Amazon Web Services (AWS) stands at the forefront of cloud computing, offering a wide range of public cloud services to businesses, organizations, and developers across the globe. With its ever-growing demand, AWS has developed an expansive and highly scalable global infrastructure that ensures customers experience high performance, low latency, and enhanced scalability. The continuous evolution of AWS’s infrastructure allows businesses to deploy applications, run services, and store data seamlessly, making AWS an indispensable tool for enterprises in the digital age.
In this article, we will explore the core components of AWS’s global infrastructure. These include Availability Zones (AZs), Regions, Edge Locations, Regional Edge Caches, Local Zones, Wavelength Zones, and Outposts. To fully leverage the potential of AWS, understanding the role and functionality of each of these components is crucial. By the end of this discussion, you will have a clear understanding of how these elements interact to form the robust cloud services AWS is known for.
AWS Global Infrastructure: A Foundation for Cloud Solutions
AWS has strategically built its infrastructure to offer resilient, flexible, and high-performance cloud services to meet the ever-expanding demands of modern businesses. Whether an organization requires data storage, computational power, or machine learning capabilities, AWS provides a reliable and efficient platform to host and scale their applications. The design of AWS’s global infrastructure involves multiple key components working in tandem, ensuring seamless cloud deployment for a variety of use cases.
Availability Zones: The Building Blocks of AWS Infrastructure
At the heart of AWS’s infrastructure are Availability Zones (AZs), which represent the physical data centers that host AWS’s cloud services. These AZs are crucial for ensuring the reliability and availability of AWS’s offerings. Each Availability Zone is designed with multiple, independent power, networking, and cooling systems, which reduce the risk of service disruption due to hardware or power failures.
Contrary to common belief, an Availability Zone is not a single data center. Rather, it consists of multiple data centers situated within close proximity to each other, connected by low-latency private fiber-optic links. This configuration allows AWS to ensure data replication and synchronization between the AZs, further enhancing the platform’s fault tolerance. If one data center in an AZ fails, the others within the same zone can continue operating, ensuring that services remain accessible.
In addition, the close proximity of multiple AZs within a region enables AWS customers to create highly resilient applications by distributing their workloads across several AZs. This approach significantly reduces the risk of downtime caused by localized failures, as resources are spread out, ensuring that if one AZ goes offline, others can take over seamlessly.
Regions: Geographical Distribution of AWS Services
AWS Regions are large geographical areas that host multiple Availability Zones. Each region is independent of the others, providing geographical redundancy, improved performance, and reduced latency for customers. AWS strategically deploys regions across the globe to ensure that businesses and users can access services with minimal latency, regardless of their location.
A typical AWS region contains at least two Availability Zones, though some regions may have even more, with regions spread across major cities worldwide. For example, businesses in Europe may prefer to deploy their services in the Ireland or Frankfurt regions to avoid the latency that would occur if their services were deployed in North America or Asia.
Regions also play a critical role in complying with local data residency and regulatory requirements. For example, AWS offers regions in various countries to help customers comply with local laws that require data to remain within specific geographic boundaries. This enables organizations to store and process sensitive data in accordance with regional legal and compliance standards, ensuring secure and lawful data management.
Edge Locations: Accelerating Content Delivery
Edge Locations are part of AWS’s Content Delivery Network (CDN), primarily used by services such as AWS CloudFront. These locations are distributed across major cities around the world, far outnumbering the Availability Zones. While Edge Locations do not host the primary compute or storage services, they play a key role in accelerating the delivery of content to end users.
When users access content through services like CloudFront, Edge Locations cache data and deliver it from the nearest location, reducing the time it takes to load websites, applications, and media. This minimizes latency, providing a faster, more efficient user experience. For example, if a user in Japan accesses a website hosted on AWS, the request will be served from the nearest Edge Location, even if the main infrastructure is located in a different region.
