Ruby Video
Talks
Speakers
Events
Topics
Leaderboard
Sign in
Talks
Speakers
Events
Topics
Use
Analytics
Sign in
Suggest modification to this talk
Title
Description
RubyConf AU 2015: http://www.rubyconf.org.au Do you need Ops in your new startup? If not now, then when? And...what is Ops? Learn how to scale ruby-based distributed software infrastructure in the cloud to serve 4,000 requests per second, handle 400 updates per second, and achieve 99.97% uptime – all while building the product at the speed of light. Unimpressed? Now try doing the above altogether without the Ops team, while growing your traffic 100x in 6 months and deploying 5-6 times a day! It could be a dream, but luckily it's a reality that could be yours.
Date
Summarized using AI?
If this talk's summary was generated by AI, please check this box. A "Summarized using AI" badge will be displayed in the summary tab to indicate that the summary was generated using AI.
Show "Summarized using AI" badge on summary page
Summary
Markdown supported
In the presentation titled "DevOps Without The 'Ops' – A Fallacy? A Dream? Or Both?" at RubyConf AU 2015, Konstantin Gredeskoul explores the concept of managing software development and operations without a dedicated Ops team. The discussion arises from his experience at Weena, an innovative platform handling substantial traffic with minimal operational hassles. Gredeskoul argues that startups can thrive without a conventional Ops team and outlines a modern approach to scaling applications effectively. Key points discussed include: - **Redefining DevOps**: Gredeskoul presents various perspectives on DevOps, indicating that many organizations struggle to grasp its implications. Traditional methods bind dev and ops with strict protocols, yet flexible approaches may work for less regulated industries. - **Scaling Traffic**: He highlights the importance of understanding scaling challenges as traffic increases. Initial hundreds of requests may snowball into thousands or even millions, necessitating an agile response. - **Culture of Ownership**: At Weena, he asserts that engineers were not just coders but were also responsible for operational tasks, blending roles to enhance resilience and reduce reliance on a dedicated Ops team. - **Infrastructure as a First-Class Citizen**: Viewing infrastructure configuration as equally crucial as feature development can streamline operations. Continuous monitoring and incremental deployment strategies help maintain uptime, while fault-tolerant architecture ensures stability. - **Utilizing Modern Tools**: The use of cloud services and automation tools like Chef allowed for significant traffic handling and uptime performance without a traditional Ops structure. - **Real-World Examples**: Gredeskoul's insights about their infrastructure detailed how their use of HAProxy and other proxies contributed to system speed and reliability by efficiently managing backend requests. - **Monitoring Metrics**: He advises on focusing alerts on significant business metrics to prevent overwhelming engineers, a practice that allows for greater operational efficiency. To conclude, Gredeskoul emphasizes that startups can operate effectively without dedicated operations if they foster a culture where everyone shares responsibility for both development and operations. By leveraging automation, focusing on fault tolerance, and employing modern cloud technologies, companies can scale up efficiently while maintaining high service availability.
Suggest modifications
Cancel