GARY ARORA
Cloud Solutions Architect
Deloitte Consulting LLP
What a Mesh!
Navigating a Mesh of Microservices in the
new Cloud-Native World with Istio
CLOUD EXPO 2018
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 2
Topics
Cloud Native
Containers
Microservices
Service Mesh
Istio
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 3
A Brief History of Computing
Virtualization
• You likely need to buy new hardware every time you
need a new server
• Often longer downtimes due to outages. Slow DR
• Unit of measure: physical servers
Physical Servers are great but…
• Run multiple virtual machines on one physical server
• Cost savings through reduced footprint, faster server
provisioning, and improved disaster recovery (DR)
• Unit of measure: virtual machines
…consider Virtualization
Hardware
CPU Memory Disk Network
Operating System
Application
Hardware
CPU Memory Disk Network
OS 1
App 1 App 2 App 3
OS 2 OS 3
Hypervisor
Operating System
Virtualization
2001
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 5
Public Cloud
• Limited by finite resources for servers, storage,
network IOPS
• Limited by budget and talent pool of in-house IT Ops
for maintenance & high-availability
On Premise Data Centers are great but…
• Potentially limitless capacity for servers, storage,
network IOPS
• Cloud providers performs all maintenance and
guarantees availability via various SLAs
…consider moving to Cloud
Virtualization
2001
Public Cloud
2006
Data
Operating System
Virtualization
Storage
Networking
Hardware
Applications
Data
Operating System
Virtualization
Storage
Networking
Hardware
Applications
Customer’sResponsibility
Customer’sResponsibility
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 6
Simply being in the cloud is
no longer enough to remain
competitive!
92% of enterprises are already
using the public cloud in 2018
Source: State of Cloud Survey, Rightscale
Cloud Native
An approach to maximizing the capabilities
of the cloud by rethinking technology
choices, architecture, and operations
CI/CD
DevOpsContainers
Microservices
Faster Time To
Market
Support rapid
Innovations
Increased
Resiliency
Agility & Scalability Increased Security Lower Costs
Benefits
Components
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 8
Containers
• VMs can take up a lot of system resources with full
virtual OS, RAM, and CPU cycles
• Startup time in minutes
• Limits the portability of applications
Virtual Machines are great but…
• Reduced IT management resources
• Startup time in milliseconds
• A portable, consistent operating environment for
development, testing, and deployment
…consider Containerization
Virtualization
2001
Public Cloud
2006
Containers
2013
Hardware
CPU Memory Disk Network
Guest
OS
App 1 App 2 App 3
Guest
OS
Guest
OS
Hypervisor
Host Operating System
Bins/Libs Bins/Libs Bins/Libs
Hardware
CPU Memory Disk Network
App 1 App 2 App 3
Docker Engine
Host Operating System
Bins/Libs Bins/Libs Bins/Libs
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 9
Container Management
• Containers cannot inherently communicate with
each other
• Containers have to be managed & deployed
appropriately
• Native auto scaling is not possible
• Distributed traffic is still challenging
Containers are great but…
• Automate packaging
• Service Discovery & Load
Balancing
• Storage Orchestration
• Self-Healing
…consider Container Management
• Batch Execution
• Secret & Configuration
Mgmt.
• Horizontal Scaling
• Automatic Rollbacks &
Rollouts
Virtualization
2001
Public Cloud
2006
Containers
2013
Container Mgmt.
2015
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 10
Microservices
2011-14
Microservices
• Scaling capabilities independently is challenging
• Changing one thing requires deploying everything
• Require extensive manual testing
Monoliths are great but…
• Independently develop and deploy services
• Organized around business capabilities
• Effective fault isolation
• Scalability and reusability
• Polygot
…consider Microservices
Database
Payment
Checkout
Shopping Cart
3rd Party Integrations
Recommendation
Product Catalog
Frontend
Shipping
Email Notifications
Users
Business Logic
Payment
Checkout Shopping Cart
3rd Party Integrations
RecommendationsProduct Catalog
Frontend
Shipping
Email Notifications
Users
Virtualization
2001
Public Cloud
2006
Containers
2013
Container Mgmt.
