Antifragility Architecture: IT Systems for Wartime Conditions

Russia’s full-scale war against Ukraine has become not only a military but also a technological turning point. What was considered unlikely scenarios just a few years ago is now shaping a new norm for designing digital systems. This involves a shift from classic reliability to antifragility and Digital Resilience.

In a reality where cyberattacks combine with the physical destruction of infrastructure, IT system architecture ceases to be a purely technical scheme. It becomes a strategic tool for preserving business and the continuity of state processes.

Default-Unsafe: when a stable environment does not exist

Classic approaches to reliability assumed that after a failure, the system should return to its initial state. Wartime experience has shown otherwise: a “normal” state may not exist at all. Network delays, infrastructure fragmentation, unstable power supply — these are now the basic operating environment.

The incident with Viasat’s KA-SAT satellite network in February 2022 was illustrative. A cyberattack on the ground control segment led to the failure of tens of thousands of terminals in Europe. This case clearly demonstrated cyber-kinetic convergence: a digital attack caused real physical consequences for energy and transport.

Systems that had alternative communication channels, backup protocols, or mesh architectures maintained operability. The rest lost it irrevocably.

Geodiversification and the end of the on-premise illusion

Today, a single point of failure (SPOF) is not just a server. It can be an entire jurisdiction, a provider, or a cloud region. True resilience means the ability to quickly switch between different geographies and providers without losing data or active sessions.

The model of “one data center + backup copy” no longer meets real threats. The future lies in multi-cloud architectures, where critical services are distributed among various global and local providers.

Resilience in logic, not in “hardware”

Redundant server duplication alone does not save. The key factor becomes architectural logic and the system’s ability to operate in limited functionality mode.

Effective approaches include:

• clear separation of the transactional core (payments, identification, accounting);
• automatic disabling of secondary services (analytics, recommendations) in case of overload;
• asynchronous data processing and advanced offline modes.

The failure of auxiliary components should not block the main process. It is logical decomposition that determines the survival of the system in a crisis.

Low-code as a tool for digital mobility

In wartime conditions, adaptation speed is often more important than formal code perfection. Enterprise-class low-code platforms allow for:

• promptly changing business processes to meet new legislative or logistical requirements;
• adapting logic without completely rewriting the system’s core;
• maintaining strict access control even during rapid changes.

UnityBase: a platform for antifragile architecture

Implementing these principles requires a foundation. Such a tool is the highly secure UnityBase platform, designed for high-load Enterprise-level systems.

UnityBase allows building scalable systems on JavaScript, focusing on the domain model. An important advantage of the platform is its infrastructure undemandingness. It works stably even on limited resources, which is critical when individual network segments or server capacities degrade.

From a security perspective, the platform meets the highest state requirements, allowing its use in critical infrastructure and electronic document management systems where data compromise is unacceptable. It combines the speed of low-code development with the power of a professional engineering platform.

The problem of Vendor lock-in: technical sovereignty

Strict reliance on a closed proprietary “engine” is a risk. In unstable conditions, businesses must have access to their data and logic. In UnityBase, logic is described by transparent models, and data is stored in standard DBMS (PostgreSQL, Oracle, etc.) without hidden formats.

This ensures technical sovereignty: you own your system, you can migrate it or deploy it in any configurations without the risk of being trapped by a single vendor.

Human factor and organizational resilience

Antifragility is not just code. It’s SRE and DevOps teams ready to work in asynchronous mode, and clear incident response protocols. Organizational resilience ensures that processes are documented and do not depend on one person, and knowledge is distributed within the team.

IQusion IT’s practical approach

The IQusion IT team develops solutions based on the presumption: “failure is inevitable.” We help businesses:

• find and eliminate architectural single points of failure (SPOF);
• rebuild processes to operate in fragmented environments;
• adapt global platforms (SAP, Oracle, etc.) to Ukrainian realities;
• use UnityBase for fast and secure digital transformation.

Resilience is not an option, but a fundamental property of architecture. In a world of modern threats, it is not the largest who survive, but those designed with inevitable disruptions in mind. This is precisely the approach we use in our projects.