Cloud
Cloud Migration in the Era of Gen AI
This whitepaper explores how GenAI is reshaping cloud migration and application modernization, addressing long-standing challenges, and unlocking new efficiencies. Additionally, we will discuss industry-leading AI frameworks, real-world applications, and a structured approach for enterprises to harness the full potential of GenAI in their cloud transformation journey.
Insights
- This white paper primarily presents the role of Gen AI in addressing the challenges faced during cloud migration and modernization.
- The paper lays out detailed Gen AI capabilities and mentions some of the industry leading AI frameworks that have impacted migration lifecycle.
- It mentions overall benefits to the organization in adopting Gen AI and AI based frameworks and tools.
- Provides an approach to integrate Gen AI into projects.
1. Introduction
Digital transformation has become a strategic investment for organizations worldwide over the last decade. The business case for cloud migration and application modernization depends on achieving cost savings, improved scalability, increased agility, and enhanced performance, thereby enabling businesses to quickly adapt to changing market demands and innovate at faster speed.
Key aspects of business case:
- Cost Reduction: Cloud Migration can lead to significant cost savings by migrating the workloads from traditional data centers to cloud service providers, reducing expenses on software licensing, hardware and staff to manage infrastructure. The pay-per-use model by public cloud allows organizations to save on upfront investments and expansion costs. Private cloud also provides organizations with predictable and stable cost models by allocating resources based on their specific needs. The model provides greater control over their infrastructure and ability to optimize resource utilization.
- Improved Agility and Scalability: Cloud provides greater agility, this allows organizations to quickly scale resources up or down based on demand, deploy new applications at a faster rate. It offers the scalability to manage the peak loads and fluctuating demands without performance degradation. Organizations may even pursue migration from one cloud provider to another to gain access to specialized services and tools. This migration approach has matured over the years and today there are tools and frameworks available for successful migration of applications and associated data from one cloud provider to another. This has provided flexibility for organizations to move across cloud providers based on their requirements and reduce dependency on a particular cloud provider.
- Enhanced Security: Cloud providers ensure systems are up-to-date and automatically apply security patches, therefore protecting the systems and applications from cyber-attacks. They invest in security infrastructure and compliance certifications thereby staying compliant with regional and global regulations.
- Faster Innovation: Modernizing applications with cloud-native technologies and APIs can enable businesses to innovate faster and develop new products and services more easily. Organizations can leverage modern hardware and innovative services regularly released by cloud providers, enhancing operations at minimal extra cost. Cloud interoperability with SaaS based software offers improved efficiency and increased flexibility. Interoperability offers streamlined workflows, faster decision making and improved collaboration.
- Improved Data backup and Recovery: Cloud providers facilitate easier and faster disaster recovery by leveraging cloud-based backup and replication services, thereby ensuring business continuity and minimizing downtime.
- Faster Time-to-Market: Modernizing applications by leveraging agile methodologies and cloud-based tools can accelerate the development and deployment of applications at improved velocity and ensure services are launched at faster speed. DevSecOps offers significant value in digital transformation engagements by integrating security into every stage of development and thereby enhancing application development speed and reliability. DevSecOps tools ensure continuous security checks and monitoring during the development process and thereby enhance the security readiness of applications and infrastructure.
- Business Alignment: Cloud providers manage the IT infrastructure, thereby freeing up organization IT staff for strategic initiatives and innovation. Modernizing applications by leveraging cloud-native technologies and microservices architecture can enable businesses to align their enterprise applications with business needs and goals. Cloud-native marketplaces provide organizations access to a wide range of solutions that can help businesses reduce time to market for launching innovative products. The marketplace acts as a central hub for cloud-native applications and provides solutions across cloud models SaaS (Software as a Service), PaaS (Platform as a Service) and IaaS (Infrastructure as a Service).
Today modernization and migration engagements have become key tenets to drive business growth and streamline operations. A sizeable budget is allocated by global organizations to ensure seamless execution of transformation projects. Maximizing ROI is one of the crucial factors for these transformation engagements.
Key Strategies for Maximizing ROI:
- Planning and Assessment: It is important to define key business objectives and expected outcomes. Conduct a comprehensive assessment of the current infrastructure and application landscape to determine the best modernization and migration strategy.
- Prioritization of Impactful Use Cases: Focus on modernization applications that provide direct benefits in terms of revenue growth, efficiency and improved customer experience.