2015
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 11
It becomes an orders of magnitude larger problem to network and debug a mesh of microservices
When Microservices Grow…
Payment
Checkout
Shopping Cart
Recommendations
Product Catalog
Frontend
Shipping
Email Notifications
Reviews User Profile
Real time Pricing
3rd Party Price Match
Order Auto-
Replenishment
Shopping History
Refunds
Customer
Personalization
Customer Service
In-store inventory
Users
Promotions Marketing Emails
Payment
Shopping Cart
Recommendations
Shipping
Email Notifications
User Profile
Real time Pricing
3rd Party Price Match
Order Auto-
Replenishment
Shopping History
Refunds
Customer
Personalization
Customer Service
Marketing Campaigns
Premium User Services
The Eight Fallacies of
Distributed Computing
1. The network is reliable
2. Latency is zero
3. Bandwidth is infinite
4. The network is secure
5. Topology doesn't change
6. There is 1 administrator
7. Transport cost is zero
8. Network is homogeneous
Source: Peter Deutsch, 1994, Sun Microsystems
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 12
Global Microservices Trend
91% are using or have plans to use
microservices
92% expect to grow their use of
microservices in the coming year
86% expect microservices to be
the default within five years
Microservices have become mainstream… …but adoption still has many challenges
99% report challenges with using
microservices
73% find troubleshooting is harder in a
microservices environment
98% of those that face issues identifying
the root cause of issues in microservices
environments report it has a direct
business impact
91% 99%
92%
86%
73%
98%
Source: Online survey by Dimensional Research & LightStep | April 2018
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 13
Service Mesh
They still require a lot of management to:
• Debug network & infrastructure issues
• Create dependency graphs & latency
• Authenticate, rate limit, access control
Microservices are great but… …consider service mesh
• Tracing
• Monitoring
• Logging
• Authentication
• Pipeline
• Resilience
• Routing
• Discovery
Microservices
2011-14
Virtualization
2001
Public Cloud
2006
Containers
2013
Container Mgmt.
2015
Service Mesh.
2018
Istio
An open services platform to manage service interactions across containers and VM-based workloads
“Kubernetes changed how we deploy applications.
Istio is going to change how we connect, manage, and secure them”
~Kelsey Hightower, Google Developer Advocate
Connect
Intelligently control the flow of
traffic and API calls between
services
Secure
Manages authentication,
authorization, and encryption
of communications
Control
Apply policies and ensure that
they are enforced, and that
resources are aptly distributed
Observe
Rich automated tracing,
monitoring, and logging for all
services
Source: Istio documentation
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 15
Istio Architecture
1. Envoy is a sidecar proxy
mediates all traffic
2. Mixer enforces access control
and collects telemetry data
3. Pilot provides service discovery,
& traffic management via rules
4. Citadel provides service-to-
service and user authentication
Source: Istio documentation
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 16
Istio Key Capabilities: Traffic Management
• Request routing
• Discovery and load balancing
• Handling failures
• Rate limiting, circuit breakers, A/B
testing
• Fault injection
• Rule configuration
Source: Istio documentation
Traffic splitting decouples from
infrastructure scaling
Content-based traffic steering
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 17
Traffic Management: Canary Deployments
• Once regularly used in coal
mining as an early detectors of
toxic gases
• Incremental rollouts to a subset
of users
• Can detect potential bugs and
disruption without affecting
every other system running.
With IstioWithout Istio
http://reviews.example.com
Prod 75% Canary 25%
Prod 90% Canary 10%
Prod 75%
Prod 90% Canary 10%
Canary 25%
http://reviews.example.com
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 18
Apply
Rules
25%
75%
reviews.example.com
http://reviews.example.com
Source: Istio documentation
Traffic & Routing
rules
Common scenarios where this is used include A/B testing or canary rollouts.
