HULK Virtualization Team 360 Cloud Computing
Heroine declaration
This article is from the HULK virtualization team. This article is mainly a record of the team’s exchanges and sharing with Autohome and Sina. The overall sharing is different from the type of conference (the nature of blowing bi), and it is more grounded, sharing the scope of the topic More extensive, including: asset management, monitoring, code release, containerization, virtualization, hybrid cloud, intelligent operation and maintenance, etc. This article mainly throws out some concerns of our team first, if you are interested in related topics , You can leave a message to reply, and we can communicate in detail.
PS: Rich first-line technology and diversified forms of expression are all in the " HULK first-line technology talk ", please pay attention!
Background introduction
In the spirit of openness and win-win, the addops team organized and participated in the technical exchange and sharing sessions with "Car Home" and "Sina". This exchange is different from the traditional technology conference. We only discuss dry goods and have very detailed discussions on specific technical points.
In the technical sharing and discussion with these two teams, we "colliding with the spark of thinking" and "learning from each other", it can be said that we have gained a lot. At the same time, the operation and maintenance teams of other companies are also welcome to actively contact us, communicate together, and grow and learn from each other.
1
Car home
We invited the operation and maintenance team of Autohome to come to our company for technical exchanges. First of all, the "Zhe Ye" of our team introduced the hulk cloud platform, focusing on some CMDB-related designs, especially how to handle various complex relationships and processes gracefully when encountering complex business lines.
Then the students from Autohome shared their CMDB system, in which the graphical display of the cabinet network topology was eye-catching.
Moreover, the students from Autohome also shared their "code publishing system", the overall architecture design is "simple and elegant", and they undertake concurrent tasks with thousands of nodes.
Our team shared the experience related to containerization, how to build a set of highly available container architecture based on k8s to meet the needs of users in different scenarios.
Afterwards, the students from Autohome shared some experiences of using radar to manage redis clusters.
Our team shared the "wonder" monitoring system based on the secondary development of open-falcon, which can support tens of thousands of nodes and provides flexible alarm and monitoring strategies.
After listening to our sharing of the monitoring system, the car home students also shared their monitoring system.
Everyone had a lot of discussions on the design and future of the monitoring system. The "directed acyclic graph" analysis of Autohome has left a deep impression on us to solve business relationships and complex business dependence on monitoring.
In the end, the sharing ended in "kind and friendly" and "unfinished". The teams of both sides exchanged technologies through sharing and learned a lot.
Sina trip
After the technical sharing meeting with Autohome, we received an invitation from Sina to exchange and share with Sina's O&M and development classmates.
We received warm welcome and hospitality from Sina classmates, and at the same time visited the Sina headquarters building just settled in. "Food is the heaven for the people". Before sharing, we first tasted Sina's delicacies. Sina's buffet working meals are rich in variety and full of appetite.
After dinner, under the leadership of Sina O&M students, our group visited the newly completed headquarters building, which is very spacious and well-equipped and convenient.
Let's enter the topic and start sharing. First, we shared some management experience and architecture solutions for 4/7 layer load balancing.
Then there are some experience in containerization.
The meeting room was full of people, the scene was hot, and the discussion was very lively
Sina students shared the "Docker-based hybrid cloud architecture and application practice" on Weibo, through the self-developed DCP system, to solve the deployment and monitoring problems of containerization in the hybrid cloud.
Next, we shared the application and evolution of virtualization technology in business.
Finally, we shared some practical experience of machine learning in operation and maintenance.
In the end, our team ended this trip to Sina with "full harvest". Thanks again to Sina O&M students for their invitation and hospitality.
By the way, when we were leaving, we heard that Wu Yifan and Xing Ye came to Sina, but the friends felt that it was important to go back to work as soon as they were out for a day, so they hurriedly returned to the company.
Conclusion
Based on the company's business, our addops team actively improves the service and quality of operation and maintenance through the hulk cloud platform, and continues to research and innovate in technology. We welcome technical colleagues to communicate with us and make progress together.