PAC 2020 Santorin - Hari Krishnan Ramachandran
•
0 likes•144 views
This document discusses common performance myths and challenges related to cloud migrations, SaaS deployments, and cloud native applications. It provides approaches to assess performance for each scenario. Some key points covered include: - Common myths like unlimited scaling without effort, cost savings from cloud alone, and SaaS vendors handling all performance. - Challenges for migrations include capacity planning, auto-scaling issues, and latency. For SaaS, risks include customization impacts and integration issues. Cloud native challenges are distributed architectures and configuration problems. - The document recommends approaches like baseline benchmarking, resilience testing, and 360-degree validation across user, cloud, data center and external layers to overcome myths and address challenges.
Report
Share
Recommended for you
Recommended for you
More Related Content
What's hot
What's hot (20)
Similar to PAC 2020 Santorin - Hari Krishnan Ramachandran
Similar to PAC 2020 Santorin - Hari Krishnan Ramachandran (20)
More from Neotys
More from Neotys (20)
Recently uploaded
Recently uploaded (20)
PAC 2020 Santorin - Hari Krishnan Ramachandran
- 1. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L How to Over come the Performance Myths in Cloud Hari Krishnan Ramachandran Cognizant
- 2. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Why Is preferred? High CAPE PEX Lowto Market Faster toTime Unlimited Infrastructure to SCALE Improved &Security Availability Increased SERVICE Coverage Available VERYWHEREe
- 3. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L How enterprises try to adopt Cloud? CLOUD MIGRATIONS (Existing App.) CLOUD DEPLOYMENTS (SaaS App.) CLOUD DEVELOPMENT (New App.) Migrating the Infrastructure (Re-host) Migrating the application with considerable code / arch. changes (Re-Architect) Cloud migrations from one provider to another COTS products deployment Direct cloud application deployments with vendor recommendations With minimal adoption of cloud components such as database, message queue and etc. Develop application from scratch using cloud tools Utilize cloud based development and deployment tools Cloud First Approach
- 4. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Common Performance Myths in Cloud Applications on Cloud will scale without hassles & Performance Testing is not relevant Moving to Cloud provides unlimited computing capacity and storage, and it will improve the responsiveness of the software applications when accessed by large number of users Moving to Cloud will save the cost Migrating all the application simply to the cloud will reduce the operational cost and will not consume additional hardware resources SaaS Vendors will take care of performance SaaS product owners will perform the right load and scalability tests that are required to meet the business performance demands of the applications SaaS Adoption will never impact “End User Experience” SaaS product integration will have no change in the performance of the applications that are accessed by business and end users Cloud Native is Cost Saving Adopting cloud native solutions such as serverless / containers will reduce the overall operating cost Performance monitoring for Cloud native is irrelevant Cloud native solutions will operate on high performance environments so application will perform better than on premise configurations
- 5. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Cloud Migrations - Challenges & Approach Capacity PlanningApplication Code Issues Cloud resources capacity planning may not be appropriate during migrations Not just the infra. Code level problems such as excessive heap usage and memory leak issues Typical Approach for Migration Performance Assessment Cloud Migration Assessment - Baseline and Benchmark Evaluate Elasticity & Auto-scaling End User Experience/Latency Impact tests Evaluate Resiliency Auto scaling Problems No proper auto scaling configurations can increase the delay between resource transfers Latency Issues Internet facing applications may not work as expected from various business locations
- 6. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Cloud SaaS Deployments - Challenges & Approach SaaS Customization Risks Any customization done on SaaS products needs to be completely tested Integration Risks Performance problems related to the SaaS application's interoperability and integration with homegrown applications located on premise Understand the performance specific SLAs signed with SaaS provider 360* Validation strategy – Component level, End to End, Customer Exp. Latency assessments On-prem Infra monitoring for Hybrid- SaaS Test Environment for SaaS Products Availability of test environment and limited configurations can make the performance results inaccurate Resiliency Abstraction SaaS applications will not be tested for resiliency due to the infrastructure access limitation thus highly vulnerable during user load spikes Typical Approach for Performance Assessment
- 7. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Cloud Native Apps – Challenges & Approach Highly Distributed Arch. Concurrency Issues Configuration Problems Deployment Strategies Cloud native are highly distributed in nature and requires effective measurement techniques Unexpected peak user load conditions could break the applications / pods / containers in production Improper configuration with no proof of performance assessments can have scalability errors Validation of deployment strategies for MicroServices should be completed with performance tests to understand the impact Typical Performance Assessment Approach Isolated Performance Assessment for Microservices / Cloud Components Single user and Multi User Performance Assessment Resilience Validation Tests Operational/Cut-over tests
- 8. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L 360* Cloud Performance validation END USER / DEVICE LAYER CLOUD LAYER DATA CENTER LAYER EXTERNAL LAYER Users Network: 3G / 4G / Broadband Browsers: Chrome / Firefox / IE and mobile versions Devices (Mobile / Desktop): Windows, Mac, Android, IOS AWS - EC2, AWS RDS, AWS Lambda Azure - Compute ,Storage, AKS , Cosmos GCP – Compute Engine , kubernetes Engine , IoT Core Web App DB IntegrationLayer IntegrationLayer Data Center APIs Enterprise Data Internet End-to-End Testing & Monitoring Using Cloud Specific & Enterprise Tools Legacy Systems INTERNET / PUBLIC NETWORK CLOUD PROVIDER NETWORK DATA CENTER NETWORK EXTERNAL SERVICES CI / CD using Jenkins & Docker Solutions
- 9. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Key Considerations for Cloud Performance ON - PREMISE [Different Cloud compute components, storage, network, tools & utilities.] Cloud Technologies & Platform CLOUD [ NFR Gathering specific to Cloud, Cloud Specific Tools, Cloud Performance Test & Engineering Methodology ] People Enablement Technology Process Tools People Testing & Monitoring Tools [ Cloud specific / Native testing and monitoring tools and open source solutions ] Non-Functional Requirements & Strategy [ Cloud based NFR’s (Latency, auto scaling, resource utilization, DR Plan) and Test strategies for cloud migration, SaaS based apps, cloud native development ]
- 10. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Some recent Cloud outages Google Cloud experienced a large-scale outage on Sunday, June 2, depriving millions of people of access to Gmail, YouTube, and Drive, and even causing third-party entities like Vimeo, Snap, and Discord to take a hit Amazon AWS experienced a power failure leading to the datacenter’s backup generators to start failing. It led to 7.5% of the EC2 instances and EBS volumes becoming unavailable In May, Microsoft had to face an outage that lasted for more than an hour. Microsoft Azure that deeply affected its cloud services which are widely used by Microsoft’s commercial customers. Cloud Platform Outage Facebook and Instagram faced outage around 14 hours earlier this year which was caused due to a server configuration change Social Network Outage Retail giant Costco (COST) got a rude and costly awakening Thanksgiving Day when its website went down for more than 16 hours. It costs the retailer nearly $11 million in lost pre-Black Friday potential. Amazon’s 2018 Prime Day glitch where it couldn’t handle traffic loads costs the giant up to $72 million. Application Outage Salesforce at 2019 left their company data unsecured, it took 3 days to fix and the sales & marketing teams came to a screeching halt during this system failure SaaS Product Outage A widespread Email outage across European data-centers that persisted over a day or two In March of 2019 Google experienced a global outage that affected its Gmail and Google Drive services, it affected Gmail Users for 4.5 hours
- 11. P E R F O R M A N C E A D V I S O R Y C O U N C I L byP E R F O R M A N C E A D V I S O R Y C O U N C I L Thank You