Get best of the week

Hype curve: what IT technologies are at their peak and what will remain stable demand

In 2017, the demand for blockchain developers grew along with the Bitcoin exchange rate. The number of vacancies declined when the market became disillusioned with the new technology. Together with representatives of Mail.ru Cloud Solutions, Red Hat, Acronis, Reksoft, SEMrush and Bilderlings figured out which technologies are popular now and it is worth spending time to study them.

Clouds, containers and microservices


Companies' spending on cloud services will soon cross the $ 1 billion mark. In the future, the market will grow by about 15% annually.

— OpenStack, . , - OpenStack , Red Hat HP, IBM. «» OpenStack.

, Mail.ru Cloud Solutions

Clouds are not only virtualization, much more popular and applied technologies. For example, the development of microservice architectures , their optimization for the realities of the cloud - the application is divided into many small pieces, so the failure of one part does not affect the operation of the application as a whole. This principle works great with containerization : when these parts of applications are packaged in containers, which allows them to be developed and updated independently of each other, as well as to ensure that the application will work the same in both the test environment and the prod.

There are several popular container orchestration systems (container management), for example, Kubernetes is very popular. There are higher levels of abstraction, vendor solutions where Red Hat OpenShift is pushed into the lead. An introduction to containerization technology should start with Docker, building a simple cluster on Docker Swarm. These are basic things, it’s worth learning the basics from them, and then moving on to serious container orchestrators.

Nikolay Butenko, architect at Mail.ru Cloud Solutions

Containers and microservices are promising because they are changing the way they deliver and deploy modern applications. They reduce the dependence on programming languages, as they allow you to write different parts of the application in different languages ​​- most often in Java and Python.

The Go programming language is gaining popularity - it is on it that the Red Hat OpenShift platform is written. There are serverless calculations, based on functions that run on the container infrastructure. They require new development approaches; in recent years, hybrid and public cloud providers have done a lot to make writing and using cloud functions easier.

Vladimir Karagazhi, Red Hat Solutions Architect Team Leader

Python Technologies


There are several promising areas that can conditionally be combined according to a key requirement for a developer - he must know Python well. These are tasks related to machine learning , neural networks, artificial intelligence and deep learning.

Python is an awesome language with virtually no limits. This is both good and bad at the same time. Good, because literally in a week you can write quite complex things in Python.
But to write a serious system, with many entities and objects, with a large number of connections between them is problematic. This requires strong analytical skills, well-structured thinking and a clear understanding of what you are doing.

Nikolay Sokornov, Director of Reksoft Software Development Department

The perspective of machine learning is also indicated by Dmitry Vdovin, senior programmer of the “Conscience” installment card development group, mentioning innovations. As an example, he cites the development of voice assistants Siri, Alice, Marousi. They work on the basis of machine learning.

Basic knowledge will never go out of style


For any IT specialist, it’s more important not to own a highly specialized tool, but to have a common understanding of the technology sphere. After all, any hype technology is built on a fundamental basis.

You need to study computer science, algorithms, data structures, design patterns, abstractions, programming methodologies. It is on this that hype frameworks and libraries are built. Understand the database - it will become easier to understand everything else.

Anton Tuzhik, Frontend Developer SEMrush

The quality of the code and the speed of its execution by a professional developer can differ dozens of times from amateur ones.

But regardless of what language you write or in what area you program, you will need a technical base: knowledge of data structures, algorithms, patterns, and design principles. Basic knowledge will be required in any case, even if you decide to switch from something standard to fashionable technology.

, . JavaScript — , . TypeScript, JavaScript. , TypeScript, JavaScript. — .

, IT- Bilderlings


Acronis platform architect Anna Melekhova believes that many technologies in IT go through the Hype Cycle curve. First, there is a huge demand for developers, they are trying to use technology in many industries. Then the recession, the number of vacancies is reduced. And stabilization - the business understands where the technology is really useful, and where it is better to use other solutions.

Anna cites blockchain as an example. He was in fashion a couple of years ago, and now he’s approaching the bottom along the hype curve - he did not live up to expectations. At the same time, the architect is sure: the blockchain will eventually take its place, and specialists in it will be in demand, just not as massively as on top of the hype.

There are tasks for which many specialists are now needed in the industry, but there are stable and very cool niche tasks. For example, hypervisors and operating systems. Niche programmers need a little, but they are highly valued, and the need for them is not reduced.

Anna Melekhova, Acronis Platform Architect

What will be in demand in IT in the 2020s


  1. Clouds, containers and microservices: VMware, OpenStack, Kubernetes, Docker Swarm.
  2. Python for building neural networks and machine learning.
  3. Basic knowledge of fundamentals is always needed: computer science, algorithms, data structures, design patterns, abstractions, programming methodologies.
  4. Demand for hype technology may decline over time, but if the technology is useful, it will occupy its niche in the market.


What else to read :

  1. , HR .
  2. .
  3. .

More articles:


All Articles