Insight and analysis on the data center space from industry thought leaders.

Six Considerations for Adopting a Microservices Architecture

The microservices approach complements cloud and DevOps very well.

Industry Perspectives

August 12, 2019

5 Min Read
DataCenterKnowledge logo in a gray background | DataCenterKnowledge

vinil (002).jpg

vinil (002)

Vinil Menon is Chief Technology Officer at Citius Tech.

KhushbooS.png

KhushbooS_0

Khushboo Shah is Senior Solution Architect at Citius Tech.

The next big wave of emerging technologies is upon us as artificial intelligence, wearables, the Internet of things (IoT), and more become prevalent. Many organizations are now faced with the challenge of managing these monolithic applications. It’s a given that speed and agility are required in today’s environment. The next architecture innovation used by large web companies like Netflix, Twitter, eBay and Amazon is microservices. According to web server vendor NGINX, 68 percent of organizations are using or investigating this approach.

Microservices architecture is gaining momentum in the IT industry for the many advantages it offers over traditional architectural approaches. For example, in healthcare, this architecture holds great promise for new solutions, such as remote patient monitoring, predictive modeling using IoT device data, automating healthcare workflows, bioinformatics analysis for precision medicine, and more. As organizations adopt a modern microservices architecture, there are six considerations that will position them for success while complementing existing cloud and DevOps infrastructure.

How a Microservices Architecture Complements Cloud and DevOps

Microservices is a software development and architectural approach that structures applications as a collection of loosely coupled, autonomous, and independently deployable services. These small, business-driven services have clearly defined interfaces to communicate, which makes application modular, easier to build and test, and efficient to deploy independently.

The microservices approach complements cloud and DevOps very well. Cloud computing has matured over the years to provide highly efficient infrastructure solutions for prototyping rapidly, supporting enormous data processing and production needs, and enabling higher service levels as compared to internal IT organizations. In parallel, DevOps enables faster delivery of quality software while bridging the gap between the developers and the operations teams.

Combining microservices architecture with sound DevOps practices empowers decentralized teams to innovate faster, control their own technology stack and standards, manage performance metrics, govern development and release cycles, and ultimately reduce the time to market. In parallel, microservices can facilitate a gradual move to the cloud by decomposing monolithic applications into microservices and deploying them on a cloud platform. With this approach, teams can more easily simulate production workloads to ensure the availability, scalability and quality of software, while releasing at shorter intervals.

Six Microservices Architecture Design Considerations

The following six considerations help position organizations for success while complementing existing cloud and DevOps infrastructure:

Service Discovery: In complex distributed systems, teams scale the instances of the services based on the load, meaning the number of service instances and their locations can change dynamically. Having a proper service discovery mechanism ensures that clients can communicate with the right service based on the service registry.

Inter-Service Communication: In a microservices architecture, services communicate either synchronously or asynchronously to complete a transaction. An inter-service communication mechanism orchestrates this communication. If poorly designed, too much communication between services results in a “chatty” application and poor performance. To optimize, employ these proven design patterns:

  • Saga Pattern: A sequence of transactions triggered by an event or message.

  • API Gateways: Abstract the calls to multiple APIs via a single API call through the API Gateway.

  • Command Query Responsibility Segregation (CQRS): Separate writes from reads using materialized views and update the views via subscribed events.

  • Event Sourcing: Store events rather than states; derive the state by replaying the events.  

  • Service Mesh: Offload network communications between services to a software component to ensure resiliency, service discovery, etc.

Data Integrity: Because each microservice has its own database, ensuring data consistency across transactions that span multiple services can be a challenge. The patterns above, such as Event Sourcing, CQRS and Saga, help achieve data consistency.

Security: Encryption mechanisms as well as strong authentication and authorization tools exist to ensure data security both at rest and in transit. Some organizations employ Identity-as-a-Service and Authorization-as-a-Service solutions. In addition, APIs need to be secured behind an API Gateway to ensure that unauthorized users cannot gain access using a token.

Monitoring and Health Check: As the number of services grows within the microservices architecture, identifying and troubleshooting issues becomes challenging. For example, a single transaction can span multiple service calls making it difficult to identify the exact cause of performance bottlenecks. Monitoring the system using distributed transaction tracing, and health check APIs ensure the applications are running efficiently. With the right monitoring and instrumentation, teams can create visualizations for key metrics, gather historical data for performance trends and receive alerts when issues arise.

Quality Assurance: Microservices handle requests by passing messages between services. As the number of services increases, automated testing is important to ensure all the interactions and communications are thoroughly tested, including unit tests, integration tests, component tests and contract tests. Further, thorough testing at each level confirms that services can run independently, in conjunction with other services, to support distributed transactions. Lastly, testing demonstrates that the overall architecture is nimble enough to support additional data sources, frameworks or libraries as needed.

Faster Innovation is the Goal

Microservices architecture is an important IT trend today for good reason. It offers many advantages over traditional architectural approaches and holds great promise. While there are challenges to overcome, organizations with a well-designed approach, thoughtfully organized and decentralized teams, and solid DevOps processes can innovate faster– by building new products and modernizing existing applications – with a modern microservices architecture.  

Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating.

Subscribe to the Data Center Knowledge Newsletter
Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily.

You May Also Like