Traffic Management Example
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 19
Istio Key Capabilities: Security
• Key and certificate management
• Perimeter proxies
• Authentication
• Transport authentication
• Origin authentication
• Mutual TLS authentication
• Authorization
• Role-based Access Control (RBAC)
• Namespace-level
• Service-level
• Method-level access control
Source: Istio documentation
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 20
Istio Key Capabilities: Telemetry & Distributed Tracing
• Telemetry is automatically
injected in any service pod
providing Prometheus-style
network and L7 protocol metrics
• Istio dynamically traces the flow
and chained connections of the
microservices mesh.
Source: Istio documentation
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 21
Summary
Higher Abstraction. Increased focus on functionality
• Servers -> Virtual Machines -> Containers -> Serverless
• Monolithics -> Microservices
• Proprietary -> Open Source
• Single Vendor -> Cross-vendor
Copyright © 2018 Deloitte Development LLC. All rights reserved. | 22
By 2020, 75% of application
purchases supporting digital
business will be “build” not
“buy”
Source: Gartner Forecast Analysis 2015
About Deloitte
Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee (“DTTL”), its network of member firms, and
their related entities. DTTL and each of its member firms are legally separate and independent entities. DTTL (also referred to as “Deloitte Global”) does not
provide services to clients. In the United States, Deloitte refers to one or more of the US member firms of DTTL, their related entities that operate using the
“Deloitte” name in the United States and their respective affiliates. Certain services may not be available to attest clients under the rules and regulations of
public accounting. Please see www.deloitte.com/about to learn more about our global network of member firms.
This publication contains general information only and Deloitte is not, by means of this publication, rendering accounting, business, financial, investment, legal,
tax, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for
any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a
qualified professional advisor. Deloitte shall not be responsible for any loss sustained by any person who relies on this publication.
Copyright © 2018 Deloitte Development LLC. All rights reserved.
@AroraGary

More Related Content

PDF
Leapfrog into Serverless - a Deloitte-Amtrak Case Study | Serverless Confere...
PPTX
Serverless Architecture at iRobot
PDF
Building a Real-Time Forecasting Engine with Scala and Akka
PDF
Project Sherpa: How RightScale Went All in on Docker
PDF
Going Reactive in the Land of No
PDF
RedisConf18 - Common Redis Use Cases for Cloud Native Apps and Microservices
PDF
How to build streaming data pipelines with Akka Streams, Flink, and Spark usi...
PDF
Extracting Value from IOT using Azure Cosmos DB, Azure Synapse Analytics and ...
Leapfrog into Serverless - a Deloitte-Amtrak Case Study | Serverless Confere...
Serverless Architecture at iRobot
Building a Real-Time Forecasting Engine with Scala and Akka
Project Sherpa: How RightScale Went All in on Docker
Going Reactive in the Land of No
RedisConf18 - Common Redis Use Cases for Cloud Native Apps and Microservices
How to build streaming data pipelines with Akka Streams, Flink, and Spark usi...
Extracting Value from IOT using Azure Cosmos DB, Azure Synapse Analytics and ...

What's hot (20)

PDF
Qlik and Confluent Success Stories with Kafka - How Generali and Skechers Kee...
PDF
Exposing and Controlling Kafka Event Streaming with Kong Konnect Enterprise |...
PPTX
Transforming the Monolith at 20M tph
PDF
Journey to the Modern App with Containers, Microservices and Big Data
PDF
Cisco's MultiCloud Strategy
PDF
MongoDB-as-a-Service on Pivotal Cloud Foundry
PPTX
Cloudsolutionday 2016: Docker & FAAS at getvero.com
PPTX
Meetup #3: Migrating an Oracle Application from on-premise to AWS
PPTX
Cloud Native Application Development-build fast, low TCO, scalable & agile so...
PPTX
Cloud Computing Design Considerations
PDF
The Future of Services: Building Asynchronous, Resilient and Elastic Systems
PDF
Matt Chung (Independent) - Serverless application with AWS Lambda
PPTX
The Problem is Data: Gwen Shapira, Confluent, Serverless NYC 2018
PPTX
Google Cloud and Data Pipeline Patterns
PPTX
Microservice.net by sergey seletsky
PDF
Kafka Summit SF 2017 - Worldwide Scalable and Resilient Messaging Services wi...
