MongoDB World 2018: Ch-Ch-Ch-Ch-Changes: Taking Your Stitch Application to the Next Level With Triggers

Ch-ch-ch-ch-changes
Taking Your Stitch Application to the Next Level
with Stitch Triggers

Industry Trends – Leading to Stitch
1370 1990

19901370 1997

19971990 2001

20011997 2007

20072001 2014

20142007 2018

20182014

Intelligent Operational Data Platform
Best way to work
with data
Intelligently put data
where you want it
Freedom to run
anywhere
● MongoDB Stitch
● MongoDB Server 4.0
● MongoDB Compass
● MongoDB Charts
● MongoDB Mobile
● MongoDB Server 4.0
● MongoDB Atlas
● Ops Manager 4.0
● Free Cloud Monitoring

The evolution of MongoDB
3.0 3.2
Document Validation
$lookup
Fast Failover
Simpler Scalability
Aggregation ++
Encryption At Rest
In-Memory Storage Engine
BI Connector
MongoDB Compass
APM Integration
Profiler Visualization
Auto Index Builds
Backups to File System
Doc-Level
Concurrency
Compression
Storage Engine API
≤50 replicas
Auditing ++
Ops Manager
Linearizable reads
Intra-cluster compression
Views
Log Redaction
Linearizable Reads
Graph Processing
Decimal
Collations
Faceted Navigation
Zones ++
Aggregation ++
Auto-balancing ++
ARM, Power, zSeries
BI & Spark Connectors ++
Compass ++
Hardware Monitoring
Server Pool
LDAP Authorization
Encrypted Backups
Cloud Foundry Integration
3.4 3.6
Change Streams
Retryable Writes
Expressive Array Updates
Query Expressivity
Causal Consistency
Consistent Sharded Sec. Reads
Compass Community
Ops Manager ++
Query Advisor
Schema Validation
End to End Compression
IP Whitelisting
Default Bind to Localhost
Sessions
WiredTiger 1m+ Collections
MongoDB BI Connector ++
Expressive $lookUp
R Driver
Atlas Cross Region Replication
Atlas Auto Storage Scaling
4.0
Multi-Document ACID Transactions
Atlas Global Sharding
Atlas HIPAA
Atlas LDAP
Atlas Audit
Atlas Encrypted Storage Engine
Atlas AWS Backup Snapshots
Snapshot Reads
Agg Pipeline Type Conversions
40% Faster Shard Migrations
Non-Blocking Secondary Reads
SHA-2
TLS 1.1+
Compass Agg Pipeline Builder
Compass Export to Code
Charts Beta
Monitoring Cloud Service
Ops Manager K8s & OpenShift
MongoDB Stitch GA
MongoDB Mobile (Private Beta)

“We set out to build a database that we would want to use, so that
whenever developers wanted to build an application, they could
focus on the application, not on working around the database.”
- Eliot Horowitz

Stitch Functions
Stitch is a collection of servers that
process application requests
Requests:
• Single actions for Database or Services
• Or executing a Stitch Function
• Integrated with Stitch’s rules
Functions:
• Scalable, hosted JavaScript (ES6) Functions
• Integrated with application context
• User, Request, Services, Values, etc.

MongoDB Query Language + Native DriversIntegrated Rules
Functions3rd Party Services
Native SDKs (JavaScript, Android, iOS)
Rest API

MongoDB World 2018: Ch-Ch-Ch-Ch-Changes: Taking Your Stitch Application to the Next Level With Triggers

When do we want functions to execute?
AUTHENTICATION
WEBHOOK/
INTEGRATION
DATABASE EVENTS
3 Contexts Where Functions
Execute
1
2
3

client.executeFunction(…)client.loginWithCredential(…)
AUTHENTICATION
WEBHOOK/
INTEGRATION
DATABASE EVENTS
1
2
3

Incoming Webhooks
(HTTP, Twilio, Github)
AUTHENTICATION
WEBHOOK/
INTEGRATION
DATABASE EVENTS
1
2
3

Database Events
Collection.insertOne(…)
AUTHENTICATION
WEBHOOK/
INTEGRATION
DATABASE EVENTS
1
2
3

