When breaking your monolith into components, services or even functions you must understand WHERE and HOW you break your existing code base and architecture into smaller units to allow it to SCALE, PERFORM and make it EASY enough to operate! In this session a Dynatrace Technical AWS Advocate, shows us how companies such as Barbri, Landbay and Citrix redefined their architecture on top of AWS; how Dynatrace was leveraged to re-platform and re-architecture; and how these lessons learned can benefit all of us to transform Fearless from Monolith to Serverless!

1. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
George Mao
Specialist Solutions Architect, Serverless, Amazon Web Services
198249
Fearless from Monolith to Serverless
with Dynatrace
Wayne Segar
Solutions Architect, Dynatrace

2. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
From Monolith to Serverless with Dynatrace
What is Serverless?
Evolution of Serverless Computing
From Monolith to Serverless with Dynatrace

3. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
What is
Serverless?
Build and run applications
without thinking about servers

4. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Challenges with Servers
Operations and management Scaling
Provisioning and utilization Availability and fault tolerance

5. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Physical Servers
Cloud
Serverless
Serverless Revolution
201620061996

6. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
AWS Lambda: Serverless Compute
• AWS Lambda lets you run code without managing servers.
• Don’t Pay for Idle time: You only pay for compute you consume
No Servers to Manage Continuous Scaling Subsecond Metering

7. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
How it Works
EVENT SOURCE SERVICES (ANYTHING)
Changes in data
state
Requests to
endpoints
Changes in
resource state
FUNCTION
Node.js
Python
Java
C#
Go

8. Splitting
Monoliths
Into
Components
Courtesy of

9. Into
Services
Courtesy of

10. Courtesy of

11. Amazon SNS
Amazon S3
Amazon API
Gateway
Amazon
SQS
Amazon
Kinesis
Amazon
DynamoDB
Business Logic
Glue between
the bricks
AWS Serverless
Leveraging standard service
and platform capabilities
Courtesy of

12. Amazon SNS
Amazon S3
Amazon API
Gateway
Amazon
SQS
Amazon
Kinesis
Amazon
DynamoDB
Microservices
to Functions
Courtesy of

13. Amazon API
Gateway
Amazon
SQS
Courtesy of
Microservices
for message
delivery

14. Amazon API
Gateway
Amazon
Kinesis
Amazon
DynamoDB
Courtesy of
Microservices
for data
persistence

15. Amazon API
Gateway
Amazon SNS
Amazon S3
Courtesy of
Microservices
for workflow
management

16. Amazon SNS
Amazon S3
Amazon API
Gateway
Amazon
SQS
Amazon
Kinesis
Amazon
DynamoDB
Microservices
On Lambda
Are Ephemeral
When the system is
idle, it shuts down and
costs nothing to run
Courtesy of

17. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Who is
• Dynatrace is at the forefront of AI-based monitoring platform
• 8 years a leader in Gartner’s MQ for Application Performance
Monitoring Suites,
5th consecutive year leader in terms of market share*
• Runs on the AWS Cloud, monitoring up to 100,000+ hosts
• 72 of the Fortune 100, trust Dynatrace incl. Verizon, Citrix
• AWS Competency Partner and available on AWS Marketplace
• Dynatrace digital assistant davis using Amazon Technology
*Gartner: Market Share Analysis: Performance
Analysis Software, Worldwide, 2016

18. 26
500
Feature Releases / Year
Deployments / Day
31000 60h
Unit & Int Tests / hour UI Tests per Build
More Quality
~120 340
Code commits / day Stories per sprint
More Agile
93%
Production bugs found by Dev
More Stability 450 99.998%
Global EC2 Instances Global Availability
Dynatrace Journey! It shaped our product!

19. Questions / Challenges on that “Migration” Journey
• Where to start?
• What to break out?
• Dependencies?
• Works as expected?
• Users happy?
• Does it scale?
• Does it perform?
• Works as expected?
• Users happy?
• How to optimize?
• How to automate?
?
? ?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
? ?
?
?
?
? ? ?
?
?
?
?
From
On-Premise Monolith
Via
Containerized Services
To
Functions & Cloud Services

20. FullStack-PoweredbyDynatraceOneAgent

21. Where to start?
What to break out?
Dependencies?
Works as expected?
Does it scale?
Does it perform?
Works as expected?
How to scale? How to optimize?
How to automate?
Dynatrace Smartscape
We tell you WHAT to migrate and
WHERE to break the monolith!
Dynatrace AI
Automatic End User, Service and
Infrastructure Root Cause Analysis
Dynatrace Automation API
Automate Rollout in your
Containers & Functions. Automate
Shift-Left and Self-Healing
Dynatrace enables Break, Shift, Re-Platform