- Cloud-Native Services: Leverage cloud-native frameworks and tools, refactoring from monolith to microservices architectures thereby improving the performance of legacy applications and faster innovation. Containerization of applications provides several benefits in the form of increased portability, efficiency, faster development cycles, consistent deployment and enhanced developer productivity.
- Manage Cloud Spend: Implement cloud management framework like FinOps to derive maximum value from cloud investments. FinOps is an operational framework that involves collaboration between finance, technology and business teams to optimize cloud spending and maximize business value. It is about creating a culture of financial awareness and accountability within the organization.
- Measure and Track: Monitor the modernization projects and keep track, identify the gaps, areas of improvement. Address the gaps and optimize the overall execution from technical and process perspective to ensure maximum ROI.
While the organizations have been at the forefront and ensuring the above strategies are part of the modernization projects, there are challenges at every stage of execution.
Some of the key challenges of cloud migration and modernization are:
- Understanding legacy systems due to lack of documentation and unavailability of core developers
- Data migration and transformation
- Managing data security
- Cost Management
- Integration with the latest DevSecOps tools and frameworks
The details of these challenges will be discussed in the section below. To address some of these challenges a technological advancement in the form of “AI” is playing an impactful role in transformation engagements.
With the advent of Generative AI (GenAI), organizations now have an unprecedented opportunity to accelerate and streamline cloud migrations by automating key processes, reducing manual effort, and enhancing decision-making.
2. Impact of Gen AI
For organizations on the path of cloud migration and application modernization, integration of AI and Gen AI based frameworks has become imperative to ensure business remain agile and future ready.
Gen AI is now a key part of cloud migration. More than automation, it provides intelligence and vision into complex processes.
- AI-powered tools help organizations in making intelligent decisions for migrating workloads
- Automation of routine tasks like application assessment and planning
- Enhancing developer productivity by automating templates for infrastructure provisioning
- Real time monitoring of infrastructure during the migration lifecycle
- AI based tools identifies the performance issues and forecasts future cloud resource needs
AI is playing a transformative role in application modernization. AI is leveraged to create smart and adaptable applications in ever changing business requirements.
- Updating legacy systems and code refactoring by Gen AI based tools
- Addressing security vulnerabilities present in legacy code base
- Upgrade the technology stack
The below section explains the lifecycle of cloud migration and application modernization engagements and role of AI and Gen AI in solving some of the challenges and overall impact created in execution of these engagements.
3. Cloud Migration Lifecycle
The Cloud Migration strategy presents a significant step for any organization’s IT department since this involves relocation of existing workloads to the cloud. It provides an opportunity to revamp the legacy software to build a new cloud native workload there by leveraging all the benefits provided by cloud.
It is important for the organization to have a detailed blueprint to ensure smooth transition.
Figure 1: Cloud Migration Steps
Breakdown of Steps:
1. Planning
Define Business Goals: Clearly define the reasons for considering cloud migration and expected outcome in terms of cost, scalability, new products etc.
Approval from sponsors: Senior leadership support and approval are key to success given the budget allocation required for large scale migration engagements.
2. Assessment
Perform Cloud Readiness Assessment: Evaluation of current infrastructure, applications, and data to determine the feasibility of migration and potential challenges.
Select Migration Strategy: Based on the application assessment migration disposition is performed based on “6 Rs” framework.
Figure 2: Cloud Migration Framework
3. Execution
Implement Patterns: Design and implement cloud-based patterns, frameworks replacing on-prem features with cloud-native features, standardization of Cloud Infrastructure.
Migration of workloads, data: Migrate applications to cloud using the migration framework. Migrate the data ensuring data integrity. Implement measures for vulnerability management of applications deployed in cloud. Ensure security measures are in place to protect data and application in cloud.
Testing: Test the migrated applications and cloud infrastructure to ensure behavior and performance is as per the acceptance criteria.
Document: Create detailed documentation of migration stages detailing the rationale for strategy, design patterns, application workflow in cloud, network diagrams.
4. Optimization
Optimize the performance of applications: Monitor and optimize the application performance in cloud. Implement cost-saving measures in the cloud environment by leveraging FinOps tools.
Automation opportunities: Identify areas to automate the tasks to improve efficiency. Leveraging Terraform to building Infrastructure as Code (IaC).