Responding to Database Changes in the Past

Responding to Database Changes in the Past
Before Change Streams:
{
"ts" : Timestamp(1526593721, 1),
"t" : NumberLong(1),
"h" : NumberLong("-2175450131096164196"),
"v" : 2,
"op" : "n",
"ns" : "",
"wall" : ISODate("2018-05-17T21:48:41.962Z"),
"o" : {
"msg" : "periodic noop"
}
}
{
"ts" : Timestamp(1526593731, 1),
"t" : NumberLong(1),
"h" : NumberLong("4757850941625744949"),
"v" : 2,
"op" : "n",
"ns" : "",
"wall" : ISODate("2018-05-17T21:48:51.962Z"),
"o" : {
"msg" : "periodic noop"
}
}

OPLOGOPLOG OPLOG
Application
W
db.users.insert({,,,})
PRIMARY SECONDARY SECONDARY
A
C
K
W
db.users.update({,,,}) db.users.update({,,,})
A
C
K
db.users.update({,,,})
What’s an Oplog?
OUT OF DATE OUT OF DATE

OPLOGOPLOG OPLOG
Application
W
PRIMARY SECONDARY SECONDARY
A
C
K
W
db.users.update({,,,}) db.users.update({,,,})
A
C
K
db.users.update({,,,})
Monitor
R R
If (insert) {
aws.s3.insertObject(…);
}
Before ChangeStreams and Triggers
TAILING THE OPLOG

Setting up a Change Stream (NodeJS)

Traditional Database Triggers
• Triggers are:
• Code executed when a certain database
event occurs
• Executed within the Database
• Trigger uses include:
• Enforcing Referential/Business logic
• Auditing and Logging
• Security/Validation Logic
• Maintaining Replication

Stitch Coordinates Change Streams
1. Sets-up Change Stream
2. Maintains Streams over time
3. Passes Change Events to
Coordinator
Stitch Infrastructure
Stitch Connection
Event Coordinator
CS 1 CS 3
CS 2

Adding Triggers to Stitch
Reduce Time to Market
Functions as a Service: No waiting on
infrastructure
Reduce boilerplate: Save code
Cross-Platform: Develop once
Existing Apps Untouched: No reverse-
engineering
Reduce Ops costs
Serverless: Zero wasted capacity
Payment: No up-front cost
Capacity on Demand: No need to over-
provision in advance
Secure by Default: Less operational effort
Reduce Dev Effort
Backend Integration Built in: No generic
backend code
HA & Scalability Built in: Hard and time
consuming to build Orchestrate Services:
Reuse what's out there
Faster Cheaper Easier

Event Driven Life-cycle
Pending SucceededOwned
Failed Dead
Event is …
Waiting for a ConsumerAttempting ExecutionCompletedAwaiting RetriesRe-Attempting ExecutionUnable to Execute
(Trigger Disabled)

Introducing: Stitch Triggers
Add Trigger
Add Trigger

Event-driven Functions for Communication
FUNCTION
Database
Update
Text/E-mail
EVENT
Push Notification

Event-driven Functions for Data Propagation
FUNCTION
EVENT
Database
Operation

Stitch Ensures Correct Execution
Database Events are:
• Generated by matching
operations ({match: { …}})
• Fully Ordered
1. Event is Generated
2. Event added to Job Queue
3. Event waits for blocking Events
to complete
4. Event waits for consumer to
attempt execution

What is the Swagstore
Standard Retail Store:
• Browse Items
• Add items to cart
• Checkout
• Request Notification for Restock
Additional Features:
• Update/text on item re-stock
• Shipping text/e-mail notification
Stitch Features
• Functions
• Triggers
• 3rd Party Services

Inventory Updates
FUNCTION
EVENT
{
ID: XXXXX
SIZE: M
COUNT: 100
}
{
ID: XXXXX
DESC: “Men’s T-shirt”
SM_COUNT: 12
MED_COUNT: 100
LG_COUNT: 23
}