By caching content closer to users, AWS’s Edge Locations reduce the load on origin servers, resulting in improved performance and bandwidth optimization. They also allow content to be delivered at scale, making them ideal for high-traffic applications and websites.
Regional Edge Caches: Bridging the Gap for Faster Content Access
AWS introduced Regional Edge Caches to further optimize content delivery. These caches sit between the CloudFront Origin servers and the Edge Locations, providing larger storage capacity for cached data. The main advantage of Regional Edge Caches is that they store content for a longer duration compared to Edge Locations.
When content expires or is not available in an Edge Location’s cache, the request is directed to the Regional Edge Cache. By retrieving data from the Regional Edge Cache, the need to access the origin servers is reduced, improving load times and decreasing latency. This additional layer of caching helps ensure that users experience consistent and quick access to content, even during peak demand times.
Local Zones: Enhancing Latency for Specific Geographies
AWS Local Zones are designed to bring AWS’s core services closer to end users in specific metropolitan areas where an AWS region may be too far away to provide low-latency access. These zones are particularly beneficial for applications that require single-digit millisecond latency, such as gaming, media, and financial applications.
For instance, in large cities such as Boston, New York City, and Miami, where the nearest AWS region may be 100 miles or more away, AWS Local Zones offer an ideal solution for businesses that need to deploy latency-sensitive applications. These Local Zones are fully integrated with the parent AWS region, allowing users to seamlessly connect to other AWS services as if they were operating within the same region.
AWS’s Local Zones are expanding rapidly, and currently, they are available in 33 metropolitan areas worldwide, with plans for further expansion. They offer the same suite of AWS services, including compute, storage, and database services, enabling businesses to deliver high-performance solutions to their users in specific regions.
Wavelength Zones: Optimizing 5G Connectivity
Wavelength Zones extend AWS infrastructure to the edge of telecommunications networks, embedded within 5G mobile networks. By integrating AWS services with 5G infrastructure, Wavelength enables ultra-low latency applications like live video streaming, autonomous vehicles, and augmented reality, which require near-instantaneous processing.
With AWS Wavelength, resources such as EC2 instances and VPC subnets can be deployed within telecom networks, significantly reducing the time it takes for data to travel to and from the cloud. This reduced latency enhances the overall experience for users, making Wavelength ideal for mobile applications, gaming, and other real-time services.
Currently, Wavelength Zones are available through partnerships with major telecom providers such as Verizon in the United States, Vodafone in Europe, and KDDI in Japan, enabling developers to build applications that leverage the power of 5G technology.
Outposts: Extending AWS Cloud to On-Premises Data Centers
AWS Outposts allow businesses to extend AWS’s cloud infrastructure into their on-premises data centers. By using the same AWS hardware as found in AWS data centers, businesses can deploy and manage AWS services on-premises, allowing for hybrid cloud configurations that combine both on-site and cloud-hosted resources.
Outposts are ideal for businesses that require a hybrid approach, such as those with existing on-premises infrastructure that need to integrate with cloud-based services. AWS manages the Outposts hardware, ensuring that customers benefit from seamless updates, maintenance, and scalability, without the need for manual patching or hardware management.
Organizations can use AWS Outposts to run core services like EC2, S3, and RDS locally, while still benefiting from the full power and flexibility of the AWS cloud.
Overview of AWS Global Infrastructure
Amazon Web Services (AWS) operates a comprehensive and robust global infrastructure that serves as the backbone for its cloud services. This infrastructure is a network of interconnected systems that work seamlessly to deliver scalable, secure, and high-performance cloud solutions. By providing a powerful cloud environment for hosting applications, managing data, and ensuring uninterrupted service availability, AWS has become a leader in cloud computing.
The AWS global infrastructure is designed with flexibility and reliability in mind, catering to a wide range of customer needs, whether they’re located in major cities or remote regions. This article will explore the key elements that make up AWS’s global infrastructure and how they contribute to delivering world-class cloud services.
