Shulou(Shulou.com)05/31 Report--

This article focuses on "how to remake cloud services with Docker and Kubernetes". Interested friends may wish to take a look. The method introduced in this paper is simple, fast and practical. Let's let the editor learn how to redo cloud services with Docker and Kubernetes.

Redo cloud service

In my previous blog, I wrote about how open source software is delivered as a service, rather than looking at it from a binary point of view. These open source services deliver a free, open, flexible service that engineers use to help create their applications without solving the problem of keeping the software running.

Today we're going to talk about a further topic, a new generation of cloud services: Anywhere Cloud Services, the cloud that is placed everywhere, thanks to Docker and Kubernetes.

Where are the cloud services now?

Today's application-level cloud services, such as AWS's Kinesis or Google's BigTable, provide comprehensive management services to engineers, where you can create an application without effort maintenance. These services are not only secure, automatically scheduled, monitored, logging in, but also integrated and updated. Personally, I love these services. But.

You can lock down these services through code (for example, you need to modify the software when modifying database backups)

Choose a public cloud service and also require the mandatory use of their hardware. So, what if your application runs on Microsoft Azure, but selfishness prefers Big Table?

What if you want to use open source software, but do not want to operate, schedule, or provide security measures for open source software?

Let's go into Anywhere Cloud Service.

Model No.2--Anywhere Cloud Services

The anywhere cloud service service is a comprehensive management service, which can be delivered anywhere, and has three characteristics:

Full hosting-users don't have to worry about keeping the service online. The service provider will assume this responsibility.

It can run anywhere-the user chooses where the service is located. It could be one of the major cloud providers, or a private cloud with standard API. There is a clear demarcation in the running infrastructure:

Be treated as service, not software-- to provide a function to the user in an obviously consumable way. For example, a No SQL service provides a way to inject data and parse the data. This service will heal itself. It dispatches itself when it is needed. It provides users with metrics, logs and alerts. This is also highly available. Installing this service is as easy as running a command line or clicking a button.

Give me some examples.

To help illustrate this idea, here are some examples we may see in the near future:

Any non-infrastructure on AWS, such as No SQL DynamoDB services, can be provided on all other cloud providers. For example, Azure, Google Cloud and so on.

Google's non-infrastructure services, such as their PaaS App Engine, can be deployed on any public cloud or even to a local data center via openstack.

With the Docker container space, startups have created services that can run your application / code on any public or private cloud.

Because they wanted to benefit from open source software, the development team decided to use the data stacks in Kafka,Spark and Cassandra. However, they do not want to hire their own consulting firms or operate on their own. As a result, some startups specialize in providing these software in any desired location, and they are using new services made by these startups.

The creators of these open source software package their software so that in addition to working with source code and binaries, the software can be as easy as configuring services. These companies will also provide "supported" versions of these service at a cost.

All of this is possible, thanks to Anywhere Cloud Service.

So, how do we do Anywhere Cloud Service (ACS)

There are two major obstacles to reaching ACS:

1. How to package and configure soft services on the premise of hardware across multiple cloud service providers in a standard and effective way.

2. How to package, run, configure, schedule and ensure the security of your service in an effective way.

Enter the link of Docker and Kubernetes

Docker allows us to quickly create, package and run software on multiple systems in a special way. By using containers, cloud service providers can run their software in any environment. But this is far from enough. We need a way to get software installed, run, and highly available. We need Kubernetes.

Once the software is packaged in a Docker container, we can turn their software into service by using the capabilities of Kubernetes. The functional goal of Kubernetes is to deal with a range of infrastructure (wherever it comes from) while providing API to create services on top of the infrastructure. Kubernetes includes load balancing, elastic expansion, rolling update, service discovery, namespace, APIs, flexible scheduling and configurable infrastructure and threshold limits, and so on. Using these capabilities, we can create declarative services and give control of those services to the user. Kubernetes improves No and NoOps, while also providing developers with a single API, allowing users to access both public and private clouds.

How do we use ACS?

First, users need to have one or more Kubernetes clusters. Install Kubernetes so that it can run in 5 minutes, or in a few days (depending on your situation). Kubernetes supports most public clouds, openstack clouds and local data centers. Users can also choose whether to run on a virtual machine or a physical machine. We can run kubernetes ourselves, or if we are looking for additional support, we can also use administrative tools to run it.

Note that here, users are in complete control, getting rid of arbitrary locks and controlling where their applications run. Thanks to regular API, these API exist in any infrastructure environment, and users can use their clusters to deliver on multiple cloud providers, or they can create a true hybrid strategy.

Second, users can easily use these services by configuring their code on the user's Kubernetes cluster. Through app store, or through developers, users can search for services that meet their needs, and then click install. The service provider will contact the Kubernetes cluster problem either in the cloud or locally, and install the service. Developers will start using the service immediately. The job of the service provider is to ensure that the service is running, healthy, and scheduled as needed.

In this new world, users have more and more control over their applications and the services they rely on. They can move our applications between clouds, on private and public clouds. Consumers can actively run applications or prepare active-standby across multiple clusters.

At this point, I believe you have a deeper understanding of "how to redo cloud services with Docker and Kubernetes". You might as well do it in practice. Here is the website, more related content can enter the relevant channels to inquire, follow us, continue to learn!

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.