NoOps in a Serverless World
0 likes344 views
The document discusses serverless computing and NoOps. It begins with an introduction to serverless computing principles, services, and benefits. It then provides examples of serverless architectures for event-driven applications and data processing. The document emphasizes that serverless enables organizations to focus on business logic rather than infrastructure management. It argues that NoOps is the next evolution of DevOps, where infrastructure tasks are fully automated. Finally, it outlines strategies for achieving NoOps in a serverless world, including through observability, automation, security automation, and local testing.
NoOps in a Serverless World
- 1. GARY ARORA Deloitte Consulting LLP NoOps in a Serverless WorldState Farm IT Symposium 2019 “Serverless”
- 2. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 2 Let’s check the room… 2. What % of enterprises have adopted the public cloud in 2019? A. 40 – 50 % B. 50 – 60 % C. 70 – 80 % D. 90 % and higher 3. What is the #1 cloud priority for enterprises in 2019? A. Optimize existing use of cloud (cost savings) B. Migrate to cloud C. Expand use of containers D. Implement a Cloud First strategy 1. Which is currently the fastest growing cloud service? A. Serverless B. Stream processing C. Machine Learning D. Container-as-a-service 4. What is the most used PaaS offering by enterprises in 2019? A. Database as a Service (Relational) B. Push notifications C. Queueing/Stream processing D. Batch processing 5. What is the number 1 reason for adopting cloud in 2019? A. Catalyst for IT innovation B. Improve speed to market C. Elasticity & scalability D. Lower costs
- 3. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 3 Let’s check the room… 2. What % of enterprises have adopted the public cloud in 2019? A. 40 – 50 % B. 50 – 60 % C. 70 – 80 % D. 90 % and higher 3. What is the #1 cloud priority for enterprises in 2019? A. Optimize existing use of cloud (cost savings) B. Migrate to cloud C. Expand use of containers D. Implement a Cloud First strategy 1. Which is currently the fastest growing cloud service? A. Serverless B. Stream processing C. Machine Learning D. Container-as-a-service 4. What is the most used PaaS offering by enterprises in 2019? A. Database as a Service (Relational) B. Push notifications C. Queueing/Stream processing D. Batch processing 5. What is the number 1 reason for adopting cloud in 2019? A. Catalyst for IT innovation B. Improve speed to market C. Elasticity & scalability D. Lower costs Source: RightScale 2019 State of the Cloud Report
- 4. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 4 Serverless 101 What is Serverless? Don’t provision or manage infrastructure Auto scale based on demand Highly available, fault tolerant, and secure Pay only for what you use and when you use it Serverless Principles Traditional IT Application Data Runtime OS Virtualization Servers Network IaaS Application Data Runtime OS Virtualization Servers Network PaaS Application Data Runtime OS Virtualization Servers Network FaaS Application Data Runtime OS Virtualization Servers Network Cloud Provider Manages You Manage
- 5. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 5 Serverless has enabled a Computing Paradigm Shift Focus on business logic Stackabstraction Serverless Unit of scale: Functions • Deploy in milliseconds • Live for seconds Containerization Unit of scale: Application • Deploy in seconds • Live for minutes/hours Virtual Machines Unit of scale: Machine • Deploy in minutes • Live for week Physical Machines Unit of scale: Physical servers • Deploy in months • Live for years Server based Cloud Native Serverless is a way to focus on business value
- 6. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 6 Compute AWS Lambda AWS Fargate Cloud Functions App Engine Functions App Service Integration Amazon API Gateway Amazon SQS Cloud Pub/Sub Cloud Tasks Logic Apps API Management Data Stores Amazon S3 Amazon DynamoDB Cloud Firestore Cloud Storage Cosmos DB Blob storage Analytics Amazon Kinesis Amazon Athena BigQuery Data Studio Stream Analytics DevOps AWS CodeBuild Amazon CloudWatch CloudBuild Stackdriver Azure DevOps Serverless is a lot more than just Function-as-a-Service (FaaS) Serverless is a design pattern that spans the entire stack. Organizations can kickoff their Serverless journey from anywhere within the software development lifecycle
- 7. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 7 Why Serverless: Key Drivers for Serverless Adoption The market forces have shaped consumer expectations to demand constant innovation. Winners of tomorrow will be those who have unlocked the benefits of cloud – enabling disruptive innovation and speed to market. 35% 38% 50% 51% 54% 56% 63% 64% Lower cost Improve security and resilience Adopt on-demand self-service Access to new technology Rapid elasticity / scalability Improve performance Improve speed / time to… Cloud as a catalyst for IT… Why Are Organizations Adopting Cloud? Source: Deloitte 2018 Global Outsourcing Survey. Deloitte. June 2018. Faster deployment cycles Faster reaction time to market No More Servers (physical or virtual) to maintain = Focus on Business Value Infinite scalability and high availability Event-driven and Microservice design Low Startup Cost No idle time costs Less Code to write & debug Initiates journey to NoOps CaseforServerless
- 8. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 8 Serverless Challenges Monitoring and debugging. Troubleshooting is Painful Cloud provider limitations requiring workarounds (e.g. cold starts, concurrency) Needs Significant Mindset Shift Serverless can quickly become expensive Vendor Lock-In. Proprietary services Migration requires considerable refactorization
- 9. Addressing Serverless Challenges Serverless can quickly become expensive Vendor Lock-In. Proprietary services Instead of trying to avoid vendor lock-in, focus on switching costs & migration effort Know which use cases are best suited for Serverless and take the Total Cost of Ownership into account
