Accelerating a Path to Digital With a Cloud Data Strategy

ACCELERATING A PATH TO DIGITAL
WITH A CLOUD DATA STRATEGY
Wednesday 30th May 2018

WELCOME & INTRODUCTION
Sacha Vansimpsen,
Senior Account Executive
sacha.vansimpsen@mongodb.com

3
SPEAKERS
Accelerating a Path to Digital with a Cloud Data Strategy
MAT KEEP
BART VAN LOOCKE
DIRECTOR OF PRODUCT
MONGODB
SALES DIRECTOR
INFOSYS LIMITED
SOFTWARE ENGINEER
DE PERSGROEP
SENIOR SOLUTIONS ARCHITECT
MONGODB
EUGENE BOGAART
PRAVEEN BISHT
SACHA VANSIMPSEN
FREDERIC BOTHY
TOYOTA MOTOR EUROPE
MONGODB
SENIOR ACCOUNT EXECUTIVE
SYSTEMS ENGINEER

• Welcome
• Executing on a Cloud Data Strategy
• Customer Story: De Persgroep
• Break
• Customer Story:Toyota Motor Europe &
Infosys Limited
• Landing in the Cloud
• Q&A& Lunch
Agenda

EXECUTING ON A CLOUD DATA STRATEGY
Mat Keep, Director, Product Team
mat.keep@mongodb.com
@matkeep

Mainframe Client-Server Web
Control & Efficiency AGILITY & INNOVATION
Cloud & Mobile
Distributed Systems
1980s Late 90s – 2000s 2015>1960s-70s
The Cloud Age
Generational Shift

Seismic Shifts — Organizational

Seismic Shifts — Application Architecture

Why do I need to rethink my underlying data layer…..?
Support the velocity of
modern app development
Exploit commodity &
cloud platforms
Deploy anywhere, on-
demand, with no lock-in

API Access Layer
Operational Data
Customers
Products
Accounts
ML Models
Shared Physical Infrastructure
App1 App2 App3
1.Development agility
2.Corporate governance
3.Data re-use
Cloud Data Strategy
Standardized, on-demand database service
Cloud Portable
Any Cloud, Any Where

Best way to
work with data
Intelligently put data
where you need it
Freedom
to run anywhere
Intelligent Operational Data PlatformIntelligent Operational Data PlatformIntelligent Operational Data PlatformIntelligent Operational Data PlatformIntelligent Operational Data Platform
Why MongoDB?

Freedom To Run Anywhere
Database that runs
the same everywhere
Coverage in any
geography
Leverage the benefits
of a multi-cloud
strategy
Avoid lock-in
Mainframe
Database as a service
ServerLaptop Self-managed in the cloud

10-step Methodology For Private DBaaS
Step 1:
Common
Workload Reqs
Step 4:
Enabling
Multitenancy
Step 5:
Security
Enforcement
Step 6:
Performance &
Uptime SLAs
Step 7:
Provisioning &
Mgmt
Step 8:
Implementation
Step 3:
Virtualization
Strategy
Step 2:
Hardware & OS
Selection
Step 9:
Cost Accounting &
Chargeback
Step 10:
Design Reviews
Discovery
Design
Deploy

MongoDB Ops Manager
Management Platform For Private DBaas
Monitoring
• Deploy, resize,
and upgrade
your
deployments
with just a few
clicks
• RESTful API to
integrate with
your enterprise
orchestration
tools
• Allocate and
create pre-
provisioned
server pools;
Cloud Foundry
Integration
Automation Backup
• Continuous
backups to
minimize your
exposure to data
loss
• Restore to
precisely the
moment you
need with point-
in-time recovery
• Dozens of charts
tracking key
performance
indicators
• Custom alerts
that trigger when
key metrics are
out of range
• RESTful API to
integrate with
your existing
APM tools

From Traditional To DBaaS
• Slow to build and launch
new applications
• Multiple copies of data
• Complex data
reconciliation controls
• High licensing costs
• Sprawling server estate
12,000+ RDBMS Instances
3,500 Systems, 40,000 Cores
1,200+ Coherence Instances

Data Fabric
Multi-tenant PaaS
Exposing APIs for data streaming and
storage
Cloud native, self-service
Modern, industry standard, open source
technologies
Intra-day releases
Multi-data center
“Data Fabric provides data storage, query and distribution as
a service, enabling application developers to concentrate on
business functionality.”
Data Fabric Clients
Java, .NET, REST
API Layer
CRUD & Streaming
App Server Layer
Java + Linux
Database
MongoDB
Messaging
Kakfa
Security
Authentication&Audit
Data Fabric