Key Components of AWS Global Infrastructure
AWS’s global infrastructure consists of several core components that collaborate to ensure optimal performance and availability. These components are specifically designed to handle varying levels of demand, secure sensitive data, and support a global customer base. Below, we explore the critical components of AWS’s infrastructure.
1. Regions
At the heart of AWS’s global infrastructure are the AWS regions. A region is a geographic area that contains multiple availability zones (AZs). Each region is physically isolated from other regions to reduce the risk of failure due to local events like power outages, earthquakes, or other disruptions. AWS has data centers in various parts of the world, allowing customers to select the region that best meets their needs in terms of proximity, performance, and compliance requirements.
AWS regions are distributed across multiple continents, offering customers the flexibility to choose where to deploy their applications and store their data. This regional approach enables customers to comply with local data residency regulations and provide their users with lower-latency access to services.
For example, AWS has regions in North America (like US East and US West), Europe (such as Ireland and Frankfurt), Asia Pacific (including Tokyo and Sydney), and other parts of the world. Each of these regions is independent, meaning that any local failure in one region does not affect the operation of others.
2. Availability Zones (AZs)
An Availability Zone (AZ) is a discrete data center within a region that is designed to operate independently yet is interconnected with other AZs in the same region. Each AZ is equipped with its own power supply, cooling systems, and networking infrastructure, ensuring that it can continue operating in the event of a failure in another AZ.
AWS typically deploys multiple AZs in each region, providing customers with high levels of availability and fault tolerance. If one AZ experiences an issue, applications and services can failover to another AZ within the same region, minimizing downtime and service disruptions. By distributing resources across multiple AZs, AWS ensures that customers can achieve greater reliability and resiliency for their applications.
This architecture also allows for the creation of highly available and fault-tolerant systems. Customers can configure their applications to replicate data across multiple AZs, ensuring that they remain operational even if an entire data center goes offline.
3. Edge Locations and Content Delivery Network (CDN)
To further enhance the performance and availability of its services, AWS utilizes edge locations that are strategically placed around the world. These edge locations are part of AWS’s Content Delivery Network (CDN), known as Amazon CloudFront. CloudFront caches content at the edge locations, enabling faster delivery of web content, applications, and streaming media to end users.
When a user accesses an AWS-hosted website or application, CloudFront automatically directs their request to the nearest edge location, reducing latency and improving load times. This global network of edge locations ensures that users from anywhere in the world can access AWS-hosted content with low latency, regardless of their physical location.
By leveraging a CDN, AWS customers can provide an optimized and responsive experience for users globally, whether they’re accessing media content, downloading software, or interacting with dynamic web applications.
4. AWS Direct Connect
To offer secure and high-performance network connections, AWS provides AWS Direct Connect. This service enables customers to establish a dedicated network connection between their on-premises infrastructure and AWS. Direct Connect allows businesses to bypass the public internet, ensuring a more secure, reliable, and consistent connection to AWS services.
With Direct Connect, customers can create hybrid cloud architectures, linking their data centers directly to AWS regions for seamless data transfer. This service is particularly beneficial for applications that require large volumes of data transfer, as it provides lower latency and better bandwidth compared to traditional internet connections.
AWS Direct Connect is available at various locations around the world, further enhancing the global reach and connectivity of the AWS network.
5. Global Databases and Data Replication
AWS provides global database services that are designed to handle mission-critical workloads with global reach. Services like Amazon Aurora Global Databases and Amazon DynamoDB Global Tables enable customers to replicate data across multiple regions to support applications that require high availability, low-latency access, and disaster recovery.
With these global database solutions, data can be automatically replicated across different AWS regions, ensuring that users in different geographical areas can access data quickly. These services are particularly useful for applications with a global user base that need to maintain consistent performance, even if one region experiences an outage.
Additionally, AWS offers various options for data backup and disaster recovery, allowing customers to implement strategies for recovering data in the event of a regional failure.
