In addition to Weibo, there is also WeChat
Please pay attention
WeChat public account
Shulou
2025-01-16 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Internet Technology >
Share
Shulou(Shulou.com)06/01 Report--
This article mainly explains "what is the difficult problem of Kubernetes application management". The content of the article is simple and clear, and it is easy to learn and understand. Please follow the editor's train of thought to study and learn "what is the problem of Kubernetes application management".
The demand of R & D for Application Management
1. YAML file of K8s all in one
Let's take a look at such a K8s yaml file, this yaml file has been simplified, but we can see that it is still relatively long.
In the face of such a YAML file widely criticized as "complex", I believe everyone can't help thinking about how to simplify it.
From top to bottom, we roughly divide them into three pieces:
One is the parameters related to capacity expansion and rolling upgrade, which should be of more concern to the students who apply operation and maintenance.
The middle piece is related to image, port and startup parameters, which should be of more concern to the developers.
The last piece you may not understand at all, usually do not need to understand, can be understood as the K8s platform students need to care about.
When we see such a yaml file, it is easy to think that we just need to encapsulate the fields inside and expose the ones that should be exposed. Indeed, we have an internal PaaS platform to do this.
two。 Only part of the field is revealed: simple but incompetent
An internal PaaS platform carefully selected some fields and made a beautiful front-end interface for users, which only gave users about 5 fields, which greatly reduced the mental burden for users to understand K8s. Then the underlying implementation renders the user's five fields into a complete yaml file in a template-like way.
The highlighted fields are roughly shown in the following figure:
I have to say that this method is very effective, for simple stateless applications, streamlining API can greatly reduce the threshold of K8s, fast and efficient docking users, PaaS platform is also smooth for everyone to use. At the same time, I also learned from some technology sharing that many other companies have simplified K8s API in a similar way.
But when the user's business begins to dock on a large scale, we will naturally encounter complex stateful applications, and users will begin to complain that the PaaS platform is not capable enough. For example, the logic of selecting master and master-slave switch for multiple instances of Zookeeper is difficult to expand in these five fields.
In the final analysis, the way of masking a large number of fields will limit the evolution of the infrastructure itself, but the capability of K8s is very powerful and flexible. We can't give up the powerful capabilities of K8s for the sake of simplification.
For example, in the current example, it is easy to think that for complex stateful applications, it should be solved through CRD and Operator in K8s.
3. CRD+Operator: K8s has strong scalability but is difficult to use.
Indeed, when we internally interface with complex applications, we also recommend them to write Operator, but there is often a conversation like this.
The engineer of the middleware said to us, I have a Zookeeper here. What kind of K8s Workload should I use to access it? We thought that K8s was so delicately designed that there was no problem that could not be solved, so we recommended them to use Operator. They were confused and said that your ability to create new words in the past few years was absolutely excellent, and they had never heard of it before.
Come to think of it, it is not difficult for the business side to understand these new concepts, but it is still very difficult to implement them on their own. We naturally feel that the business side should focus more on their business itself, so we have to help them write together.
It can be seen that we urgently need a unified model to solve the demand of R & D for application management.
Thank you for your reading, the above is the content of "what is the problem of Kubernetes application management". After the study of this article, I believe you have a deeper understanding of what the problem of Kubernetes application management is, and the specific use needs to be verified by practice. Here is, the editor will push for you more related knowledge points of the article, welcome to follow!
Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.
Views: 0
*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.
Continue with the installation of the previous hadoop.First, install zookooper1. Decompress zookoope
"Every 5-10 years, there's a rare product, a really special, very unusual product that's the most un