PPTX
Building the Serverless Container Experience: Kevin McGrath, Spotinst, Server...
PDF
Building Scalable Real-Time Data Pipelines with the Couchbase Kafka Connector...
PPTX
RedisConf17 - Smartwaiver - Using Redis for Kiosk Registration Command and Co...
PDF
Death of the dumb pipes: Using Apache Kafka® for Integration projects
Qlik and Confluent Success Stories with Kafka - How Generali and Skechers Kee...
Exposing and Controlling Kafka Event Streaming with Kong Konnect Enterprise |...
Transforming the Monolith at 20M tph
Journey to the Modern App with Containers, Microservices and Big Data
Cisco's MultiCloud Strategy
MongoDB-as-a-Service on Pivotal Cloud Foundry
Cloudsolutionday 2016: Docker & FAAS at getvero.com
Meetup #3: Migrating an Oracle Application from on-premise to AWS
Cloud Native Application Development-build fast, low TCO, scalable & agile so...
Cloud Computing Design Considerations
The Future of Services: Building Asynchronous, Resilient and Elastic Systems
Matt Chung (Independent) - Serverless application with AWS Lambda
The Problem is Data: Gwen Shapira, Confluent, Serverless NYC 2018
Google Cloud and Data Pipeline Patterns
Microservice.net by sergey seletsky
Kafka Summit SF 2017 - Worldwide Scalable and Resilient Messaging Services wi...
Building the Serverless Container Experience: Kevin McGrath, Spotinst, Server...
Building Scalable Real-Time Data Pipelines with the Couchbase Kafka Connector...
RedisConf17 - Smartwaiver - Using Redis for Kiosk Registration Command and Co...
Death of the dumb pipes: Using Apache Kafka® for Integration projects
Ad

Similar to Navigating a Mesh of Microservices in the new Cloud-Native World with Istio (20)

PPTX
Virtualization and cloud computing
PPTX
The Journey of IT – Mainframe to Serverless
PPTX
Istio Mesh – Managing Container Deployments at Scale
PPTX
Manging Container Deployments at Scale
ODP
Implementing Private Clouds
PDF
Istio as an enabler for migrating to microservices (edition 2022)
PDF
Introduction to Istio Service Mesh
PPT
lecture5 Cloud Computing with Utility .ppt
PPTX
Making sense of microservices, service mesh, and serverless
PPTX
Cloud Computing
PPTX
Planning a Successful Cloud - Design from Workload to Infrastructure
PPTX
Cloud.pptx
PPTX
Speaker Presention by Irena Bojanova of the University of Maryland University...
PPTX
Microservices at Scale with Istio
PPTX
Api service mesh and microservice tooling
PPTX
CompTIA Cloud Plus Certification Bootcamp June 2017
PDF
Make Java Microservices Resilient with Istio - Mangesh - IBM - CC18
PDF
Istio presentation jhug
PPTX
vm provisioning
PPTX
2 vm provisioning
Virtualization and cloud computing
The Journey of IT – Mainframe to Serverless
Istio Mesh – Managing Container Deployments at Scale
Manging Container Deployments at Scale
Implementing Private Clouds
Istio as an enabler for migrating to microservices (edition 2022)
Introduction to Istio Service Mesh
lecture5 Cloud Computing with Utility .ppt
Making sense of microservices, service mesh, and serverless
Cloud Computing
Planning a Successful Cloud - Design from Workload to Infrastructure
Cloud.pptx
Speaker Presention by Irena Bojanova of the University of Maryland University...
Microservices at Scale with Istio
Api service mesh and microservice tooling
CompTIA Cloud Plus Certification Bootcamp June 2017
Make Java Microservices Resilient with Istio - Mangesh - IBM - CC18
Istio presentation jhug
vm provisioning
2 vm provisioning
Ad

Recently uploaded (20)

PDF
Rapid Prototyping: A lecture on prototyping techniques for interface design
PDF
Transform-Quality-Engineering-with-AI-A-60-Day-Blueprint-for-Digital-Success.pdf
PPTX
Internet of Everything -Basic concepts details
PPTX
Module 1 Introduction to Web Programming .pptx
PDF
A hybrid framework for wild animal classification using fine-tuned DenseNet12...