6. Security and Compliance
Security is a critical aspect of AWS’s global infrastructure. The company takes a multi-layered approach to security, incorporating physical security at its data centers, network security at every layer, and robust encryption for data in transit and at rest. AWS also provides a wide range of security services and tools that help customers protect their data and applications.
In addition, AWS’s infrastructure complies with numerous industry standards and regulatory requirements, such as GDPR, HIPAA, SOC 1, 2, and 3, and ISO 27001. Customers can leverage AWS’s compliance programs to meet their own regulatory requirements and ensure that their applications and data are secure and compliant.
Benefits of AWS Global Infrastructure
AWS’s global infrastructure offers numerous benefits to businesses, including:
- High Availability and Redundancy: The presence of multiple regions and availability zones ensures that applications can remain highly available even in the face of hardware failures or other disruptions.
- Scalability: AWS’s infrastructure is built to scale easily, allowing businesses to expand or contract their resources as needed, without the need for significant upfront investment.
- Global Reach: With AWS data centers located in multiple regions around the world, businesses can provide services with low latency to customers in virtually any part of the globe.
- Security and Compliance: AWS offers a secure environment for running applications, with encryption, monitoring, and compliance with global standards to protect sensitive data.
- Cost Efficiency: The AWS model allows businesses to pay only for the resources they consume, making it easier to manage costs while still benefiting from a global network of data centers.
Availability Zones (AZs)
At the core of Amazon Web Services’ (AWS) robust global infrastructure are the Availability Zones (AZs), which function as the physical data centers hosting a majority of AWS’s cloud services. AZs are designed with a high level of redundancy to guarantee the continuous availability of services and to minimize the risk of downtime. Each AZ is equipped with its own independent power supply, networking infrastructure, and cooling systems, ensuring that an issue in one zone doesn’t impact others in the region.
Typically, a region in AWS consists of multiple Availability Zones, each of which is strategically located within close proximity to the others. This geographic clustering ensures that AWS can provide low-latency connections between zones, promoting better communication and data replication across them. Despite this proximity, each AZ is completely isolated from others, providing a layer of protection. In the event of an issue such as a power failure or a natural disaster affecting one AZ, the isolation helps ensure that the other zones remain unaffected.
A critical feature of AZs is their ability to support fault tolerance and high availability. AWS’s design ensures that data can be replicated across multiple AZs, significantly improving the resilience of services. For instance, services like AWS Relational Database Service (RDS) rely on multi-AZ deployments to replicate data across different Availability Zones. This replication process enhances data durability and increases uptime by making sure that, even if one zone goes down, the data is still available from another zone.
AWS’s emphasis on high availability through the use of AZs is a fundamental component of its cloud service offering. Customers can architect applications in a way that makes use of multiple AZs within a region to safeguard against potential disruptions. This approach not only improves fault tolerance but also ensures the business continuity of cloud-hosted applications. By deploying applications across multiple Availability Zones, AWS users can mitigate risks such as network disruptions or hardware failures, ensuring that services remain operational, regardless of challenges faced by one particular AZ.
Best practices in deploying applications on AWS recommend distributing workloads across more than one AZ within a region. This strategy is an essential part of maintaining the reliability and performance of applications, especially for critical workloads that require uninterrupted service. Even if one AZ experiences an outage, the failover mechanism built into the multi-AZ architecture ensures that the workload is redirected to another zone, thus minimizing the impact on end users.
Additionally, the low-latency fiber-optic links between Availability Zones play a significant role in enabling seamless data replication and application scaling. AWS customers benefit from this network architecture, as it enables high-speed, synchronous replication of data, enhancing the overall performance of applications and ensuring that end users can access up-to-date, consistent information, no matter where they are.
The design of AWS Availability Zones also allows for scalability. With multiple AZs within a region, customers can scale their workloads by utilizing the resources of several zones, thereby avoiding overloading a single zone with traffic. This elasticity makes it easy to accommodate growing demands, whether in terms of user traffic, storage needs, or computational power.