Results
• £m license cost avoidance (Coherence)
• Plans to decommission hundreds of servers
• Coherence
• Oracle/SQL Server databases
Cost
Reduction
• 2 foundational applications refactored off
Coherence
• Supporting data needs of a dozen applications
Simplification
• Velocity: Develop new applications in days
• No need for database administration
• self-service data service
• Promotes collaboration and data sharing
Velocity
Slide taken from https://www.mongodb.com/presentations/mongodb-days-uk-building-an-enterprise-data-fabric-at-royal-bank-of-scotland-with-mongodb

20
You have 2 choices: Self-Managed or DBaaS
Cloud Migration
or Cloud First
Self-Managed
Aka “Lift and Shift”
Database as
a service
Fork in
the road

21
Cloud Migration
or Cloud First
Self-Managed
Database as
a service
Fork in
the road
1. Provision instances and storage
2. Configure HA
3. Configure security
4. Configure backup/restore
5. Monitoring & alerting
6. Ongoing upgrades & maintenance

22
Cloud Migration
or Cloud First
Self-Managed
Database as
a service
Fork in
the road
2. Configure HA
Choose instance, hit deploy,
wait a couple of minutes

MongoDB Atlas
On-Demand, Elastic, Fully Managed
• Run for You
• Resilient & Secure
• Multi-Cloud

24
MongoDB Atlas unlocks agility & reduces cost
Self-service, elastic,
and automated
Global and highly
available
Secure by default
Comprehensive
monitoring
Managed disaster
recovery
Cloud agnostic

MongoDB Atlas Powering
Microservices Architecture
Problem Why MongoDB ResultsProblem Solution Results
Over 35 different apps accessed by
10,000+ unique customers on AWS
Each experiment produces millions of
“rows” of data, which led to suboptimal
performance with incumbent databases
RDS & Aurora slow, added code
complexity
DynamoDB limited query functionality &
expensive
MongoDB Atlas managed database
service
Flexible document model allows
storage of multi-structured data
Expressive query language and
secondary indexes allow ad-hoc
analysis of experiment data
Scalability to handle growing data
volumes
Thermo Fisher customers now obtain
real-time insights from mass
spectrometry experiments from any
mobile device or browser; not possible
before
Improved developer productivity with
40x less code, improved performance
by 6x
Easy migration process & zero
downtime. Testing to production in
under 2 months

Cloud Data Strategy:
Wrapping Up

Conclusion
1 Cloud is more than just a new
platform: microservices, agile,
DevOps enable new ways of
delivering apps faster
2 Cloud data strategy is an
essential building block
3 MongoDB is the foundation for
cloud data

Resources To Get Started
Spin up a cluster on the
Free Tier today
Download the Whitepaper

MONGODB 4.0:
MULTI-DOCUMENT
ACID TRANSACTIONS

SAFE HARBOUR STATEMENT
This presentation contains “forward-looking statements” within the meaning of Section 27A of the
Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as
amended. Such forward-looking statements are subject to a number of risks, uncertainties, assumptions
and other factors that could cause actual results and the timing of certain events to differ materially from
future results expressed or implied by the forward-looking statements. Factors that could cause or
contribute to such differences include, but are not limited to, those identified our filings with the
Securities and Exchange Commission. You should not rely upon forward-looking statements as
predictions of future events. Furthermore, such forward-looking statements speak only as of the date of
this presentation.
In particular, the development, release, and timing of any features or functionality described for
MongoDB products remains at MongoDB’s sole discretion. This information is merely intended to
outline our general product direction and it should not be relied on in making a purchasing decision nor
is this a commitment, promise or legal obligation to deliver any material, code, or functionality. Except
as required by law, we undertake no obligation to update any forward-looking statements to reflect
events or circumstances after the date of such statements.

MongoDB: Already Used Across Every Industry and
Use Case...
eCommerce
Travel Graph &
Recommendation System
Analytics
Internet of Things
Product Catalog
Artificial Intelligence
Digital Transformation
Drug Sequencing
Gaming
Single View
Mobile
Database as a Service
...without multi-document ACID transactions

Relational Database
Related data split across multiple records and tables.
Multi-record transactions essential
Different databases take different approaches
Document Database
Related data contained in a single, rich document.
Transaction scoped to the document
Data Models and Transactions

