Why MongoDB

Architecture and Technology Overview

Rhiannon Ludington | Account Executive | [email protected]

Matt Davis | Senior Solution Architect | [email protected]
 ● Intro to the MongoDB Data Platform

● The Document Database model

Agenda
● Distributed Architecture

● Modern Database Platform

● How to get started

Anthem has partnered with MongoDB since 2017 in order to increase the
pace of innovation in the buildout of new, mission critical applications.

MongoDB PS provided
application lifecycle best 65 Highlights
practices,design guidance Applications developed ● Enterprise-wide MongoDB center of
in 12 months excellence was established
Professional Services
provided training for DBAs,
developers and architects
25+ ● PS helped Anthem to realize its cloud
development strategy
Development teams
● Custom training programs and
enabled
application reviews enabled
MongoDB PS led the developers and architects to build
development of MongoDB 3 months → 46s more scalable and reliable projects,
faster
center of excellence Average cluster provision
time
Developers Biggest Problems Are Data Problems

Rigid, complex relational Monolithic, fragile relational Niche NoSQL data stores
data models inhibit systems fail to deliver arrived as bandaids, but
application evolution resilience and scale discard essential
developer functionality
 MongoDB Developers love
PostgreSQL using MongoDB
Elasticsearch
Intuitive and flexible
Redis document data model

Firebase

Powerful query engine and

Most Wanted Database for 4 Years idiomatic data API

2020 Stack Overflow Developer Survey

Distributed systems architecture

for resilience & scale-out
At the core is the database for modern applications

● Transactional guarantees at a
global scale
● Intuitive and flexible data model
● Unique data distribution
capabilities
● MongoDB Query Language (MQL)
is built for nearly any workload

Transactional Intuitive & Flexible Data Model Distributed

We fully manage it for you in the cloud

● Fully managed database lifecycle with

MongoDB Atlas.
● Multi-cloud, available in ~ 80 regions
across AWS, GCP, Azure
● Sophisticated security controls and
next-gen end user privacy
● Autopilot features such as auto-scale,
performance advisor, and more
● Built-in data access, movement,
manipulation services for rapid
application development
And support for data at the edge

● Persist data at the edge or on

mobile devices with Realm, the
world’s fastest mobile database
● Keep data between Realm and
MongoDB Atlas in sync with
Realm Sync, the edge-to-cloud
synchronization service with
built-in conflict resolution
Interactive data visualization for MongoDB data

● MongoDB Charts is the fastest

and easiest way to create
visualizations of MongoDB data
● Share, embed, and collaborate
on live data
● Support for the richness of
document model, including
nested and hierarchical data
Integrated full-text search capabilities

● Atlas Search allows you to

implement full-text search on
top of your data in cloud with no
need to replicate your data
elsewhere and no additional
systems to learn or manage
● Atlas search queries use the
MongoDB Query Language
Tier, query and analyze your data using MQL

● Auto-archive aged data into Atlas Data

Lake
● Blend, query, and analyze the
structured / unstructured data in your
cloud object storage using the
MongoDB Query Language
● Support for federated queries means
you can submit a single query and
analyze operational data in MongoDB
Atlas alongside your data in S3
An integrated data layer that runs on all clouds

● Flexible data model & natural query

language so teams can focus on
building
● An integrated & consistent experience
for both developers and operations
● Transparent data movement — sync
and replication — across data services
● Multi-cloud infrastructure; deploy on
GCP, AWS, and Azure
Integrations with your most strategic investments

● Seamless integrations with your

technology stack —
● Infrastructure as code:
Terraform, Kubernetes
● IDEs: VS Code, Jetbrains
● Monitoring: Datadog
● And much more…
 The Evolution of MongoDB
Stack Overflow Survey: Most wanted database
3.0 & 3.2 3.4 3.6 4.0 4.2 4.4 5.0

