della
doppia catena fino al final deploy e acceptance test.
Wave 1 Wave 3
TO BE
Analisi di Dettaglio Applicativa
E Database
Conversione configurazioni delle
componenti infrastrutturali (LB, VPN, AD)
Wave 2
Service
Readiness
TO BE
Re-platformingg
Re-hostingg
Wave 4 Wave 5
PILOTING
Re-platforming o
Re-hosting secondo
Complessità del cambiamento
Deployment
Setup
Lift & Shift
Cloud Setup Assessment Testing

A REAL CASE OF A CLOUD JOURNEY
Benefits
1. Reduction from 30 to 5 servers
2. Reduction Managment effort 70%
3. Reduction Deployment Time
Option 1 – Cloud Migration
Lift & Shift
Option 2 – Cloud Aware
 PaaS Deployment Automation
 Workload Optimization
 Management Optimization
 Infrastructure as a Code
Next Steps– Cloud Native
 Application Assessment
 PaaS/CaaS
 Cloud Native
 Microservices

Contact
Fausto Pasqualetti
Head of Cloud & Edge
Cloud Infrastructure Services
Capgemini Italy
fausto.pasqualetti@capgemini.com
https://www.linkedin.com/in/faustopasqualetti/
https://twitter.com/faustopasq

Thank you

Cloud Adoption Strategy .. the starting point
Cloud natives
apps
Traditional
apps
Data Center
Public Cloud
A traditional Insurance IT Department
 Owns its DataCenter
 Hosts Traditional Application to provide Business Services
 Relies on Public Cloud to target specific needs (such as offloading)
or to provide tactical and ad hoc solutions
major features Business needs and
Innovation priorities
are changing the
game
 Do More!!
 Faster !!
 With Less !!
In a context of continous
acceleration

Cloud Adoption Strategy .. the target point
Cloud natives
apps
Traditional
apps
Data Center
An Innovation-Ready Insurance IT Department
 Doesn’t really matter where DataCenter and/or Applications are
 Hosts Hybrid model and manage Cloud Native Apps aligned on
behaviour of Business needs
 A lot more automation both from a release management and
provisioning perspectives
major features
Public Cloud
… on top of traditionals ...
Speed of
Deployment
Frequency of
Deployment
Deployment
Success Rate
Financial
efficiency

Cloud Adoption Strategy .. Our Value in
roadmap implementation
From the initial Cloud Assessment to Target Design, Transformation and Running, Capgemini Cloud CoE (Center of
Excellence) brings you complete knowledge, capabilities, methologies and experiences to choose right path for the
right needs, optimizing time and costs
Avoiding Cloud Adoption Worst
practices such as:
 Move Infrastructure to Cloud
automatically brings to Cost decrease
 «Fast & Furious» Cloud Adoption without
taking into account the topologies of
Service, just Tranform to Cloud
 No need to fine analyse and optimize
before, just go to Cloud and everything
will work well

6Rs Cloud Journey

Cloud Native Pre-requisites
Competencies, capabilities, alignment
Group IT must Offer: PaaS platform, Standard Practices & Integrated Tools, Cloud Native Culture, API Based Architecture
Traditional
(Pets)
Cloud
Native
(Cattle)
DevOps Continuous Delivery Microservices
Spread Reponsibilities
Separate tools, Processes,Culture
Shared Responsibilities
Common tools, processes,culture
Static, release once 6 months
Code bugs in production
High Release Rate
Better Code Quality
Monolotic coupled components
Slow deployment cycle depending of testing team
Independant coupled components
Automated deploy without waiting individual components

Azure RDBMS Migration

Journey to Cloud: Fast Track to Azure

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Similar a Journey to Cloud: Fast Track to Azure

Similar a Journey to Cloud: Fast Track to Azure (20)

Último

Último (20)

Journey to Cloud: Fast Track to Azure

Notas del editor