- 10. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 10 Case Study 1: Event Based Cloud Native Architecture for instant search, real-time data anomaly detection, and analytical solution with microservices for ingesting and processing 1 billion+ records annually Our experience building a Serverless Search & Analytics solution for an Investment Bank Kafka Users On-Premise AWS Cloud On-Premise Region VPC Amazon DynamoDB ElasticsearchAmazon Kinesis Data Streams Amazon Kinesis Data Firehose Parquet (Archiving) Analytics rules Cleanse, Add to ES Enrich Search, Fetch from ES Detect anomalies /search /alerts AWS Direct Connect Kibana Realtime Analytics Dashboard
- 11. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 11 Case Study 2: Hybrid Cloud Native Architecture with ECS for workflow, real-time data validation, and Informatica for ETL Key Differentiators Traditional SQL/Mongo: • $XXX - $XXXX per month fixed • Infrastructure provisioning and regular maintenance required AWS Dynamo: • $1.25 per million write request • $0.25 per million read request • Auto scaling • Zero maintenance Flat files S3 Raw Bucket Informatica Data Validation No SQS Raw Queue Lambda Invoke Informatica S3 Error Bucket Is Valid? DynamoDB Error Metadata Apply Rules Yes S3 Success Bucket DynamoDB Curated Metadata Glue S3 Catalog Athena Curated Data ECS Container Microservice Downstream Consumers Analytics Rules Ingress SQS Success Queue SQS Fail Queue Lambda Add to S3 Lambda Add to S3 & Dynamo Step 1: Raw Data Ingress Step 2: Data Validation Step 3: Data Curate Step 4: Data Consumption UI Our experience building a Hybrid Serverless Workflow solution for an Insurance provider
- 12. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 12 NoOps in a Serverless World
- 13. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 13 Why NoOps: Shift IT’s focus from Operations to Outcomes Serverless Umbrella term for a spectrum of cloud-based options available to organizations wishing to get out of the business of managing servers NoOps Hyper-automation of cloud computing where core systems administration tasks are fully automated setting talent free to focus on driving business outcomes …accelerates… …enables… Maintaining IT systems, backups, patching, managing data, network, and handling other essential but time-consuming tasks suck up considerable chunk of IT budgets. Serverless and NoOps seeks to change that by automation
- 14. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 14 NoOps is the next evolution of DevOps – A North Star DevOps is a cultural shift that promotes collaboration between operations and development teams and demands a linked toolchain of technologies to facilitate faster delivery of change DevOps The next evolution of cloud computing where developers don’t need to interface with operations team because underlying system infrastructure and security management tasks are fully automated maximizing development time and speeding up release cycles NoOps Driven by Collaboration Driven by Self-Service Infrastructure Automated Dev Ops Dev ∞
- 15. Road to NoOps in a Serverless World Real Time Observability, Telemetry, and Monitoring • Setup tracing tools to visualize and track the real-time performance of services • Integrate tools for real-time high- resolution metrics to spot outliers quickly. Down to the exact invocation • Use custom labels for intelligent logging, filtering and advance searching • Setup auto-triaged alerts based on custom triggers End-to-end App Automation • Leverage deployment templates & frameworks with built-in best practices to create functions, APIs, databases, and event source mappings • Automate backups and disaster recovery mechanisms through intelligent routing • CI/CD toolchain integration A large event driven serverless application may have 100s of services firing off millions of events per minute. NoOps is critical in ensuring application sustainability during test, deploy, log and monitor Automate Security • Setup rules engine for cloud security, cost optimization, and governance • Automate security guardrails ensuring real time enforcement of security policies (e.g. encryption, access requirements) Local Debugging And Testing • Develop and test locally in prod-like environment • Tighten development loop by finding issues locally before deploying to the cloud.
- 16. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 16 Leverage CNCF’s Serverless Landscape to find and build a robust tool ecosystem Road to NoOps in a Serverless World: Selecting the right tools
- 17. Example: Automated Build, Integration, and Deployment workflow using a Cloud Agnostic Framework App automation example using Serverless Framework
- 18. Our Perspective on pursuing Serverless Serverless Use Cases Most Suited • API driven apps (Web, Mobile, IoT) • Real-time data processing & analytics • Unpredictable workload (batch, chat bots) • Internal tooling and operations • Modernizing legacy applications through refactorization Least Suited • Always-on constant workloads • Long-running functions • Unsupported language/runtime • Legacy apps w/o sufficient refactoring • Serverless = Technology + Mindset shift The first thing you should consider when adopting serverless is your people • Selecting right cloud provider(s) and tool ecosystem Leverage fully managed services Prioritize frameworks and create reusable templates • Integrate end-to-end automation early on CI/CD tool chain, Backup & DR Application observability with logging, metrics, and tracing Setup alarms and notifications • Know Serverless patterns and anti-patterns Design event driven and asynchronous solutions Factor in technology limits & pricing for specific services Adopt the principle of least privilege for granular security • Continue learning & adapting. Serverless evolves rapidly
- 19. Copyright © 2019 Deloitte Development LLC. All rights reserved. | 19 Serverless is gaining significant momentum to drive Cloud Enabled Innovation in the market place By 2022, most platform as a service (PaaS) offerings will evolve to a fundamentally serverless model, rendering the cloud platform architectures dominating in 2017 as legacy architectures Gartner Forecast 2017 Serverless Architecture Market worth $14.93 billion by 2023 Billions Source: MarketsandMarkets 46 percent of IT decision makers are already using or evaluating serverless computing Source: Cloud Foundry Foundation
- 20. 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. Thank you. Gary Arora Senior Manager | Chief Cloud Architect GaryArora@deloitte.com