Where Should the Container Ecosystem be Heading this Year

In making the container ecosystem experts are trying to take the developers away from the core infrastructure where damage can be done. They are putting policies and rules around the way in which they deal with continuous integration, continuous development and continuous testing. By placing things in operations, the abstraction also continues and there is still monitoring the containers to this abstraction layer which is difficult.

There is still an ongoing talk on how the infrastructure itself works. For example if the consumer wants a different service mechanism than what the provider is offering, they can enable you, the consumer to plug that service in and use it. If you want a different way of storing data on different types of devices they want to be able to provide that.

You want to create a system that’s very modular. You want to allow people to take out certain solutions and plug in other solutions and be able to do that in a fairly straightforward way without causing a lot of pain.

Container ecosystem is moving quickly. One of the focuses on the ecosystem development needs to be on the industry agreeing on the right primitives because it needs to be built from the bottom up, and when you take things like governance, if creators are not working in a declarative way, it becomes difficult to model policy. It also becomes difficult for different vendors to provide solutions around governance if the primitives are not agreed upon.

Better understanding of CI (Continuous integration) and CNCF (Cloud Native Computing Foundation), where things are headed with regard to container run times, image formats and cluster managers also need to be sorted out in coming months. Then you can move the focus of everyone on to the next set of problems.

You can’t always wait for the standardization bodies to come up with the best possible outcome, but weighing solutions on technical merits will at the end result in the right solution getting adopted. One year from now probably we will see an evolution, and either organic or inorganic the community will start gathering around some of these options.

It’s good and healthy that there are several options now for essentially every level of the containers. From the runtime to the orchestration. It’s clear now that containers are here to stay and that there are still many unsolved problems with containers. Fifteen years ago the same thing happened with virtual machines and VMware became the standard for virtualization by winning. A strong ecosystem was created that is still alive now with many vendors and many winners. The same thing is going to happen with containers especially to make the enterprise comfortable and cozy with using containers.

Right now you can compare the containers to the industrial revolution of cars. Cars were invented and nobody drove them and then the industrial revolution hit and everybody had cars. Container technologies and technologies like container technologies will give us this advent of clouds of clouds appearing to where there are millions of clouds around the world running, and where you have a SaaS (Software as a Service) in a particular cloud, multiple SaaS talking to each other inside those clouds and then there are tens of thousands of those clouds.

Then there will be the emergence of immutable infrastructure operationally and we can see that starting to emerge out of the OS and out of the containers and we’ll see it continue go up.

The promise of containers has always been portability and the idea that you can deploy your application in one place, in one area of different space of possible runtime environments, and that you can also deploy your containers in the other axis of different stages in the software development lifecycle and move them around easily between those areas. We are already seeing customers asking for multi cloud and hybrid cloud deployments. It’s apparent that containers are the key to unlock that. However there’s many much more work to be done to truly enable them.

From a point of view of standards and plug ability it’s important for an ecosystem to have open standards. One example is Docker plugins which succeeded in getting a plugin system in the standards led by successful implementations and rough consensus and working code, and on that basis and on the basis of technical merit we can create a better ecosystem for everyone and reduce duplication of effort.

UseFull Links




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s