PDF
SaaS reusability assessment using machine learning techniques
PDF
LMS bot: enhanced learning management systems for improved student learning e...
PDF
Connector Corner: Transform Unstructured Documents with Agentic Automation
PDF
Altius execution marketplace concept.pdf
PDF
NewMind AI Weekly Chronicles – August ’25 Week IV
PDF
Electrocardiogram sequences data analytics and classification using unsupervi...
PDF
The AI Revolution in Customer Service - 2025
PDF
5-Ways-AI-is-Revolutionizing-Telecom-Quality-Engineering.pdf
PDF
Aug23rd - Mulesoft Community Workshop - Hyd, India.pdf
PDF
CXOs-Are-you-still-doing-manual-DevOps-in-the-age-of-AI.pdf
PDF
Transform-Your-Streaming-Platform-with-AI-Driven-Quality-Engineering.pdf
PPTX
Build automations faster and more reliably with UiPath ScreenPlay
PDF
Early detection and classification of bone marrow changes in lumbar vertebrae...
PDF
Advancing precision in air quality forecasting through machine learning integ...
PDF
A symptom-driven medical diagnosis support model based on machine learning te...
Rapid Prototyping: A lecture on prototyping techniques for interface design
Transform-Quality-Engineering-with-AI-A-60-Day-Blueprint-for-Digital-Success.pdf
Internet of Everything -Basic concepts details
Module 1 Introduction to Web Programming .pptx
A hybrid framework for wild animal classification using fine-tuned DenseNet12...
SaaS reusability assessment using machine learning techniques
LMS bot: enhanced learning management systems for improved student learning e...
Connector Corner: Transform Unstructured Documents with Agentic Automation
Altius execution marketplace concept.pdf
NewMind AI Weekly Chronicles – August ’25 Week IV
Electrocardiogram sequences data analytics and classification using unsupervi...
The AI Revolution in Customer Service - 2025
5-Ways-AI-is-Revolutionizing-Telecom-Quality-Engineering.pdf
Aug23rd - Mulesoft Community Workshop - Hyd, India.pdf
CXOs-Are-you-still-doing-manual-DevOps-in-the-age-of-AI.pdf
Transform-Your-Streaming-Platform-with-AI-Driven-Quality-Engineering.pdf
Build automations faster and more reliably with UiPath ScreenPlay
Early detection and classification of bone marrow changes in lumbar vertebrae...
Advancing precision in air quality forecasting through machine learning integ...
A symptom-driven medical diagnosis support model based on machine learning te...

Navigating a Mesh of Microservices in the new Cloud-Native World with Istio

  • 1. GARY ARORA Cloud Solutions Architect Deloitte Consulting LLP What a Mesh! Navigating a Mesh of Microservices in the new Cloud-Native World with Istio CLOUD EXPO 2018
  • 2. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 2 Topics Cloud Native Containers Microservices Service Mesh Istio
  • 3. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 3 A Brief History of Computing
  • 4. Virtualization • You likely need to buy new hardware every time you need a new server • Often longer downtimes due to outages. Slow DR • Unit of measure: physical servers Physical Servers are great but… • Run multiple virtual machines on one physical server • Cost savings through reduced footprint, faster server provisioning, and improved disaster recovery (DR) • Unit of measure: virtual machines …consider Virtualization Hardware CPU Memory Disk Network Operating System Application Hardware CPU Memory Disk Network OS 1 App 1 App 2 App 3 OS 2 OS 3 Hypervisor Operating System Virtualization 2001
  • 5. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 5 Public Cloud • Limited by finite resources for servers, storage, network IOPS • Limited by budget and talent pool of in-house IT Ops for maintenance & high-availability On Premise Data Centers are great but… • Potentially limitless capacity for servers, storage, network IOPS • Cloud providers performs all maintenance and guarantees availability via various SLAs …consider moving to Cloud Virtualization 2001 Public Cloud 2006 Data Operating System Virtualization Storage Networking Hardware Applications Data Operating System Virtualization Storage Networking Hardware Applications Customer’sResponsibility Customer’sResponsibility
  • 6. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 6 Simply being in the cloud is no longer enough to remain competitive! 92% of enterprises are already using the public cloud in 2018 Source: State of Cloud Survey, Rightscale
