Microservices
Microservices
In their article called "Microservices", James Lewis and Martin Fowler talk about the new trending and future of the design and architecture of software: The microservices architecture. In a general panorama, instead of having one big monolithic application, one must create several different small applications that coordinate with each other and create the bigger perspective for the entire system.This grants flexibility in the entire structure because changes are less expensive, as instead of rebooting the entire system each time there's a change in the requirements, one can just look into the specific microservice that needs to be changed and do it.
The biggest challenge to tackle in order to begin the implementation of this structure, however, is to manage the synchronous processes, and therefore, create a reliable controller that administrates and instructs the different modules that conform each of the microservices. However, I think it's pretty doable nowadays, and just a matter of time before this new form of architecture prevails on the others.
The biggest challenge to tackle in order to begin the implementation of this structure, however, is to manage the synchronous processes, and therefore, create a reliable controller that administrates and instructs the different modules that conform each of the microservices. However, I think it's pretty doable nowadays, and just a matter of time before this new form of architecture prevails on the others.
Obviously, there are a lot of design patterns and strategies that are developed with the focus on monolithic architecture, and another goal should be to focus on multiple layered design and the proper division of work in order to find out more efficient methods and new patterns apliable to this new architecture.
Finally, another problem I see is that right now, as the great and vast majority of enterprises and companies have implemented monolithic-architecured software and have been running on them for many years, they won't be very inclined in modifying or changing their whole infrastructure (hardware and software) for another architecture that even when it is proved it is more efficient, is not even needed because the previous one fullfilled its job perfectly. I do think it's gonna take more than that to introduce this new architecture in the industry, and it is up to us, the new generation of IT's, to evolve the industry and invite others to adapt into the new models.
Comments
Post a Comment