Just like relational transactions
● Multi-statement, familiar relational syntax
● Easy to add to any application
● Multiple documents in 1 or many collections
ACID guarantees
● Snapshot isolation, all or nothing execution
● No performance impact for non-transactional
operations
Schedule
● MongoDB 4.0, Summer ‘18: replica set
● MongoDB 4.2: extended to sharded clusters
MongoDB Multi-Document Transactions

Major engineering investment over 3+ years touching every part of
the server and drivers
● Storage layer
● Replication consensus protocol
● Sharding architecture
● Consistency and durability guarantees
● Global logical clock
● Cluster metadata management
● Exposed to drivers through API enhancements
We’re 85% done…...
Our Journey to Acid Transactions

SUMMARY: MongoDB UNIQUELY Delivers…...
Scale-out, data locality, and
resilience of distributed systems
ACID transactional guarantees
of relational databases
Developer productivity
of document databases
Freedom to Run Anywhere

NEXT STEPS
Sign up for the
beta program
www.mongodb.com/transactions

EXECUTING ON A CLOUD DATA STRATEGY
Bart Van Loocke, Software Engineer
De Persgroep

38
Customer case study
De Persgroep

39
▪ Application Developer
@DePersgroep
▪ Freelance Developer
Bart Van Loocke

40
Why MongoDB Atlas
➔ Move to the cloud
➔ De Persgroep Engineering Culture /Principles & Practices

43
De Persgroep engineering culture

Autonomous Squad
➔ Small, co-located, self organised
➔ End-to-end responsibility for the stuff
they build, from design to
maintenance.
➔ Within the scope of its mission, a
squad is empowered to decide what to
build, how to build it and how to work
together while doing it.

45
Autonomous Squad
Each squad chooses his own technology
How does De
Persgroep do X?
Depends on
the Squad

49
➔ 24/7 support for production issues.
➔ Support the Service desk
➔ Fix recurring issues
➔ Performance monitoring and tuning

58
On Premise MongoDB
▪ MongoDB 2.6
▪ 1200GB
▪ 85 Databases
▪ MongoDB 3.2
▪ < 50GB
▪ 80 Databases
Upgrade was imposible: too many squads/applications involved!
▪ ...

59
MongoDB 2.6 on-premise > Cloud
Problem:
no migration tools available (in 2017)
Some different strategies:
➔ copy data
➔ duplicate data
➔ incremental update
➔ ..

60
Copy data @deploy time
DB 1
Application
DB 1
Application
DB 2
DB 2
Application
copy data
Small applications with downtime

61
Duplicate data @runtime
DB 1
Application
DB 1
Application
DB 2
DB 2
Application
Applications with mostly CREATE actions

62
Incremental update @deploy time
DB 1
Application
DB 1
Application
DB 2
DB 2
Application
Applications with mostly UPDATE actions
DB 2
copy data
{migrated : ‘N’}
migrate
d
Y
migrate
d
Y

64
Migration of clusters inside Atlas
Atlas Data Migration

66
Isolation
Originally the goal was to have a cluster / application / environment.
However this is not a feasible strategy > high costs
DEV/TEST/ACC
APP A APP B APP C
DEVTESTACC
APP A APP B APP C
APP A APP B APP C
Cluster
PROD
APP A APP B APP C
PROD
Cluster
2 Clusters per domain (non-prod and prod)

67
CPU Steal
M10 and M20 instances are running on t2-family machines (AWS)
Example: M10 > t2.small > 12 credits/hour > 20%

69
Some Numbers
Squads using MongoDB Atlas: 22
#Clusters: 92
DB Size: 2950GB
Data Size: 2070GB

70
Some Numbers
Instances:
M2: 2
M10: 53
M20: 17
M30: 19
M40: 1
versions:
3.2: 4
3.4: 81
3.6: 7

LEVERAGING MONGODB ATLAS
Praveen Bisht
Infosys Limited

TOYOTA MOTOR EUROPE AND INFOSYS
30th May 2018
75
An example case of leveraging MongoDB Atlas to deliver high quality,
predictable and scalable database services..

