[Seminar co-sponsored by Yahoo!] MixLeap Study #68 - Learning about Kubernetes operation from case studies of various companies
2021.10.28
Beyond will collaborate with the event "Mix Leap Study" operated by Yahoo Japan Corporation to hold an online study session on the theme of Kubernetes
Date and time
2021/11/18 (Thu)
18:55 Live streaming information
19:00 Start online
About the event
"Mix Leap Study" is a study session held at Yahoo's Osaka office with the aim of helping creators (engineers and designers) in the Kansai area grow by exposing them to various Yahoo's proprietary technologies and the industry's cutting-edge technologies. is.
We aim to become a crossroads of people and information by bringing together students interested in IT and working adults working at IT companies to share information with each other.
This time's theme
This time, it is a study session with the theme of "Kubernetes operation".
Three engineers who are in charge of Kubernetes operations in their daily work will talk about their use cases, operational issues, future developments, etc.
In the second half of the event, there will also be a panel talk on the theme of "Kubernetes operation challenges" submitted by participants.
- Considering using/introducing Kubernetes
- I am involved in Kubernetes and would like to know about other companies' case studies.
This event is especially recommended for engineers like you!
time table
Presentation time: 15 minutes / Q&A: 5 minutes
| time | Content | speaker |
|---|---|---|
| 18:55 - 19:00 | Live streaming information | |
| 19:00 - 19:05 | Opening explanation | Michitaro Nakajima (Yahoo Japan Corporation) |
| 19:05 - 19:25 | “About the behavior of Pods and Worker Nodes during high loads in GKE” | Yuki Teraoka (Beyond Co., Ltd.) |
| 19:25 - 19:45 | "Learn Kubernetes settings by comparing manifests" | Keisuke Yabuta (Yahoo Japan Corporation) |
| 19:45 - 20:05 | “Kubernetes operation near-miss case study” | Takuya Yamada (Yahoo Japan Corporation) |
| 20:05 - 20:25 | panel talk | |
| 20:25 - 20:30 | closing |
*Depending on the progress, the end time may vary slightly.
Speaker information
Speaker: Yuki Teraoka
Joined Beyond in 2016 and is currently in his 6th year
as an Infrastructure Engineer MSP, where he troubleshoots failures while also designing and building infrastructure using public clouds such as AWS.
Recently, I have been working on building container infrastructure such as Docker and Kubernetes, and using Hashicorp tools such as Terraform and Packer as part of operational automation, and I also play the role of an evangelist who speaks at external study sessions and seminars.
| title | About the behavior of Pods and Worker Nodes under high load in GKE |
|---|---|
| overview | When using Kubernetes to run some kind of service or system, you will generally use managed services provided by AWS, GCP, etc. This time, I will use GKE as an example to organize the behavior of Pods and Worker Nodes when the Cluster is under high load during operation, and talk about how to deal with and follow the load. Masu. |
Speaker: Keisuke Yabuta
Joined Yahoo in 2017.
Engaged in the development and operation of platforms that support Yahoo's services. Focused
on back-end development, examining system and network architecture, and recently also involved in front-end development.
| title | Learn Kubernetes settings by comparing manifests |
|---|---|
| overview | I will introduce examples of two products that handle Kubernetes, and I will talk about why they are the way they are based on differences in their manifests, and what settings should be done and what settings should not be done. |
Speaker: Takuya Yamada
After joining Yahoo in 2017, he was in charge of developing and operating an image distribution platform.
He is currently responsible for developing the Approximate Neighborhood Search Engine Platform.
| title | Case studies of near misses in Kubernetes operations |
|---|---|
| overview | I would like to introduce a near-miss experience that particularly left an impression on me during my three and a half years of operating and developing Kubernetes. I would like to look back at the causes and countermeasures taken at the time, and share some know-how to prevent problems. |