5. Monitoring
Monitor: Implement tools for continuously monitoring the application and infrastructure. This is important to manage the overall health of the migration program.
Proactive Improvements: By utilizing the monitoring tools look for areas of overall improvement in the application ecosystem.
4. Application Modernization Lifecycle
Application modernization presents organizations with an opportunity to overhaul the entire legacy ecosystem encompassing application architecture and underlying platform infrastructure. The bulk of work is done in migrating the monolithic applications to microservices based architecture. The tangible benefits are in the form of improved velocity for new feature delivery, replatforming the application from on-premises to cloud by using microservices and DevOps based release cycle. The decision to migrate to cloud post application modernization solely depends on the business value.
The key to modernization success is based on the following steps:
Figure 3: Application Modernization Steps
Breakdown of Steps:
1. Planning
Define Modernization Strategy: Clearly define the reasons for considering cloud migration and expected outcome in terms of cost, scalability, new products etc.
Assess As-Is State: Senior leadership support and approval is key to success given the budget allocation required for large scale migration engagements.
Create a roadmap: Create a detailed roadmap involving timeline, engineers, tools and frameworks required for modernization projects.
2. Implementation
Build Digital Skillset: Shift in mindset is critical for the success of engagement. The team should be open to reskill. Engineers need to be provided with opportunities to upskill in the digital space like DevOps, Cloud etc.
Microservices Architecture: Migrate and build new systems by utilizing microservices based architecture to create a scalable, robust, reliable application ecosystem. Focus on creating future ready applications.
Containerization: Implement containerization methodology which is now the industry standard in modern application landscape. Containerization significantly improves the overall software development and deployment lifecycle.
3. Operations
Continuous Integration and Delivery (CI/CD): Implement DevSecOps based principles for automated build, test and deployment of applications. The entire process is streamlined with the help of automation.
Monitoring: Monitor the overall performance of the modernized application and maintain overall health.
Governance: Establish policies and procedures for managing modernized applications.
The lifecycle steps discussed above for both Cloud Migration and Application Modernization overlap and provide a standard template which is implemented by organizations embarking on their Digital Transformation journey.
5. Strategic Role of AI
AI is playing a pivotal role in cloud migration and application modernization space by automating complex tasks, enhancing decision-making, and optimizing costs leading to faster and efficient transformations. AI in conjunction with cloud computing is significantly lowering the barriers involved in legacy migrations to cloud.
Look at AI benefits:
Figure 4: AI Benefits
- Automation & Efficiency: AI helps in automating tasks performed during migrations like application assessments there by reducing efforts. Predicting and preventing potential issues and thereby reducing downtime and improving overall efficiency. Tools to generate code snippets and test cases in further improving overall productivity of the developers.
- Optimization: AI can dynamically allocate infrastructure resources based on demand and application needs, thereby ensuring optimization and cost effectiveness during cloud migrations.
- Improved Decision Making: AI can analyze enormous amounts of data and provide insights which can help organizations while driving migration and modernization strategies. AI presents predictive analysis that can help in mitigating risks during execution.
- Enhanced Security: Algorithms can identify security threats and act accordingly thereby protecting the overall organization application landscape.
6. Key Challenges in Migration and Modernization
As organizations embark on their migration and modernization engagements, there are several challenges encountered during the journey.
The challenges can be categorized under the following high-level areas:
1. Technical Knowledge:
Legacy system comprehension: Understanding the legacy systems has always been a challenge given the nature of code base, architectures, outdated technologies, lack of documentation and unavailability of original developers of application. This makes it difficult to modernize and upgrade the system.
2. Process:
Lack of planning: Organizations need to have solid strategy for migration or modernization engagements. Lack of planning can lead to technical hurdles and budget overruns.
Migration and Transformation: Migration of data from on-prem or legacy to cloud can be a highly complex and time-consuming activity.
3. Implementation:
Managing security vulnerabilities: Legacy code may have known security vulnerabilities due to older versions of technology used in application. They can pose serious threats while moving to cloud.
Integration: As part of application modernization implementation and integration of DevOps and latest technologies can be difficult especially due to lack of skilled engineers.
4. Business:
Budget and Effort: The cost encountered for migration and modernization can be expensive and time consuming, this has led organizations to prioritize during critical times and thereby put the project on hold due to lack of funds.
