- DevOps：Continuous Delivery，Integration，and Deployment with DevOps
- Sricharan Vadapalli
- 424字
- 2021-07-02 16:17:55
DevOps application - business scenarios
Application of DevOps varies for multiple scenarios, with accrued benefits as listed:
- Automation of development cycle: Business needs are met with minimal manual intervention, and a developer can run a build with a choice of open tools through a code repository; the QA team can create a QA system as a replica, and deploy it to production seamlessly and quickly.
- Single version of truth - source code management: There are multiple versions of the code, but it is difficult to ascertain the appropriate code for the purpose. We lack a single version of the truth. Code review feedback is through emails and not recorded, leading to confusion and rework.
- Consistent configuration management: We develop, test, and build source code on different systems. Validating the platforms and compatibility versions of dependencies is manual and error-prone. It's really challenging to ensure all the systems speak the same language, and have the same versions of the tools, compilers, and so on. Our code works fine on build systems but doesn't when moved to production systems, causing embarrassment regarding business deliverables, and cost overheads to react.
- Product readiness to markets: We have a process to develop code, test, and build through defined timelines. There are many manual checks and validations in the process; the integrations between different groups cause our commitments and delivery dates to be unpredictable. We wish to know how close our product is to delivery and its quality periodically, to plan in advance rather than being reactive.
- Automation of manual processes: We are following manual processes, which are often error prone, and wish to enhance efficiency by following an automation process wherever applicable. Testing cycle automation, incremental testing, and integrating with the build cycle will expedite product quality, the release cycle, and infrastructure service automation such as creating, starting, stopping, deleting, terminating, and restarting virtual or bare-metal machines.
- Containers: Portability of code is the primary challenge. The code works in development and QA environments, but moving to production systems causes multiple challenges such as code not compiling due to dependency issues, build break down, and so on. Building platform agnostic code is a challenge, and maintaining multiple platform versions of development and QA platforms is a huge overhead. Portable container code would alleviate these kinds of issues.
- On-premise challenges: We have many on-premise systems. There are multiple challenges, from capacity planning to turnaround time. The Capex and operational expenses are unpredictable. Cloud migration seems to have multiple choices and vendors, so there needs to be an efficient adoption method to ensure results.
推薦閱讀
- ArchiCAD 19：The Definitive Guide
- Seven NoSQL Databases in a Week
- Windows 8應用開發(fā)實戰(zhàn)
- 計算機網(wǎng)絡應用基礎
- Python Data Science Essentials
- Creo Parametric 1.0中文版從入門到精通
- 工業(yè)控制系統(tǒng)測試與評價技術
- 西門子變頻器技術入門及實踐
- Blender 3D Printing by Example
- R Machine Learning Projects
- 學練一本通：51單片機應用技術
- 網(wǎng)絡存儲·數(shù)據(jù)備份與還原
- 工業(yè)機器人操作
- Win 7二十一
- 多媒體技術應用教程