Three critical concerns about Linux container Technology

Three critical concerns about Linux container Technology
GuideOver the past two years, Linux's container technology has dominated the world's enterprise IT market, and for good reasons-in addition, the problem they solve is that virtualization cannot be used in large-scale applications in the application software development and computing industries, and allow the enterprise world to truly embrace DEVOPS concepts and microservices (years of service-oriented architecture dream have passed ).

As an extension of the operating system, especially a part of the Linux kernel extension, Linux containers can be created and formatted at runtime and then used to coordinate network and storage resources on hosts across clusters. The final result is a set of lightweight, dynamic, and secure application services, each of which is contained in a Linux container, and can run with itself or with applications of other containers to create more flexible, but more complex enterprise applications.

These capabilities of containers present significant competition threats to some of the current IT industry segments, and are another heavy blow to the proprietary computing stack (mainly for Linux), which is obvious. However, Linux containers and the Docker container technology, which are closely related to them, are also facing some undeniable threats, even before these technologies mature. At present, the views on these technologies are quite positive, so what is threatening the future of container technology?

In general, there are three issues related to Linux container adoption and containerized infrastructure deployment:

1. Standard fragmentation
2. Proprietary code and false Open Source
3. Container packaging ing (container packaging ing indicates containerized packaging and brainwashing of things that may not be related to container, first from cloud packaging ing)

Fragmentation

Fragmentation can be said to be the most dangerous threat to Linux container adoption in the business world. If there are no clear and easy-to-use standards, especially for image formats and orchestration levels, most enterprises will give up embracing this technology. Simply put, no IT decision makers want to support a failure task. For example in the Consumer World, suppose someone has stored a full warehouse all with HD-DVD players and accessories, and then finds that Blu-ray has won the final market. The loss will be exponential growth, which is the reality of the Enterprise World and the technical standard for Linux containers.

Although fragmentation is currently the most serious threat to Linux container adoption, it is also one of the most easily solved problems. Two new foundations controlled by the Linux Foundation have been established, and they have recently started to run to help eliminate the haze of fragmentation in the Linux container world.

Although these two organizations are a fantastic start, from the perspective of the complexity of the complete lifecycle of containers, they need to be continuously concerned from a standard perspective. With the continuous improvement of technology and adoption for enterprises, it is very likely that there will be more "Battle of formats" types of melee, on a large scale, it depends on how the entire open-source community helps regulate and compile these arguments to form general standards and practices. With the initial development of the two foundations and the best practices of the Post-open source community, we can hope that standardization will continue to open up without compromising or slowing down innovation.

"Semi-open" actually means "fully closed"

The threat of fragmentation also raises another important question. When talking about containers, it actually contains open-source kernels or peripheral products of "fake open-source" Linux containers. Despite all the "Linux" names, containers can be widely used in proprietary and open stacks. This will cause problems. When proprietary code and services gradually enter the containerized solution based on their methods, they will be considered completely open. The threat of "false openness" is not a new thing. First, we have seen it in Unix and most recent cloud computing solutions, especially PaaS and openstack-based solutions, however, the top layer is based on proprietary open source technologies.

Although Linux containers are still in the early stages of the ecosystem adopted by the enterprise IT market (although the pace is accelerating ). If the proprietary technology is applied to the containerized technology, and this is almost certainly the case, then the sentiment in the IT field will become more appealing. Is this an innovation or a continuation of the old proprietary system? From a closed protocol stack to a high license, innovation will be greatly reduced, adding fake open code based on the creation of blood, sweat, and tears in the open community will quickly suppress open source enthusiasm and innovation.

Container loading ing

In the exciting days of the cloud computing boom (it can be said that this boom is still in progress), the concept of cloud cleaning was born. In fact, an IT supplier will adopt marketing materials based on existing products and "processing ing", and use the specifications table and other cloud words for packaging to persuade customers and potential customers, this existing supplier is a player in the emerging world of cloud computing.

Now, we see the same occurrence of iner loading, which is slightly related to the container boom (or completely unrelated) suppliers and solutions are trying to enter the market in their way. The threat here is far greater than fragmentation or false open source. It is a threat that subverts what the container actually is.

For example, the concept of containers and virtual machines is confusing. Of course we can run a container on a virtual machine or reverse, but the two technologies solve different problems. The combination of virtualized infrastructure services and application code provides abstraction. containers are the best way to allow clean isolation and ideal calls in lightweight software environments, serving as the main service delivery method.

These three are currently the largest, but not the only ones. We are facing a threat to the growth of the Linux container technology ecosystem. This does not mean that these problems will prevent the use of container technology and the fragmentation issue has been located, in addition, IT leaders have experienced the war between Unix and Cloud, and can understand how to handle the differences between open-source core and offline packaging products with caution. However, it is important to remember that the path to innovation is always full of thorns. The development of Linux is the same as that of cloud computing. Now it is the turn of Linux containers.

Ultimately, IT depends on open-source communities, business circles, and startups, and builds IT equipment vendors to overcome these obstacles by innovating around Linux containers, through cooperation and dedication, container technology can truly realize their enterprise application potential.

Original article address: http://geek.csdn.net/news/detail/59741author: Wei

Reposted at: http://www.linuxeye.com/linux/2881.html :linuxeye

This article Reprinted from: http://www.linuxprobe.com/linux-container-technology/