7. Generative AI Impact on Migration and Modernization
Generative AI plays a transformative role in migration and modernization engagements by addressing some of the current challenges in execution. There is a paradigm shift with integration of Gen AI which has improved the overall experience from discovery to execution and monitoring.
| Current Challenges | Role of Gen AI |
|---|---|
| Legacy system comprehension |
|
| Migration and Transformation |
|
| Managing security vulnerabilities |
|
| Budget and Effort |
|
| Integration |
|
Key Benefits:
- Productivity Improvements: Gen AI can accelerate software engineering activities by automating tasks in the form of code generation, extracting insights from documents and generating unit test cases. Research by McKinsey & Company has indicated that Gen AI can accelerate modernization by 40% while improving the quality of output.
- Faster time-to-market: GenAI tools in the form of “co-pilots” have matured over time and there has been a sharp rise in overall utilization of these tools by organization. Automation features of the tools have streamlined the overall development process and improved the velocity of modernization, moving applications to production at a faster rate.
- Right Size Team: Gen AI automating core repetitive tasks has allowed developers to focus on complex tasks which need human oversight and other strategic initiatives.
Generative AI is disrupting the SDLC. This statement holds true because of the numerous benefits in the form of automations provided by Gen AI and AI models; however, the role of human-in-the-loop is irreplaceable. The human factor is required to ensure accuracy, reliability and adhere to ethical and regulatory standards.
Some aspects of human-in-the loop:
- Provide real-time feedback and enable refinement of Gen AI tools.
- Identify and address bias in AI systems.
- Ensuring transparency and explainability of Gen AI systems.
- Validate the output and ensure standards are met.
- Set ethical standards and ensure AI systems comply too same.
- Identify and mitigate risks associated with Gen AI such as drift in responses, incorrect information etc.
The next section covers in detail how Gen AI and AI Models are leveraged in each phase of migration and modernization life cycle.
8. Generative AI capabilities across migration lifecycle
The section will cover generative AI and AI capabilities leveraged across the migration lifecycle.
Cloud migration can be primarily divided into 4 phases:
Figure 5: Cloud Migration Phases
Phase 1 – Discovery and Assessment
| Activities | Role of AI/Gen AI | AI powered Tools | Benefits |
|---|---|---|---|
| Discover on-premises application architecture components and dependencies | Generative AI provides an approach to collect infrastructure, application data and map out dependencies and relationships Generate visualizations and documentation |
|
|
| Generate legacy application documentation | Gen AI by leveraging LLM (Large Language Models) can be used to understand the legacy code base and generate documentation |
|
|
| Discover legacy application business logic | LLMs* can delve into the legacy code and discover patterns and help in understanding complex business logic and workflows | Custom built tool can be developed to extract the pattern with help of LLM and Gen AI |
|
| Transformation Planning | Create a wave plan for migrating and modernizing workloads to cloud |
|
|
*LLMs in combination with RAG based approach provides required intelligence to scan through the code and enable developers to understand the workflow.
*Infosys Mainframe Scan is an Application Discovery, Reverse Engineering and Data Lineage tool with cool IDE feature to Modernize Mainframe. It accelerates the application transformation of your hybrid cloud environment. Improves business alignment in deriving requirement and reduces Application development and Maintenance cost. Discovering hidden issues within the Application increases the stability of the system.
Phase 2 – Cloud Readiness
| Activities | Role of AI/Gen AI | AI powered Tools | Benefits |
|---|---|---|---|
| Convert Monoliths to microservices |
|
|
|
| Generate target database/schema model |
|
Database Migration Service
|
|
| Generate OpenAPI documentation |
|
Gen AI tool kit using LLM |
|
| Generate HLD/LLD |
|
|
|
| Cloud Infrastructure readiness |
|
|
|
*Infosys Database Migration solution manages life cycle activities of database migration with end-to-end automation capability. It has brought together indigenously developed deep learning models, proven Infosys products and open-source utilities into one platform. It helps in process simplification with a rich UI, and it facilitates accelerating over all database and data migration augmented with various AI / ML models, connectors, accelerators and utilities.
Phase 3 – Migration
Code Migration
| Activities | Role of AI/Gen AI | AI powered Tools | Benefits |
|---|---|---|---|
| Migrate monolith to microservice based templates |
|
|
|
| Implement new features/use cases |
|
|
|
| Migrate source database to target |
|
Database Migration Service
|
|
| Conduct Code Reviews |
|
|
|
*Infosys Code Migrator is an Infosys IP tool which helps in accelerating the Organizations code conversion from one technology to another. This platform supports both rules based & Gen AI based code migrations.