2015 2016 2017 2018 2019 2020 2021

Doc-Level Concurrency Views Change Streams Replica Set Transactions Distributed Transactions Union Time-Series collections
RAFT / Fast Failover Graph Processing Retryable Writes Atlas Global Clusters Client-Side Field Level Custom Agg Expressions Clustered indexes
$lookup Zones ++Aggregation ++ Schema Validation 40% Faster Shard Migrations Encryption Refinable Shard Keys Window functions
Materialized Views Compound Hashed Shard Keys New temporal operators
Ops Manager Auto-balancing ++ Expressive $lookUp Atlas HIPAA
Wildcard Indexes Mirrored Reads Live resharding
Compression Linearizable Reads Query Expressivity Atlas LDAP
Global PIT Reads Hedged Reads Client-Side FLE multi-cloud KMS
≤50 replicas Decimal Causal Consistency Atlas Audit Large Transactions Resumable Initial Sync Long running snapshot reads
Aggregation ++ Intra-cluster compression Consistent Sharded Atlas Enc. Storage Engine Mutable Shard Key Values Time-Based Oplog Retention Atlas Serverless preview
Encrypted and In-Memory Log Redaction Secondary Reads Atlas Backup Snapshots Atlas Data Lake (Beta) Simultaneous Indexing Majority write concern default
storage engines Spark Connector ++ Query Advisor Type Conversions Atlas Auto Scaling (Beta) Hidden Indexes Schema validation diagnostics
BI Connector BI Connector ++ End to End Compression Snapshot Reads Atlas Search (Beta) Streaming Replication New MongoDB Shell GA
Compass Compass ++ WiredTiger 1m+ Collections Non-Blocking Sec. Reads Multi-CAs Global Read/Write Concerns Resumable index builds
MongoDB BI Connector ++ Expressive Updates Rust & Swift Drivers GA Rewritten Swift driver
Hardware Monitoring SHA-2 & TLS 1.1+
Apache Kafka Connector TLS 1.3 & Faster Client Auth Rewritten C# LINQ provider
WiredTiger LDAP Authorization R Driver Compass Agg Pipeline Builder
MongoDB Charts GA OCSP Stapling PyMongoArrow API
Encrypted Backups Charts (post GA) Compass Export to Code Retryable Reads & Writes Kerberos Utility Charts on Data Lake
Atlas X-Region Replication
Acquisition Atlas Auto Storage Scaling
Free Monitoring Cloud
Service
New Index Builds Atlas Online Archive x509 certificate rotation
10x Faster stepDown Auto-Scaling Auditing ++
+ Integration Ops Manager K8s Beta Storage Node Watchdog Schema Recommendations Atlas K8S controller
Zstandard Compression AWS IAM Auth & Atlas x509 Ops Manager 5.0
ACID Federated Queries Ops Manager migration wizard
Atlas Search function scores &
MongoDB Atlas Transactions synonyms

Versioned API
Rapid Releases
 Intuitive: Contrasting data models

{
"_id" : ObjectId("5ad88534e3632e1a35a58d00"),
"name" : {
"first" : "John",
"last" : "Doe" },
"address" : [
{ "location" : "work",
"address" : {
"street" : "16 Hatfields",
"city" : "London",
"postal_code" : "SE1 8DJ"},
"geo" : { "type" : "Point", "coord" : [
51.5065752,-0.109081]}},
+ {...}
],
"dob" : ISODate("1977-04-01T05:00:00Z"),
"retirement_fund" : NumberDecimal("1292815.75")
}

Tabular (Relational) Data Model Document Data Model

Related data split across multiple records and tables Related data contained in a single, rich document
 Documents are Universal
JSON Documents are the modern standard in today’s application stacks

JSON Tabular Key-Value Text Geospatial Graph Time Series Events

Documents

Model and Query Data Any Way You Need

Point | Range | Geospatial | Rich Search | Aggregations | JOINs & UNIONs | Graph Traversals

All wrapped in a single API, giving a consistent experience for any workload
 Building FHIR Applications with MongoDB
{
"resourceType": "Condition",
"id": "example",
"clinicalStatus": {
"coding": [
Next-gen standards framework created by HL7 {
"system": "http://terminology.hl7.org/CodeSystem/condition-clinical",
"code": "active"
}
Supports JSON as de facto Standard },
]