TOYOTA IN EUROPE
• Began selling cars in 1963
• 9 manufacturing plants in 7countries
• Network of 30National Marketing and Sales Companies in 53 countries.
• Network of over 3000Toyota and Lexus authorised retailers.
• Directly employ over 20,000people across Europe.
• In 2017 sold 1,001,700vehicles in Europe
• 45%of Toyota Motor Europe (TME) sales in 2018-Q1 are hybrid electric vehicles

INFOSYS IS A ‘$10BN+’ GLOBAL, SCALABLE AND DIVERSE
ORGANIZATION WITH ‘200,000+ EMPLOYEES’ AND ‘1,200+ CLIENTS IN
45 COUNTRIES’..
77
• 97.6% of our revenues come
from existing customers
• Shared business vision and
commercial objectives
• 97% of our projects are delivered in
time and to budget
Customers
are the anchor for
everything we do, and the
metric of our success
Predictability
helps our customers achieve
programme objectives in time
and within budget
• Investment in purposeful AI based platforms to
drive automation
• Innovation to deliver change through DT and
ZD
Innovation
$500 million innovation
fund

INFOSYS RELATIONSHIP WITH TOYOTA
78
Toyota Motor North America - Since 2007
Toyota Motor Europe- Since 2009
Toyota Kirloskar Motor , India – Since 2008
Toyota Financial Services - Since 2010
Toyota Great Britain - Since 2013
Toyota Financial Services, UK - Since Jan 2017
Toyota Sweden - Since 2016
Long term strategic partnership with global overview across 3 continents
A multi-discipline, multi geography relationship which has matured over the last nine years from a
projects based engagement to a strategic partnership across Europe, North America and Asia
Toyota Connected - Since 2016
Toyota Motor Manufacturing France - Since 2016

MONGODB FOOTPRINT IN TOYOTA MOTOR EUROPE
79
15+ B2B & B2C Applications
45 Toyota Websites
39 Lexus Websites
50+ Environments

HOW DID WE END UP WITH MONGODB ATLAS
80
1. Features
2. Support
3. Cost
4. Migration Effort
Evaluation criteria

ATLAS MIGRATION – OUR APPROACH
Mongodb 2.6 cluster on mlabs
Upgrade in-place
from 2.6 to 3.0
on mlabs
Mongodb 3.0 cluster on
mlabs/Heroku
Option 1 - Migrate from mlabs 3.0 to
Atlas 3.2 (using MongoMirror)
Mongodb 3.2 cluster on Atlas
Application remediation/testing for upgrade
to 3.0
1 2
• Switch DB connections to Atlas clusters
• Application testing for upgrade to 3.2
Option 2 – Export data files from mlabs 3.0 to temp env and
import to Atlas 3.2

MONGODB ROLE IN CURRENT TOYOTA MOTOR EUROPE CLOUD
ECOSYSTEM
Microservice
Microservice
Microservice
Microservice
Microservice
Microservice
Microservice
MicroserviceKafka
Connectors
Schema
Registry
Microservice
Microservice
Microservice
Microservice
Public Cloud 1 Public Cloud 2

83
THANK YOU
© 2018 Infosys Limited, Bengaluru, India. All Rights Reserved. Infosys believes the information in this document is accurate as of its publication date; such information is
subject to change without notice. Infosys acknowledges the proprietary rights of other companies to the trademarks, product names and such other intellectual property
rights mentioned in this document. Except as expressly permitted, neither this documentation nor any part of it may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, electronic, mechanical, printing, photocopying, recording or otherwise, without the prior permission of Infosys Limited and/ or any
named intellectual property rights holders under this document.

LANDING IN THE CLOUD
MIGRATING FROM RELATIONAL, ON PREMISES TO MONGODB IN THE CLOUD
Eugene Bogaart, Senior Solutions Architect
eugene@mongodb.com

Migrating from Relational, On Premises, to MongoDB in the Cloud.

DATA MODELS DIFFERENCES
Relational Database
Document Database

RECIPE: STEP 1
• Design Data Model,
○ New Model, requires understanding of the data
○ Application Impact, simplify data access and ORM

88
Developer like ORM's
Thousands
of lines of
configuration code….
Total pain in the...
Customer
Opportunit
y
Contact
ARR Address Contact Roles
Opportunity
Team
Phone Phone
Objects
Tables
Lead
NameName
Activity
History
Open
Activities
Customer
Detail
Summary
Object Relational Mapping Layer...which needs to be
updated whenever
the schema
changes

89
With MongoDB, without ORM
Customer
Customer
Opportunity
Opportunity
Contact
Contact
Lead
Lead
Objects
Database

RELATIONSHIP REMODELING
Design Decision
Embedding
Referencing
1. Top Down vs. Bottom Up
2. Use Case / Access Patterns
3. Entity Relationships
4. Cardinality & Direction
5. MongoDB Document Size

CONVERT DATA
Relational Database
Document Database

RECIPE: STEP 2
• Design Data Model,
○ New Model, requires understanding of the data
○ Application Impact, simplify data access and ORM
• Convert Data
○ Bulk, merge data in new model
○ Change Data Capture, if large dataset involved which
cannot be offline
○ Custom, opportunity to do some data cleansing