Validation
| Activities | Role of AI/Gen AI | AI powered Tools | Benefits |
|---|---|---|---|
| Generate unit test cases |
|
|
|
| Generate API test cases |
|
|
|
| Generate functional test cases |
|
|
|
| Test Execution and Management |
|
|
|
Deployment
| Activities | Role of AI/Gen AI | AI powered Tools | Benefits |
|---|---|---|---|
| Generate Docker files for containerization |
|
|
|
| Generate CI/CD scripts Generate CI/CD pipelines |
|
|
|
| Security |
|
|
|
| Documentation |
|
|
|
| Deployment playbooks |
|
|
|
Phase 4 – Post Migration
| Activities | Role of AI/Gen AI | AI powered Tools | Benefits |
|---|---|---|---|
| Monitoring |
|
|
|
| Logging |
|
|
|
| Proactive Incident Resolution |
|
|
|
| Optimization |
|
|
|
The above-mentioned AI tools are a small subset from ever increasing list of frameworks and tools that are being developed by various organizations. Organizations have also started building in-house solutions by leveraging LLMs in combination with the Retrieval Augmented Generation (RAG) framework.
8.1. AI-driven Automation in DevSecOps for Cloud Migration
The integration of AI into DevSecOps has streamlined cloud migration by safeguarding data, applications and infrastructure in cloud environments. AI in DevSecOps has brought a significant shift towards automating security tasks, proactive vulnerability detection and compliance checks to maintain robust security in complex environments.
Some of AI-driven solutions in DevSecOps
- Vulnerability Management: Real time analysis of code and infrastructure configurations by AI algorithms can be used to identify vulnerabilities. Models can identify patterns that indicate risk and provide remediation. Automated scanning can happen at each stage of the development life cycle and proactively detect vulnerabilities there by reducing risk in the production environment.
- Predictive Threat Detection: AI models have the capability to forecast potential security risks by analyzing historical incidents, threat patterns and security trends. Automate incident response workflows and respond to security threats in real time.
- Anomaly Detection: AI models by monitoring user and system patterns can detect anomalies or potential issues. This will enable teams to detect threats like unauthorized access leading to malicious attacks.
- Intelligent Incident Response: Machine learning is used by AI-powered tools to analyze incidents and perform classification, further isolate them based on severity and provide suggestions to resolve the incidents Automation can dynamically patch vulnerabilities, rollback updates etc.
- Automated Compliance and Policy Adherence: AI tools can automate compliance checks against industry standards (GDPR, HIPAA, PCI DSS) to meet regulatory requirements. Automate enforcement of policies in the CI\CD pipelines. Detection of non-compliance issues and provide remediation for same.
Some of the industry’s leading DevSecOps tools powered by AI:
- Checkmarx
- Snyk
- DeepSource
- Veracode
8.2. Role of Gen AI in FinOps
Gen AI is playing a crucial role in FinOps and thereby creating a paradigm shift in cloud cost management. As organizations over the years have been upgrading cloud infrastructure and building new applications in cloud, financial management is critical for maximizing ROI and business value of cloud. Gen AI is making FinOps more valuable by automating tasks, analyzing and summarizing data from multiple sources and enabling proactive and cost-effective cloud resource management.
Cloud providers have started integrating Gen AI with FinOps tooling for the engineers to ask questions in natural language to query the data. It enables streamlining processes, identify spending patterns and provide insights into cloud spending.
Some of the Gen AI use cases in FinOps are:
- Data Analysis and Automation: Gen AI can automate data processing and analyze huge amounts of data related to cloud usage, spending trends to identify patterns and provide avenues for cost optimization. Provide visibility to overall cloud resources and associated spends.
- Improved Decision-making: Gen AI can aid in improved decision making by providing insights and recommendations based on the analyzed data. Provide more accurate forecasts thereby enabling organizing to manage the budgets.
- Cost Optimization: Gen AI based tools can identify underutilized cloud resources, services and provide recommendations to save cloud cost. Tools can detect anomalies in the spending and proactively alert the teams and manage the financial risk.
Gen AI has significant potential to transform FinOps into a proactive cost management model and help organizations drive maximum value for investments.
