About how such a model came to the fore and what are its features, says Alexander Triguba, Operations Director of Mobius Technologies.
The pandemic as a catalyst
Any infrastructure, be it cloud or on-premises, is essential for businesses to be able to solve everyday tasks. At the same time, each company seeks to find its own effective balance between cost and the resulting set of IT services. Cloud solutions can be rented from a provider or deployed at your own facilities, striking a balance between reliability and availability. It’s no secret that local infrastructure requires its own engineers and specialists who will accompany it, constantly investing in hardware solutions, software and communication channels. If the choice is made in favor of an external service, then the entire burden of maintenance and responsibility is transferred to the IT service provider. Each solution you choose has its pros and cons – continuous availability, stability, performance, and security.
Since all organizations are different, with different scale of activity, business objectives and financial capabilities, there is no one-size-fits-all model that can satisfy all business needs. For some, it is critical to place all services inside and consciously bear all the costs, while others choose the pay-as-you-go strategy, when you can rent “here and now”, reducing costs and consciously accepting the risks of the chosen solution.
Finding the optimal balance allows a hybrid model that provides for the simultaneous use of local and cloud computing resources. Implementing a hybrid IT infrastructure provides organizations with many benefits: increases business resilience and agility, improves customer service processes, and accelerates testing and development of new products.
The hybrid model received a serious impetus to development during the pandemic. This is not to say that it has not been resorted to before, but the new restrictions forced even those who are accustomed to the traditional approach and were just starting to get acquainted with cloud solutions to look in a new way.
Many adopted a replacement strategy and did not seek to change established architectural IT solutions, realizing that managing a hybrid IT infrastructure is not an easy task. In such situations, only force majeure circumstances could push for a change of decision. In this case, the pandemic turned out to be such a catalyst that gave a new impetus to the development of digitalization of services and services.
Due to coronavirus restrictions, everyone was forced to go online, business willy-nilly began to look for new models of work in new conditions and new ways of interacting with partners and employees. Those who did not have cloud services created their own from scratch or started using public ones on the go. By scaling their resources into the cloud, the business began to build hybrid structures, creating new digital products for the business.
Nobody was ready for such a dynamic growth of cloud services, but over time, it became clear that the hybrid model is convenient and economically justified even without taking into account coronavirus restrictions. When restrictions began to be lifted, some external services reached a new level of digitalization, EDI and EDI became a striking example. The business refused some of the cloud solutions and transferred the accumulated experience to the internal infrastructure, and integrated some of the cloud services into its corporate solutions.
Thanks to the pandemic, many have unwittingly become testers and architects of new services, customer experience has been accumulating, digital solutions have appeared at a cosmic speed, and competition between digital providers has begun. These factors made it possible to painlessly offset the departure of foreign cloud services from the Russian market. When some of the services became unavailable or moved into the risk zone, the business quickly migrated to previously tested or created local solutions.
Now almost any business uses a hybrid infrastructure consisting of three large segments: a private (local) circuit, external cloud solutions and government digital services.
Work without restrictions
In any infrastructure, the question of limitations always arises, and they need to be asked at the stage of its design and scaling. When choosing a solution, it is important to take into account the factors of business growth and justified investments in IT, trying to get the maximum utilization in the current moment and be easily transformable according to business needs, ensuring digital resilience in relation to external storms and challenges.
For many, the pandemic has been a test of digital resilience. If earlier many cloud solutions worked smoothly, then with the advent of large corporations, they simply fell, unable to withstand the load. As a result, their suppliers were forced to expand capacity, look for new architectural solutions with dynamic control, allowing them to survive different peak loads.
When designing the architecture, it is better to initially lay down the possibility of scaling through cloud services that allow you to dynamically manage capacity, reduce costs, and transfer part of the services or resources outside. You also need to simulate peak loads and test action plans for different situations.
When planning or transforming an IT infrastructure, it is better to involve technology consulting, which will reduce the time to develop a comprehensive architectural solution and get an adapted solution for the needs of a particular business, taking into account previous experience in similar projects, as well as a full project estimate and specification. In addition, even at the pilot project stage, you can test the designed solution, conduct load tests at the partner’s facilities and use ready-made technology consulting solution libraries, rather than invest in an experiment.
Common Mistakes When Building a Hybrid IT Infrastructure
One of the typical mistakes in building a hybrid infrastructure is the poor integration of local and cloud resources. For example, a drop in performance during working hours when replicating between environments and business applications, insufficient bandwidth of communication channels, low performance during peak or seasonal loads.
The second mistake is the publication of critical services in the cloud part and the lack of alternative data exchange between services. Often a business solution for maintaining master data is published in the cloud, and if the cloud service is unavailable, the main business applications and processes become unavailable.
The third mistake is ensuring the availability and security of an external service from external factors. An ordinary DDos attack can completely paralyze a business, while a banal payment delay story can stop a service.
The fourth mistake is that infrastructure deployments often forget about legacy applications and technologies that the company already has. They can be difficult to migrate to a hybrid model.
Another weak point is the lack of qualified employees to manage hybrid IT, as well as the lack of working out secure access to critical information with separation of access rights, the presence of restrictions on connections from a certain pool of IP addresses, etc., but this is a separate big topic. .
In conclusion, we can conclude that hybrid IT infrastructure has proven itself well in terms of peak load distribution, rapid deployment of test environments, and cost savings when there is a shortage of own capacities. Practice has shown that in order to correctly build a hybrid infrastructure, it is necessary to conduct a pre-project survey, identify the main business risks, coordinate SLA services with the business, assess all risks both when hosting local services and when hosting them in cloud resources.