"verificationStatus": {
"coding": [
{
Rich Data Structures "system": "http://terminology.hl7.org/CodeSystem/condition-ver-status",
"code": "confirmed"
}
]
},
"category": [
{
"coding": [
{
"system": "http://terminology.hl7.org/CodeSystem/condition-category",
"code": "encounter-diagnosis",
"display": "Encounter Diagnosis"
…….

https://www.mongodb.com/blog/post/building-fhir-applications-with-mongodb-atlas
 Fully Indexable
Fully featured secondary indexes - Document optimized - Extended beyond RDBMS experiences
Index Types Index Features
• Primary Index • TTL Indexes
– Every Collection has a primary key index – Single Field indexes, when expired delete the
document
• Compound Index
– Index against multiple keys in the document • Unique Indexes
– Ensures value is not duplicated
• MultiKey Index
– Index into arrays • Partial Indexes
– Expression based indexes, allowing indexes on
• Wildcard Index
subsets of data
– Auto-index all matching fields, sub-documents &
arrays • Case Insensitive Indexes
– Supports text search using case insensitive search
• Text Indexes
– Support for text searches. Atlas Search offers • Sparse Indexes
Lucene-based inverted indexes – Only index documents which have the given field
• GeoSpatial Indexes
– 2d & 2dSphere indexes for spatial geometries
• Clustered Indexes
– For time series collections, pre-sorted by timestamp
for low latency queries
 Aggregation Features
A feature rich framework for data transformation and Analytics

Pipeline Stages Operators

• $match • $lookup • Mathematical • Conditionals • String
– $add, $abs, – $and, $or, $eq, $lt, – $toUpper, $toLower, $substr,
• $group • $merge $subtract, $multiply, $lte, $gt, $gte, $strcasecmp, $concat, $split, etc.
$divide, $log, $log10, $cmp, $cond,
• $facet $stdDevPop, $switch, $in, etc • Literals
• $project $stdDevSam, $avg,
– $exp, $let, $literal, $map, $type, etc
• $geoNear $sqrt, $pow, $sum, • Temporal
• $search $zip, $convert,
$round, etc
– Window Functions • Regex
• $graphLookup – $dateAdd, $dateDiff, – $regexFnd, $regexMatch, etc
• $sort
$dateSubstract,
• Array $dateTrunc • Trigonometry
• $setWindowFields – $push, $reduce, – $dateFromParts,
$reverseArray, $dateToParts, – $sin, $cos, $degreesToRadians, etc
• $unionWith $addToSet, $dateFromString,
$arrayElemAt, $dateToString, • Custom Aggregation
• $unwind $slice, etc. $dayOfMonth, Expressions
$isoWeek, $minute,
• ...and more $month, $year, etc.

Practical MongoDB Aggregations eBook

MongoDB Multi-Document ACID Transactions

Multi-node transactional guarantees delivered at scale

• Multi-statement, familiar relational syntax
• Easy to add to any application
• Multiple documents in 1 or many collections and
databases, across replica sets and sharded clusters

ACID guarantees
• Snapshot isolation, all or nothing execution
 MongoDB Replica Sets
Replica Set – 2 to 50 copies

Application
Self-healing
• Typical failover in 5 seconds or less
• Retryable reads and writes to catch temporary exceptions
Driver

Data Center Aware, Tunable Durability and

Consistency
Primary

Addresses availability considerations:

Secondary Replication
• High Availability
Secondary • Disaster Recovery
• Maintenance

Workload Isolation: operational & analytics

 MongoDB Workload Isolation

Rich MongoDB Query API and distributed Analytics

Transactional architecture allows you to run both Consumers
Applications Transactions and Analytics on the same
cluster with no resource contention.

