

# Design principles

 The following design principles can help you achieve and maintain efficient workloads in the cloud. 
+  **Use managed services to reduce operational complexity:** Reduce the effort of advanced technology implementation by delegating complex tasks to your cloud vendor. Rather than requiring your team to learn how to host and run specialized technologies, consume these technologies as services. For example, NoSQL databases, media transcoding, and machine learning all require specialized expertise but are available as optimized, performance-tuned services in AWS. This lets your team focus on product development instead of resource provisioning and management, while benefiting from continuously improved performance and scalability. 
+  **Deploy workloads closer to your customers:** Go global in minutes by deploying your workload across multiple AWS Regions around the world. This provides lower latency and a better experience for your customers at minimal cost. 
+  **Build scalable architectures:** Most workloads have fluctuating access patterns. In the cloud, you can scale up and down in minutes. For example, you can use serverless services to build scalable architectures and remove the undifferentiated heavy lifting by scaling based on your workload requirements. 
+  **Experiment more often:** With virtual and automatable resources, you can quickly run comparative testing using different types of instances, storage, or configurations. Capture the lessons learned from these experiments and continuously fine-tune your architecture as your workload requirements change. 
+  **Consider mechanical sympathy:** Use the technology approach that aligns best with your goals. For example, consider your data access patterns when you select a database or storage for your workload, and choose the right compute and instance types to meet your computational needs based on business KPIs. Apply the same thinking to how you move and access data: use caching to avoid unnecessary round trips to the data layer, and prefer services that cache by default or integrate directly with managed caching services. 
+  **Design for change and variable demand:** Build loosely coupled, distributed architectures so that purpose-built components can scale independently and you can mix serverless and containerized services based on the performance needs of each one. Design these components to absorb traffic spikes and tolerate partial failures, using techniques such as dynamic scaling, throttling, back-off with jitter, load shedding, and circuit breakers to maintain consistent performance while using resources efficiently. 

 These techniques work together to support consistent performance under variable load and partial system failures, helping maintain reliability while efficiently utilizing resources. 