  • 7. Cloud Native An approach to maximizing the capabilities of the cloud by rethinking technology choices, architecture, and operations CI/CD DevOpsContainers Microservices Faster Time To Market Support rapid Innovations Increased Resiliency Agility & Scalability Increased Security Lower Costs Benefits Components
  • 8. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 8 Containers • VMs can take up a lot of system resources with full virtual OS, RAM, and CPU cycles • Startup time in minutes • Limits the portability of applications Virtual Machines are great but… • Reduced IT management resources • Startup time in milliseconds • A portable, consistent operating environment for development, testing, and deployment …consider Containerization Virtualization 2001 Public Cloud 2006 Containers 2013 Hardware CPU Memory Disk Network Guest OS App 1 App 2 App 3 Guest OS Guest OS Hypervisor Host Operating System Bins/Libs Bins/Libs Bins/Libs Hardware CPU Memory Disk Network App 1 App 2 App 3 Docker Engine Host Operating System Bins/Libs Bins/Libs Bins/Libs
  • 9. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 9 Container Management • Containers cannot inherently communicate with each other • Containers have to be managed & deployed appropriately • Native auto scaling is not possible • Distributed traffic is still challenging Containers are great but… • Automate packaging • Service Discovery & Load Balancing • Storage Orchestration • Self-Healing …consider Container Management • Batch Execution • Secret & Configuration Mgmt. • Horizontal Scaling • Automatic Rollbacks & Rollouts Virtualization 2001 Public Cloud 2006 Containers 2013 Container Mgmt. 2015
  • 10. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 10 Microservices 2011-14 Microservices • Scaling capabilities independently is challenging • Changing one thing requires deploying everything • Require extensive manual testing Monoliths are great but… • Independently develop and deploy services • Organized around business capabilities • Effective fault isolation • Scalability and reusability • Polygot …consider Microservices Database Payment Checkout Shopping Cart 3rd Party Integrations Recommendation Product Catalog Frontend Shipping Email Notifications Users Business Logic Payment Checkout Shopping Cart 3rd Party Integrations RecommendationsProduct Catalog Frontend Shipping Email Notifications Users Virtualization 2001 Public Cloud 2006 Containers 2013 Container Mgmt. 2015
  • 11. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 11 It becomes an orders of magnitude larger problem to network and debug a mesh of microservices When Microservices Grow… Payment Checkout Shopping Cart Recommendations Product Catalog Frontend Shipping Email Notifications Reviews User Profile Real time Pricing 3rd Party Price Match Order Auto- Replenishment Shopping History Refunds Customer Personalization Customer Service In-store inventory Users Promotions Marketing Emails Payment Shopping Cart Recommendations Shipping Email Notifications User Profile Real time Pricing 3rd Party Price Match Order Auto- Replenishment Shopping History Refunds Customer Personalization Customer Service Marketing Campaigns Premium User Services The Eight Fallacies of Distributed Computing 1. The network is reliable 2. Latency is zero 3. Bandwidth is infinite 4. The network is secure 5. Topology doesn't change 6. There is 1 administrator 7. Transport cost is zero 8. Network is homogeneous Source: Peter Deutsch, 1994, Sun Microsystems
  • 12. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 12 Global Microservices Trend 91% are using or have plans to use microservices 92% expect to grow their use of microservices in the coming year 86% expect microservices to be the default within five years Microservices have become mainstream… …but adoption still has many challenges 99% report challenges with using microservices 73% find troubleshooting is harder in a microservices environment 98% of those that face issues identifying the root cause of issues in microservices environments report it has a direct business impact 91% 99% 92% 86% 73% 98% Source: Online survey by Dimensional Research & LightStep | April 2018
  • 13. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 13 Service Mesh They still require a lot of management to: • Debug network & infrastructure issues • Create dependency graphs & latency • Authenticate, rate limit, access control Microservices are great but… …consider service mesh • Tracing • Monitoring • Logging • Authentication • Pipeline • Resilience • Routing • Discovery Microservices 2011-14 Virtualization 2001 Public Cloud 2006 Containers 2013 Container Mgmt. 2015 Service Mesh. 2018
