Cloud Modernization Strategies for Coexistence with Monolithic and Multi-Domain Systems: A Target Rollout Approach

This article discusses a strategy for modernizing complex domains with monolithic databases through four phases: decomposition, prioritization, coexistence, and targeted rollout. The approach offers benefits such as improved scalability, availability, and security of data, the reduced workload on legacy systems, early cutover of dependent systems, and incremental cost savings through on-demand pricing for data storage. Additionally, the strategy provides early availability of analytical data for data-driven decision-making through machine learning and artificial intelligence-led transformation, quick wins, and early data validation.

Modernization refers to the process of updating an organization's systems, processes, and infrastructure to keep up with changing business needs and technological advancements. In today's fast-paced digital landscape, modernization is crucial for organizations to stay competitive and meet the expectations of their customers.

However, modernization can be a complex process, especially when it comes to monolithic databases. Monolithic databases are large, centralized systems that handle all data and transactions for an organization. As a result, they can be difficult to scale, maintain, and update and can hinder an organization's ability to innovate and respond to changing market demands.

Modernization Strategies for Monolithic Databases Supporting Multi-Domains and Capabilities

Modernizing a monolithic database system can be achieved by adopting a cloud-based architecture, using automation tools, and breaking the monolith into smaller pieces. Moving the system to the cloud can enhance scalability, resilience, and security, while automation tools can streamline the modernization process. Breaking the monolith into smaller pieces can facilitate the modernization of individual components without affecting the whole system. Lastly, modernizing data analytics and reporting capabilities can help organizations gain valuable insights and improve their operations.

Here are a few challenges:

Business Continuity for Mission Vital and Critical Workloads

Maintaining business continuity during cloud modernization is a critical challenge as it can lead to service disruptions and extended downtime, which can disrupt operations. This challenge can take different forms, such as application downtime, data loss, performance issues, integration challenges, and security risks. To overcome these challenges, organizations must plan and implement their cloud modernization strategies carefully, ensuring that they have proper backups, testing, and disaster recovery plans in place to ensure business continuity throughout the process.

Challenges and Strategies for Rollout Windows and Prioritization in Multi-Domain Transformations

Multi-domain transformations pose challenges when it comes to rollout windows and prioritization, including limited windows due to dependencies, prioritization conflicts, resource constraints, communication and coordination difficulties, and the need to balance risk and rewards. To address these challenges, a comprehensive plan is essential, with clear communication and coordination strategies, risk assessments, and a prioritization framework. This plan may also require strategies to facilitate early cutovers for decoupling-dependent applications.

Challenges in Moving From a Monolithic Database to a Microservices-based System of Records: Identification of Data Discrepancies and the Importance of Quick Wins

Moving data from a monolithic database to a microservices-based system of records is a complex undertaking, and automating all corner cases for legacy applications built over several decades is impossible, even for subject matter experts. While the modernization team can automate new microservices and web interfaces, covering all function scenarios remains a challenge. Therefore, it is crucial to identify data discrepancies early on. If available, a UI or analytics tool can ensure the target data model accurately represents the data. To avoid risks like data loss, corruption, or misrepresentation, it is essential to implement Quick Wins by validating the data in a phased approach.

Business Sees Modernization as Mere technology Upgrades and Cost-Saving Opportunity

Modernization in IT organizations usually involves upgrading outdated technology with more efficient solutions. However, businesses often prioritize gaining insights from data as the primary benefit of modernization. While moving applications to the cloud and upgrading technology can improve accessibility, mobile usage, and security, they often fail to address core business problems. As a result, businesses may lack the commitment to these modernization efforts due to their focus on cost savings, web enablement, system availability, security, and other factors instead of data insights.

To overcome this challenge, modernization teams need to focus on understanding data needs and how to consolidate them into the new architecture. In addition, they should provide early access to data to gain commitment from stakeholders and ensure a smooth transition to the new system.

Testing Challenges With Legacy Application Modernization

Modernizing applications can be a daunting process that can cause anxiety among IT teams. The lack of knowledge, documentation, and subject matter experts can contribute to this feeling of unease. Even if a company implements automated testing for existing systems, it is still challenging to ensure that all possible scenarios have been tested. One of the challenges of modernizing to cloud-based systems is converting relational schema-based databases to non-relational schema-less databases. However, legacy applications may not have test automation tools to validate all use cases, making it difficult to maintain data consistency across different data stores during modernization. In turn, this can lead to missed business rules that result in significant design changes. To avoid these problems, modernization teams need to gain early access to a mechanism for testing this data on the cloud with real-time feeds to compare against existing applications. By doing so, IT teams can gain confidence that the modernized system will be consistent with the original application and meet business requirements. 

Scalability, Reliability, Performance, and Availability Limitation on Existing Systems

The existing system is most overburdened by the ever-increasing business needs, and data availability and performance are limited by the current architecture. This not only impacts the core domain but also applications and services supporting domains. Although modernization may bring benefits by increasing the availability and scalability of data, it is very time-consuming and can span for years. 

Modernization should not put additional pressure on already constrained resources. 

Target Rollout and Coexistence Challenges for Applications Supporting Multiple Domains

When an organization decides to modernize its existing systems, one of the most significant challenges it faces is the rollout of the new system while ensuring coexistence with the old system. 