9. Industry leading Gen AI and AI frameworks
This section will cover some industry leading AI based frameworks that have enabled organizations to streamline cloud migration and modernization through automation.
9.1. Discovery and Assessment
9.1.1. AWS Application Discovery Service
| Features | Description |
|---|---|
| Discover on-premise infrastructure | AWS Service collects server and database configuration information. |
| Identify server dependencies | Application Discovery Service uses agents to record network activity of server. The data collected from inbound and outbound activity can be used to identify server dependencies. |
| Measure server performance | Measuring memory usage and network performance of the applications and associated processes. This information is used as a baseline while migrating to AWS. |
| Migration Plan | Generate a migration plan for Application Migration Service including rightsizing of instances and network configuration. |
9.1.2. Azure Migrate
| Features | Description |
|---|---|
| Map on-premise environment | Collect all the on-premises information related to application, databases, and hosted infrastructure. Analyze the data and generate cost estimate for migration to Azure. |
| Create a plan | Generate a plan with cost effective resources required in Azure by assessing on-premises infrastructure. |
9.1.3. Matilda Cloud
| Features | Description |
|---|---|
| Automated Discovery | Detects all supported operating systems, services, databases and software automatically |
| Topology and Dependency Mapping | Identifies connections, topology, and dependency for applications |
| Architecture Assessment | Current and proposed architecture analysis based on 5Rs – Rehost, Replatform, Refactor, Retire and Retain |
| Application Insights and resource utilization | Application utilization, infrastructure information. Memory and network utilization across environment |
9.2. Migration – Build
9.2.1. GitHub Copilot
| Features | Description |
|---|---|
| Code Completion | Tool provides templates, suggestions for code completion etc. |
| Code Refactoring/Fix | Provide suggestions for implementing code refactoring, potential improvements. Propose a fix for bugs in code. |
| Copilot Chat | Provide guidance and support for coding tasks. |
| Code Review | Provides suggestions as part of code review to support developer write high quality code. |
| Pull request summaries | AI-generated summaries of the changes that were made in a pull request, which files they impact, and what a reviewer should focus on when they conduct their review |
| Knowledge Bases | Tool has a feature which enables developers to create a repository of documentation which can be used as a context while working with Copilot. |
| Code documentation | Generate code documentation for multiple languages |
| Automated Test Generation | Assist in developing unit and integration tests |
9.2.2. Amazon Q Developer
| Features | Description |
|---|---|
| Code Generation | Generates real-time code suggestions ranging from snippets to full functions |
| Transform workloads | Legacy code bases to modern versions of languages.
|
| Code Refactoring/Fix | Amazon Q can identify and fix bugs. Refactor for improved maintainability. |
| Inline Chat | Provide conversational support, allowing users to ask questions and suggestions.. |
| Code Review | Amazon Q Developer automates code reviews, allowing you to detect and resolve code quality issues such as logical errors, anti-patterns, code duplication, and security vulnerabilities in your applications. |
| Deployment-ready IaC | Amazon Q Developer can help generate deployment-ready infrastructure as code (IaC) for AWS CloudFormation, AWS Cloud Development Kit (AWS CDK), or Terraform |
| Automated Documentation | Generate in-depth documentation within source code including data-flow diagrams. |
| Automated Test Generation | Q Developer can iteratively generate unit tests |
9.2.3. Gemini Code Assist
| Features | Description |
|---|---|
| Code Assistance | Tool based on demand generates code blocks. Auto completes the code. |
| Code transformation | Provides features like fixing code issues and helping developers by giving code explanations. |
| Natural Language Chat | Get answers on coding questions, receive guidance on coding practices. |
| Local codebase | Gemini Code Assist generates code that’s more relevant to the application by grounding responses with context from local codebase and current development session. Perform large-scale changes to the codebase, including adding new features, updating cross-file dependencies, helping with version upgrades, comprehensive code reviews, and more. |
| Automated Test Generation | Generate unit test cases for code. |
9.2.4. Azure Database Migration Service
- Azure Database Migration Service (DMS) is a fully managed service that simplifies, guides and automates database migrations to Azure, supporting various databases like SQL Server, MySQL, PostgreSQL, Oracle and MongoDB.
- Database-sensitive migration service moves data, schema and objects to Azure.
- Highly resilient and self-healing migration service provides reliable outcomes with near-zero downtime.