Primary Secondary Secondary Secondary Secondary

{Analytics Node} {Analytics Node}

23
 Sharding Architecture
Application

Driver

••• Mongos Mongos •••

Shard 1 Shard 2 Shard N

High availability Primary Primary Primary
- Replica sets
Secondary Secondary ••• Secondary

Secondary Secondary Secondary

Horizontal scalability
- Sharding
 Secure wherever you put your data
All MongoDB Atlas customer projects are deployed into their own VPC for network
isolation. Private network peering is available for databases on all three clouds.

Business Trust Needs Security Features

Authentication SCRAM, X.509, LDAPS, AWS IAM, IP Access Lists

Authorization RBAC, Read-Only Views, Field-Level Redaction

Auditing Admin, DML, DDL, DCL, Role-based

In-Flight: TLS 1.2+
At-Rest: Encrypted Hardware, Volume and Database Storage Engine
Encryption (AES-256)
In-Use: Client-Side Field Level Encryption
Key Management: Cloud Services (AWS KMS, Azure KV, Google Cloud KMS)
Compliance
 Broadest Reach: Private, Hybrid, Public
Self-managed
Laptop Mainframe Server in the cloud Database as a Service

Self-Managed: Ops Mgr Database as a Service

• Automation • Self-service & fully automated
• Monitoring Alerting • Global, elastic multi-cloud clusters
• Backup/Restore • Highly available and secure out of the box
• Patching • Operational tooling built in: monitoring, backups
• Scaling
• Performance Advice • On AWS, Azure, GCP, 80+ regions total
• Kubernetes Operator

Complete optionality: same codebase, same APIs, same tools, wherever you run
 Lightweight, fully-featured
database that’s embedded on
the device
Object-oriented database that developers
find easy and intuitive to work with

Easier for companies to enhance app design, build

new features, and deliver a good user experience
— creating a competitive advantage

On-device database allows developers to move

Realm Mobile compute to the mobile device, which reduces

server-side costs

Database Database can be encrypted with a single API call

 A platform that does the
heavy-lifting
MongoDB Realm’s platform provides a layer of
abstraction when syncing data between the client
and server.

The datastore is always up-to-date meaning fewer

syncing errors. Apps work when offline.

Automatic Conflict Resolution means no need to write

complex conflict conflict resolution code or
networking code.

Pre-set rules provide out-of-the-box conflict

Realm Sync resolution

Automatic Schema Generation means no need to

translate your data from MongoDB to Realm with
schema mapping.
 ● Build fast, relevant, full-text search on
top of your data in the cloud with
Atlas Search
● Create powerful search queries that
take advantage of the rich MongoDB
aggregation framework
● By integrating search into your data
layer, you can eliminate costly
replication and simplify your
application infrastructure
● Atlas Search is available for all
Atlas Search clusters running MongoDB 4.2+
Deployment on your terms
 MongoDB Kafka Connector

Kafka Cluster MongoDB Database

Receives events from Writes documents
Kafka Topic(s) to DB collection

SINK: topicA MongoDB

topicB Connector

topicC

MongoDB Database Kafka Cluster

Receives documents Writes events to
from DB collection Kafka Topics(s)

SOURCE: Change MongoDB

topicA

Streams Connector topicB

topicC
cloud.mongodb.com
 MongoDB University
MongoDB University has had over 1.5 million registrations from 196 countries and 13,000+ organizations.
Our courses address needs throughout the project lifecycle. We provides every member of your
organization with the training they need, by role. From basic fundamentals to advanced deep dives, your
team will receive the training they need.
Developer Learning Path

Online Courses: Learning Paths:

● 24x7 access globally ● Streamlined curricula
● Self-paced for different roles
● Hands-on exercises ● Take the guesswork
● In-depth & up-to-date out of learning for
your team

Certification: Business Dashboard: Operations Learning Path

● Team speak a ● Measure individual
common language and group
● Continuing performance with
professional growth course analytics and
● Improve productivity reporting.
 Summary: MongoDB Uniquely Delivers…

Unified
Fastest way Freedom and experience
to innovate flexibility for modern
applications