Moreover, each AZ is supported by dedicated AWS teams that actively monitor the health of the infrastructure and ensure that systems remain operational. In addition, AWS’s use of advanced technologies such as machine learning and automation helps to predict potential issues and prevent outages before they can affect services.
For businesses operating globally, leveraging AWS’s Availability Zones provides the flexibility to ensure high availability and low-latency access for customers in different regions. AWS customers can deploy their applications in the region closest to their target audience, benefiting from the geographically distributed AZs that AWS offers. Whether serving customers across North America, Europe, or Asia, AWS enables companies to run their applications closer to where users are located, providing them with faster response times and a better overall experience.
Regions: Defining Geographic Areas for AWS Cloud Services
Amazon Web Services (AWS) organizes its vast infrastructure into geographic areas known as regions. These regions consist of multiple Availability Zones (AZs) that work together to deliver cloud services efficiently. Each AWS region is designed to operate independently, providing resilience by ensuring that any failure in one region does not affect the services of another region. This decentralized structure is a key element in making AWS a reliable and scalable cloud platform, offering customers high availability and continuity of service.
AWS has strategically placed its regions around the world to ensure that customers can access cloud services with minimal latency, no matter their location. The physical proximity of these regions to their end users plays a significant role in the speed and performance of cloud-based applications. For businesses with a global presence, selecting the right AWS region is essential for providing an optimal user experience. For instance, a business operating in Europe would gain significant advantages by deploying its resources in AWS regions such as Ireland or Frankfurt. Choosing a region closer to the target audience reduces network latency and ensures that applications run smoothly and efficiently, offering users faster response times.
Each AWS region is made up of multiple Availability Zones, which are essentially isolated data centers within that region. These AZs are connected by low-latency, high-bandwidth fiber-optic links, ensuring data replication and service availability even in the event of a failure in one AZ. The availability of multiple AZs within a region ensures fault tolerance, allowing AWS customers to design highly available and resilient architectures. For businesses, the use of multiple AZs within a region guarantees that their services remain operational, even if one data center experiences a disruption.
The autonomy of each AWS region is another important factor in the platform’s resilience. Since AWS regions are isolated from each other, an issue in one region does not have a cascading effect on other regions. This means that a failure, whether caused by natural disasters, power outages, or network issues, in one region will not impact the services running in another. AWS’s global network of regions allows organizations to build disaster recovery solutions that can span across multiple regions, ensuring business continuity in case of emergencies.
Another key consideration for businesses when selecting an AWS region is regulatory and compliance requirements. Each region adheres to the legal and regulatory standards specific to the geographical area in which it operates. This is particularly important for industries that handle sensitive data, such as healthcare, finance, and government. Different countries and regions have varying laws regarding data residency, privacy, and protection, and AWS provides tools to help businesses stay compliant with these regulations.
For example, organizations in Europe must comply with the European Union’s General Data Protection Regulation (GDPR), which sets strict rules on how personal data is handled. AWS regions in the EU, such as those in Ireland and Frankfurt, are fully equipped to help businesses meet these requirements by ensuring that data remains within the region and complies with local laws. Similarly, businesses in the U.S. may need to adhere to regulations such as the Health Insurance Portability and Accountability Act (HIPAA) for healthcare data, or the Sarbanes-Oxley Act (SOX) for financial data. AWS has designed its regions to support these requirements, giving businesses the flexibility to select the most appropriate region for their data storage and processing needs.
Furthermore, AWS regions offer flexibility in terms of where businesses store and process data, allowing them to optimize their infrastructure for both performance and compliance. By choosing the right region, businesses can ensure that they meet data residency and data sovereignty requirements, mitigating the risks associated with non-compliance. This is especially crucial for companies that handle customer data across different jurisdictions and must adhere to a variety of legal frameworks.