22. Dynatrace Helps to Decide What to Migrate/Shift
Dynatrace Smartscape
• Auto-dependency
mapping of internal &
external services
• Auto-load detection
and baselining
• Auto-resource
consumption detection
Automate T-Shirt Size Categorization for your Migration Plan

23. Tightly coupled! Shall we really distribute/extract?
Dynatrace “virtually” breaks your monolith!
Dynatrace helps to decide WHERE to break the
monolith

24. Getting to know the Monolith Architecture
?
?
?
?
?
?
?
?
?
?
?
Service
Endpoints
Depending
Services
Service Flow

25. Step #1 Dependency Information

26. Step #2 Detecting Service Endpoints
Endpoints CPULoad

27. Step #3 Understanding Service Flow per Endpoint
#1 Filter on specific endpoint
#2 See Execution Flow to Depending Services
#3 Dependencies
& Call Patterns
#4 Call Pattern Details

28. Step #3 Understanding Service Flow per Endpoint
#1 Filter on specific endpoint
#2 See Execution Flow to Depending Services
#3 Dependencies
& Call Patterns
#4 Call Pattern Details

29. What We Learned So Far!
/review
/finish
/config
/payment
Backend
Service X
DB A
Q 1
Endpoints +
Throughput
Dependencies &
Call Patterns
Flow & CPU
Consumption

30. Next: Where to Break the Monolith?
/review
/finish
/config
/payment
Backend
Service X
DB A
Q 1
?
?
?
?
?
?
?
?
?
?
Entry Points? Entry Points?

31. Step 4: Finding Entry Points with CPU Sampling
#1 Filter on specific endpoint
#2 Search for specific code
#3 Found Entry Point to Package

32. Step #5 Define Custom Service Entry Points
#1 Service Name
#2 Method

33. We “virtually” broke the monolith!
/review
/finish
/config
/payment
Backend
Service X
DB A
Q 1
User
Catalog
Check
Avail.
Credit-
Card
Billing
Order
Journey
Booking
Configuration
Payment

34. Step #6 Learn from Dynatrace Data
Tightly coupled! Shall we really distribute/extract?

35. Dynatrace validates & optimizes scalability
Low Load Peak LoadRegular Load
How does each service scale? Does it correctly fail over??

36. Dynatrace Baselining Validates Before vs After!
How is Performance & Resource Consumption per Service Endpoint?
Before After

37. Dynatrace APIs: How to DevOps? Shift-Left? Self-Healing?
pushDynatraceDeploymentEvent
Pushes Deployment Info to Dynatrace Entities
validateBuildDynatraceWorker
Compares Builds and Approves/Rejects Pipeline
pushDynatraceDeploymentEvent
Pushes Deployment Info to Dynatrace Entities
validateBuildDynatraceWorker
Validates Production and Approves/Rejects Pipeline
handleDynatraceProblemNotification
Executes Auto-Remediating Actions, e.g: Rollback
Build 6
Build 7
Production
Production
Auto-Approve!
Auto-Reject!
Auto-Approve!
Auto-Reject!
https://github.com/Dynatrace/AWSDevOpsTutorial

38. Planned Architectural Service Flow
Actual Architectural Service Flow
Identify / Optimize Architectural Patterns
Recursive Calls, N+1 Call Pattern, Chatty Interfaces, No Caching Layer …
Dynatrace Service Flow Validates your Architecture!

39. WHAT RUNS WHERE?
Monitoring Data and Tags for each EC2 instance through CloudWatch
Dependencies to Cloud Services
Who is calling RDS? DynamoDB? Lambdas?
Dynatrace Cloud Monitoring Validates Cloud Services

40. Dynatrace Cloud Monitoring Validates Cloud Resource Usage
Automated Cloud Resource Usage and Error Detection

41. Dynatrace Container Monitoring
Automatic Injection & Meta Data Extraction!
Automatic Service Monitoring within the Containers

42. Method Level Visibility: No Code Change Required!
Dynatrace End-to-End through Lambda

43. From Monolith via Microservices to Serverless – and into the Mainframe
Dynatrace End-to-End through your Stack
Monolith
Serverless
Micro-
services
Legacy /
3rd-Party

44. Dynatrace RUM Validates User Behavior Before vs After!
How are Conversions, Bounce Rates, Interaction Behavior, JavaScript Errors, Crashes?

45. Dynatrace AI: When something fails, find out: Where, Why and How Faster
If you break things!
ChatOps with the AI

46. • Signup through AWS Marketplace
• Run as SaaS or On-Premises
• Visit us at our booth
• Follow us @dynatrace
Try

47. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Fearless from Monolith to Serverless with Dynatrace
Thank You
Please complete the session
survey in the summit mobile app.