This is particularly challenging when the applications support multiple domains, as any disruption in one domain can potentially impact the other domains.

Another challenge in target rollout and coexistence is maintaining data consistency and integrity between the old and new systems. During the transition period, the old and new systems may operate simultaneously, and any changes made in one system need to be reflected in the other system to ensure data consistency and integrity. This requires the implementation of effective data synchronization mechanisms and strict data validation rules.

Furthermore, it is essential to ensure that the coexistence period is as short as possible to minimize any potential disruptions to the organization's operations. This requires a well-planned and executed transition strategy that includes effective testing and validation procedures, as well as training for the end users.

When it comes to modernizing the airline and retail industries, a targeted rollout strategy may involve the following:

Airline Modernization

• Geographically Targeted Rollout: This strategy entails introducing new technologies or applications in specific regions or markets rather than worldwide all at once. Doing so can help test the demand for the product or service and identify any local challenges or opportunities.
• Channel-Targeted Rollout: This strategy involves introducing new technologies or applications through specific channels, such as mobile apps or travel agencies. This can help reach customers where they prefer to conduct their transactions and reduce the need for physical ticketing offices.

Retail Modernization

• Geographically Targeted Rollout: This strategy involves introducing new technologies or applications in specific regions or markets rather than nationwide all at once. This can help test the demand for the product or service and identify any local challenges or opportunities.
• Channel-Targeted Rollout: This strategy involves introducing new technologies or applications through specific channels, such as online or mobile apps. Doing so can help reach customers where they prefer to conduct their transactions and reduce the need for expensive physical stores.

In summary, target rollout and coexistence challenges are significant in the context of modernizing applications that support multiple domains. Effective planning, testing, validation, and training are essential to ensure a smooth transition and minimize any potential disruptions to the organization's operations.

Database Migration to Cloud Before Modernization

Migrating a monolithic database system to the cloud for modernization purposes can provide benefits such as enhanced scalability, better handling of usage changes, and improved disaster recovery capabilities. Nevertheless, businesses may have reservations about transferring their applications to the cloud due to factors such as legacy systems, complex business logic, inadequate functional comprehension, outdated applications and code, data, tight coupling, and the need for extensive testing within a limited timeframe.

Solution 

The strategy involves four main phases: decomposition, prioritization, coexistence, and targeted rollout. 

The first phase, decomposition, involves breaking down the monolithic database into smaller components that can be moved to the cloud in a more streamlined manner. The solution proposes to leverage the traditional Structured Design Matrix (SDM) approach to map various systems and calculate their dependency count and prioritization based on impact.

Prioritization is the second phase, which involves identifying the components that need to be modernized first based on their criticality and complexity. Prioritization means shifting the systems with higher impact score system to the top-left corner (earlier in the roadmap) while the most dependable systems will automatically move to the bottom right corner (later in the roadmap) in the chart. 

Figure1: Structured Design Matrix 

The third phase is coexistence, which involves the migration of the identified components to the cloud while ensuring that they can coexist with the legacy systems that have not yet been modernized. The solution further proposes to establish a one-way continuous data migration/sync from on-premises to the cloud to support the phased rollout of various applications to leverage incremental benefits while coexisting with minimized business impact for existing systems. By establishing one-way sync, you can move dependent applications to read data from the cloud, and hence you will be able to further improve your matrix and move other applications earlier in the roadmap. 

Figure 2: Coexistence Strategy Flow Chart

The final phase, targeted rollout, involves the deployment of the modernized components in a targeted manner based on the identified priorities.

The rollout can be divided into the below phases based on the process describes in Figure 2.

Phase 1: Establishing the one-way continuous data migrations/sync from on-premises to the cloud to roll out the least dependent applications requiring simple cutover (no targeted rollout). Redirect the downstream simple, complex applications simple cutover (no targeted rollout) by leveraging the wrapper services to read from the cloud. 

Phase 2: Rollout of simple application requiring targeted rollout by leveraging the wrapper services to update on-premises and read from the cloud and incrementally moving the write wrapper to point to the cloud database.

Phase 3: Rollout of complex application requiring targeted rollout by leveraging the wrapper services to update on-premises and read from the cloud and incrementally moving the write wrapper to point to the cloud database.

Phase 4: Retirement of the legacy system.

Benefits of the Approach Include:

• Early cutover of dependent systems that require only read data
• Early cutover of straightforward dependent systems that require target rollout
• Prioritization based on rollout windows for targeted rollout
• Availability of data for insights, AI transformation, and analytics
• High availability, scalability, performance, and security of data and read API 
• Quick wins and early data validation
• Gradual reduction of traffic to on-premise database
• Incremental cost benefits for data storage through on-demand pricing

Conclusion

By utilizing this approach, the solution gradually decreases the need for a complicated application in the future roadmap to writing to an on-premises database since all dependent systems have already been shifted to the cloud. 

This approach offers numerous benefits, including early availability of analytical data for data-driven decision-making through machine learning and artificial intelligence-led transformation, early enablement of downstream applications, gradual reduction of workload on the on-premises system, and gradual improvement in availability, security, cost, and performance.