9.2.5. AWS Database Migration Service
- AWS Database Migration Service (AWS DMS) helps to migrate databases and data warehouses to AWS with minimal downtime and zero data loss. The migration service is executed in a secure manner.
- Supports homogeneous database migrations from source database to the equivalent and compatible engine in Amazon.
- Supports heterogeneous database migrations for schema conversions from commercial engines, such as Microsoft SQL Server, to Amazon Aurora PostgreSQL-Compatible Edition and Amazon Relational Database Service (Amazon RDS) for PostgreSQL.
- DMS Schema Conversion in AWS Database Migration Service (AWS DMS) makes database migrations between distinct types of databases more predictable. Automatically converts your source database schemas and most of the database code objects to a format compatible with the target database.
9.2.6. Codacy
| Features | Description |
|---|---|
| Automatic Issue Detection | Identify code issues, security vulnerabilities and code style violations |
| Suggestions | Codacy uses Generative AI to provide smart suggestions and recommendations for improving code |
| Suggest Fixes | Suggest specific fixes or refactoring patterns |
9.3. Migration – Test
9.3.1. Functionize
| Features | Description |
|---|---|
| Autonomous Test Creation | Test plans, test cases and workflows are automatically build using plain English text. |
| Diagnosis and Triaging | Identify potential test issues and offer actionable solutions |
| Autonomous Maintenance | Tests dynamically adapt to application changes, minimizing maintenance efforts and ensuring continued reliability. |
| Report and Document | Tool can auto generate test case metrics and coverage reports. |
| Automated Test Generation | Generate unit test cases for code. |
9.3.2. Applitools
| Features | Description |
|---|---|
| Test Creation |
|
| Test Execution |
|
| Test Validation |
|
| Test Analysis |
|
9.4. Migration – Deploy
9.4.1. Harness
| Features | Description |
|---|---|
| Continuous Integration |
|
| Continuous Delivery |
|
| Infrastructure as Code Management |
|
| Incident Response |
|
9.5. Post Migration
Some of tools that use Generative AI in Observability
- Grafana
- New Relic
- Dynatrace
| Features | Description |
|---|---|
| Anomaly Detection | Gen AI can effectively identify patterns in huge sets of data and provide warnings. |
| Root cause analysis | Identify the cause of the problem by performing analysis on the data |
| Predictive Maintenance | Proactively scheduled maintenance tasks by analyzing previous data that led to failures. |
10. Steps to integrate Generative AI into projects
Gen AI provides great opportunities for the organization with impact across productivity, growth, and cost. However, it is important to follow well defined steps to integrate Gen AI into migration and modernization projects.
Key steps to integrate Gen AI
Figure 6: Steps to Integrate Gen AI
The overall success of cloud migration and modernization engagements also depends on a few other strategic factors that need to be applied across the life cycle of projects.
- Governance: Establish a well-defined framework of policies, procedures and guidelines for usage of cloud resources and operations. Ensure responsible and ethical use of Gen AI systems.
- Continuous Innovation: Leveraging the power of cloud and AI to build innovative services and solutions to stay relevant in a competitive environment.
- Hardening: Strengthen the cloud environments to minimize vulnerabilities by implementing security controls. Protect data and ensure business continuity. Ensure AI models leveraged in projects are resilient and protected against malicious inputs.
- Certification: Certification programs are an integral part for overall success as they validate engineers' knowledge and skills to use Cloud and AI for building solutions. It demonstrates proficiency and ensures they are up to date with the evolving technology landscape.
- Compliance: It is important to have practice in place to protect data and enforce security best practices. Regular audit to identify any compliance issues and address the same
- Trusted Advisor: Providing guidance and expertise on leveraging cloud native architectures and Gen AI technologies effectively. Identify use cases, create strategies and implement solutions to help businesses to enhance their operation.
- Implementation Best Practices: Leverage the expertise of organizations who have architected and executed large scale digital transformation programs for different clients. They have a playbook of the best practices that can be applied during the project lifecycle.
- Partnerships: Collaborations with cloud service providers like Amazon Web Services, Microsoft Azure and Google cloud to leverage expertise to drive innovation. Leveraging expertise of organizations who have built a hub for AI research and development.
11. Challenges of using Generative AI in Cloud Migrations and Modernizations
Generative AI offers considerable advantages for migrations and modernizations; however, it also has challenges that need to be addressed.