HOW TO GET INTO THE CLOUD?
Cloud Migration
or Cloud First
Self-Managed
Database as
a service
Fork in
the road
2. Configure HA
Choose instance, hit deploy,
wait a couple of minutes
“You have 2 Choices…”

HOW TO GET INTO MONGODB ATLAS?
Migrate existing deployments running
anywhere into MongoDB Atlas with minimal
impact to your application. Live migration
works by:
● Performing a sync between your source
database and a target database hosted in
MongoDB Atlas
● Syncing live data between your source
database and the target database by
tailing the oplog
● Notifying you when its time to cut over to
the MongoDB Atlas cluster
Live Migration

RECIPE: STEP 3
• Migrate data to Cloud,
○ MongoDB Altas Live Migration
■ Wizard run from destination
○ MongoDB Mirror
■ Run from source

Accelerating a Path to Digital With a Cloud Data Strategy

DEMO CONTEXT
Existing Database in Tabular database with employee information.
Contains, past and present information with managers, departments, titles and salary.
Main application is a for managers who can view who reports to them
And view company history of their direct (and indirect) reports.

○ New Data Model
○ Convert Data
○ Altas Live Migration
DEMO SETUP

○ New Data Model
○ Convert Data
○ Altas Live Migration
WRAP UP DEMO

• Design Data Model -> shift in paradigm
• Convert you data -> some manual work
• Migrate to MongoDB Atlas -> tooling provided

SUMMARY: MONGODB UNIQUELY DELIVERS…...
Scale-out, data locality, and
resilience of distributed systems
ACID transactional guarantees
of relational databases
Developer productivity
of document databases
Freedom to Run Anywhere

Morphia
MEAN Stack
Support for Popular Languages and Frameworks
DRIVERS & ECOSYSTEM

MONGODB CONNECTOR FOR BI
• Run analytical queries on live
MongoDB data
• Can run against your primary or
secondaries
• Runs on any server, or co-locate with
your database process for optimal
performance

HOW TO GET OFF THE GROUND
3rd Party Tools

All MongoDB Atlas nodes are single-tenant and
deployed into their own VPC for security isolation.
In-flight security:
● TLS/SSL for in-flight data encryption
● Authentication and authorization access controls
with SCRAM-SHA1
● IP whitelists
At-rest security:
● Encrypted storage volumes
● AES-256 (CBC mode) hardware encryption with
Seagate Self-Encrypting Drives
Under the hood
SECURE IN THE CLOUD

MongoDB 4.0: Multi-Document ACID
Transactions

Under the hood
Out-of-the-Box Operational Tooling
● Fine Grained Monitoring and Alerts
● Real-Time Performance Panel
● Data Explorer
● Query/Performance Optimization
● Continuous Backup/Point-in-Time
Restore
● Query-able Backup
● Enterprise APM Integration
● Programmatic API Management

Highly Available by Default
Under the hood
● A minimum of three data nodes per
replica set/shard are automatically
deployed across availability zones for
high availability
● If your primary node does go down for
any reason, the self healing recovery
process in MongoDB Atlas will typically
occur in under 2 seconds

The Latest MongoDB Features
Under the hood
● MongoDB Atlas comes out-of-the-box with
MongoDB 3.2 and MongoDB 3.4 available. When
maintenance releases become available, MongoDB
Atlas will automatically upgrade your cluster, using a
rolling process to maintain cluster availability at all
times
● When new versions of MongoDB are released,
MongoDB Atlas will also allow you to make
seamless upgrades without risking downtime

ELASTIC SCALABILITY & ON-DEMAND
PROVISIONING
Under the hood
● WiredTiger storage engine with compression and fine-
grained concurrency control to meet the most demanding
SLAs
● MongoDB Atlas supports automatic-sharding, giving you the
ability to scale up or out with no impact to your app
● MongoDB Atlas allows you to pick the shard key that best
suits your application needs
● All MongoDB Atlas clusters are single-tenant and deployed
on servers allocated specifically to the cluster

Integrated services and
functions for complex,
multi-stage workflows
Native SDKs for
Android, JS, and iOS
apps
Direct Database
Access
MongoDB Stitch

Accelerating a Path to Digital With a Cloud Data Strategy

Recommended

More Related Content

What's hot (20)

Similar to Accelerating a Path to Digital With a Cloud Data Strategy (20)

More from MongoDB (20)

Recently uploaded (20)

Accelerating a Path to Digital With a Cloud Data Strategy