Shipping Updates
FUNCTION
Shipping Status
Updated
Your package
has shipped
EVENT

AUTHENTICATION
BROWSE
SELECT ITEM
IN
STOCK?
CART
CHECKOUT
UPDATE NOTIFY
Initialization
• Before we leverage any features of Stitch,
we must initialize our API with the Stitch
App Key
import React from 'react'
import ReactDOM from 'react-dom'
import './index.css'
import App from './App'
import registerServiceWorker from './registerServiceWorker'
import { BrowserRouter } from 'react-router-dom'
import { StitchClientFactory } from 'mongodb-stitch'
const appId = 'ecommercechatbot-glwkl'
let stitchClientPromise = StitchClientFactory.create(appId)
stitchClientPromise.then(stitchClient => {
let db = stitchClient.service('mongodb', 'mongodb-
atlas').db('swagstore')
let props = { stitchClient, db }
ReactDOM.render(
<BrowserRouter>
<App {...props} />
</BrowserRouter>,
document.getElementById('root')
)
registerServiceWorker()
})

AUTHENTICATION
BROWSE
SELECT ITEM
IN
STOCK?
CART
CHECKOUT
UPDATE NOTIFY
Authentication
• Leverage integration with Google Sign-in
to authenticate the user and set the user’s
profile context.

AUTHENTICATION
BROWSE
SELECT ITEM
IN
STOCK?
CART
CHECKOUT
UPDATE NOTIFY
Browse
• Leverage SDK db.collection() to access
SWAG-STORE database and products
collection
import React, { Component } from 'react'
import { ProductDeck } from '../components/Products'
class ProductsPage extends Component {
constructor(props) {
super(props)
this.state = {
products: []
}
}
getProducts() {
this.props.db
.collection('products’)
.find({ category: this.props.match.params.category })
.execute()
.then(products => {
this.setState({ products })
})
.catch(err => {
console.log(err)
})
}
componentDidMount() {
this.getProducts()
}
componentDidUpdate(prevProps, prevState, snapshot) {
if (this.props.match.params.category !== prevProps.match.params.category) {

AUTHENTICATION
BROWSE
SELECT ITEM
IN
STOCK?
CART
CHECKOUT
UPDATE NOTIFY
Click on Product
• Reveals Product Page
getProduct() {
this.props.db
.collection('products’)
.findOne({ id: this.props.match.params.id })
.then(product => {
this.setState({
product,
notifying:this.props.notify
.includes(this.props.match.params.id)
})
this.props.handleBrowsedProduct(product)
})
.catch(err => {
console.log(err)
})
}

AUTHENTICATION
BROWSE
SELECT ITEM
IN
STOCK?
CART
CHECKOUT
UPDATE NOTIFY

UPDATE NOTIFY
IN
STOCK?
AUTHENTICATION
BROWSE
SELECT ITEM
CART
CHECKOUT

Event Subscriptions
in App Structure
• Event Subscriptions are a
top-level Stitch asset
• Can be constructed in the UI
or with the CLI

Change Streams Future Work
• Non-Atlas/Multiple Application per Application
• Events that span Multiple MongoDB Instances
• Timed Events
• Execute on a certain Interval
• Execute at a given time
• For now, can use Webhooks with CRON service

What do change events look like?
Event Subscription (Trigger) Event Source (Actual event document)

Example Trigger
Name of the Trigger
Type: DATABASE, AUTHENTICATION
Type of Operation to Trigger
Database
Collection
Match Criteria
Full Document or Partial
Function to be called on Trigger

Example Trigger Event Document
Type of Database Operation
Full Document impacted by operation
Database
Collection

MongoDB World 2018: Ch-Ch-Ch-Ch-Changes: Taking Your Stitch Application to the Next Level With Triggers

Recommended

More Related Content

What's hot (12)

Similar to MongoDB World 2018: Ch-Ch-Ch-Ch-Changes: Taking Your Stitch Application to the Next Level With Triggers (20)

More from MongoDB (20)

Recently uploaded (20)

MongoDB World 2018: Ch-Ch-Ch-Ch-Changes: Taking Your Stitch Application to the Next Level With Triggers