- Security and Privacy: Training Gen AI models need a vast amount of data which can include proprietary information. Code generated using AI can introduce IP risks and security concerns. It is important for organizations to ensure standards are met before pushing the code into production.
- Inadequate Data: Large amounts of quality data is required for Gen AI models to produce better results. Lack of quality data can be a hindrance in the implementation of models.
- Bias in generated code: Gen AI models can produce biased data based on training data. This can lead to unfair outcomes, hence it’s important to have a human-in-loop in the workflow.
- Model Inaccuracy: Models can produce incorrect data, or drift over time. This will lead to retraining models which can be costly.
- Infrastructure requirements: Demand for very high resource requirements to deploy Gen AI models can be a hindrance for organizations with smaller budgets. Organizations need to have a check and ensure expected ROI, otherwise it will lead to an impact on budgets.
- Right Skill: Lack of specialized skills to implement Gen AI and AI based solutions can be a challenge and lead to missed opportunity of leveraging benefits provided by Gen AI.
- Regulatory and Compliance: AI-driven tools need to comply with regulations around governance, data privacy and transparency. The tools need to explain decision making processes to ensure they are compliant.
12. Future Trends – Role of AI in Cloud Migration and Modernization
Agentic AI is an emerging technology that can act autonomously by making independent decisions and executing tasks with minimal human intervention. The ability to automate complex workflows, real time decision making makes the system more efficient and fosters innovation.
Role of Agentic AI in Cloud Modernization:
- Automated Cloud Infrastructure Management: Agentic AI can automate infrastructure specific tasks like provisioning, scaling, predictive load balancing and managing cloud resources. This can lead to saving effort of IT engineers managing cloud infrastructure.
- Self-Healing: Proactively monitor environments, identify issues and take corrective actions, thereby reducing downtime and improving the system reliability.
- Migration: Agentic AI can analyze dependencies, identify migration issues and automate data migration processes.
- Enhanced Security: Detect and respond to security threats in real-time, thereby protecting cloud infrastructure and applications.
Conclusion
Digital transformation has become a strategic imperative for organizations worldwide which has led to significant improvements in scalability, agility, and cost-effectiveness, they have also introduced complex technical challenges across the migration and modernization lifecycle. With the advent of Generative AI (GenAI), organizations now have an unprecedented opportunity to accelerate and streamline cloud migrations by automating key processes, reducing manual effort, and enhancing decision-making. This paper presents a detailed view on the capabilities of Gen AI in each phase of migration lifecycle. The impact of some of the leading Gen AI and AI based tools in modernization projects. Transformation brought by AI-driven automation in DevSecOps has also been discussed in this paper.
References
Throughout the preparation of this whitepaper, information and insights were drawn from a range of reputable sources, including research papers, articles, and resources. Some of the key references that informed the content of this whitepaper include:
- https://aws.amazon.com/
- https://codeassist.google/
- https://azure.microsoft.com/
- https://www.ibm.com/think/topics/application-modernization
- https://www.ibm.com/think/topics/ai-transformation
- https://www.ibm.com/consulting/cloud-accelerator
- https://www.finops.org/
- https://cloudsecurityalliance.org/
- https://www.functionize.com
These references provided the foundation upon which the discussions, insights, and recommendations in this whitepaper were based.
- “Accelerating Cloud Migration Efficiency with Generative AI”- Bijit Ghosh, https://medium.com/@bijit211987/accelerating-cloud-migration-efficiency-with-generative-ai-fdc7eaea1509
- “Agentic AI: The New Trend in Generative AI”-Vishnu Mohandas, https://blogs.infosys.com/emerging-technology-solutions/artificial-intelligence/agentic-ai-the-new-trend-in-generative-ai.html#:~:text=Agentic%20AI%20systems%2C%20consisting%20of,for%20addressing%20real%2Dworld%20problems.
- “AI for IT modernization: Faster, cheaper, better” - AI agents for IT modernization | McKinsey
- “The Evolution of DevSecOps with AI”- Rahul Kalva, The Evolution of DevSecOps with AI | CSA
- “Top GenAI Testing Tools” - Top GenAI Testing Tools | GeeksforGeeks
Subscribe
To keep yourself updated on the latest technology and industry trends subscribe to the Infosys Knowledge Institute's publications
Count me in!