  • 14. Istio An open services platform to manage service interactions across containers and VM-based workloads “Kubernetes changed how we deploy applications. Istio is going to change how we connect, manage, and secure them” ~Kelsey Hightower, Google Developer Advocate Connect Intelligently control the flow of traffic and API calls between services Secure Manages authentication, authorization, and encryption of communications Control Apply policies and ensure that they are enforced, and that resources are aptly distributed Observe Rich automated tracing, monitoring, and logging for all services Source: Istio documentation
  • 15. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 15 Istio Architecture 1. Envoy is a sidecar proxy mediates all traffic 2. Mixer enforces access control and collects telemetry data 3. Pilot provides service discovery, & traffic management via rules 4. Citadel provides service-to- service and user authentication Source: Istio documentation
  • 16. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 16 Istio Key Capabilities: Traffic Management • Request routing • Discovery and load balancing • Handling failures • Rate limiting, circuit breakers, A/B testing • Fault injection • Rule configuration Source: Istio documentation Traffic splitting decouples from infrastructure scaling Content-based traffic steering
  • 17. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 17 Traffic Management: Canary Deployments • Once regularly used in coal mining as an early detectors of toxic gases • Incremental rollouts to a subset of users • Can detect potential bugs and disruption without affecting every other system running. With IstioWithout Istio http://reviews.example.com Prod 75% Canary 25% Prod 90% Canary 10% Prod 75% Prod 90% Canary 10% Canary 25% http://reviews.example.com
  • 18. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 18 Apply Rules 25% 75% reviews.example.com http://reviews.example.com Source: Istio documentation Traffic & Routing rules Common scenarios where this is used include A/B testing or canary rollouts. Traffic Management Example
  • 19. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 19 Istio Key Capabilities: Security • Key and certificate management • Perimeter proxies • Authentication • Transport authentication • Origin authentication • Mutual TLS authentication • Authorization • Role-based Access Control (RBAC) • Namespace-level • Service-level • Method-level access control Source: Istio documentation
  • 20. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 20 Istio Key Capabilities: Telemetry & Distributed Tracing • Telemetry is automatically injected in any service pod providing Prometheus-style network and L7 protocol metrics • Istio dynamically traces the flow and chained connections of the microservices mesh. Source: Istio documentation
  • 21. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 21 Summary Higher Abstraction. Increased focus on functionality • Servers -> Virtual Machines -> Containers -> Serverless • Monolithics -> Microservices • Proprietary -> Open Source • Single Vendor -> Cross-vendor
  • 22. Copyright © 2018 Deloitte Development LLC. All rights reserved. | 22 By 2020, 75% of application purchases supporting digital business will be “build” not “buy” Source: Gartner Forecast Analysis 2015
  • 23. About Deloitte Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee (“DTTL”), its network of member firms, and their related entities. DTTL and each of its member firms are legally separate and independent entities. DTTL (also referred to as “Deloitte Global”) does not provide services to clients. In the United States, Deloitte refers to one or more of the US member firms of DTTL, their related entities that operate using the “Deloitte” name in the United States and their respective affiliates. Certain services may not be available to attest clients under the rules and regulations of public accounting. Please see www.deloitte.com/about to learn more about our global network of member firms. This publication contains general information only and Deloitte is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. Deloitte shall not be responsible for any loss sustained by any person who relies on this publication. Copyright © 2018 Deloitte Development LLC. All rights reserved. @AroraGary