In addition to compliance, the physical security of data centers within AWS regions is a top priority. AWS invests heavily in the security of its data centers to protect both customer data and its infrastructure. Data centers are equipped with state-of-the-art physical security measures, including surveillance, biometric access controls, and multi-factor authentication for personnel access. Additionally, AWS regions benefit from the platform’s strong focus on data encryption, both in transit and at rest, providing customers with an added layer of security.
One of the benefits of AWS’s global network of regions is the ability to deploy applications and services in a way that is geographically distributed to optimize performance, availability, and compliance. Organizations can choose to deploy their applications in specific regions to meet local market needs, improve latency, and adhere to regulatory requirements. Whether a company is serving customers in North America, Europe, Asia, or elsewhere, AWS provides the necessary infrastructure to support a global operation.
For instance, a business headquartered in the U.S. that wants to provide services to European customers can deploy its applications in AWS’s U.S. East region and replicate critical data to an EU-based region to ensure compliance with GDPR. This setup also allows the business to deliver fast, responsive services to users in Europe while maintaining data sovereignty.
AWS’s expanding global footprint also means that regions are continuously added to meet the increasing demand for cloud services. As of 2023, AWS operates 32 regions worldwide, with more regions in development to support growing global needs. This continual expansion ensures that AWS customers have access to new geographic areas, giving them more options for deploying their cloud services in ways that best meet their specific requirements.
Edge Locations: Bringing Services Closer to Users
Edge Locations are strategically positioned AWS sites located in major cities worldwide. These sites act as Content Delivery Network (CDN) endpoints and are used by services like AWS CloudFront and AWS Lambda@Edge to reduce latency and deliver content quickly to end users. Edge Locations cache data, providing quicker access to frequently requested content and reducing the load on origin servers.
For example, if you host a website in AWS and use CloudFront as a CDN, your users will be directed to the nearest Edge Location when they request your content. This reduces the time it takes to load your website, improving the user experience.
Although Edge Locations do not host core AWS services like EC2 or S3, they play a critical role in speeding up the delivery of content and enhancing performance for end users around the world.
Regional Edge Caches: Bridging the Gap
Regional Edge Caches are another vital part of AWS’s global infrastructure, introduced to provide faster access to cached content. Positioned between AWS Origin servers (where your primary data resides) and Edge Locations, Regional Edge Caches provide a larger, more persistent cache for data that is not frequently accessed.
When data expires in the Edge Location cache, the request for that data is sent to the Regional Edge Cache, which stores a larger volume of content. This helps reduce the frequency of accessing the origin server, thereby improving the performance and reducing latency for users.
Local Zones: Delivering Services Near Major Urban Areas
AWS Local Zones extend AWS services to major metropolitan areas that are not close to an AWS region. Local Zones bring essential compute, storage, networking, and database services closer to end-users in large cities where AWS does not have a full region. For example, AWS Local Zones are used to serve areas like Boston, New York, and Miami, where the distance to the nearest AWS region would otherwise result in unacceptable latency for latency-sensitive applications.
These Local Zones are connected to their parent AWS regions through secure, high-speed connections, ensuring seamless integration with other AWS services. They are ideal for use cases such as gaming, media, and other latency-sensitive applications where low-latency access to cloud resources is crucial.
In addition to Local Zones, AWS also offers Dedicated Local Zones, which are tailored infrastructure solutions for specific customers or communities. These zones provide more secure and compliant environments, making them suitable for industries like government or healthcare where data sovereignty and security are of utmost importance.
Wavelength Zones: Powering 5G Applications
AWS Wavelength brings AWS services to the edge of telecommunications networks, specifically embedded within 5G networks. These Wavelength Zones enable ultra-low latency applications by placing AWS compute and storage resources closer to end-users, inside the telecommunications network.
Wavelength is designed for applications that require near-instantaneous processing, such as live video streaming, interactive gaming, and autonomous vehicles. By reducing the number of network hops and eliminating the need for traffic to traverse the public internet, Wavelength ensures faster data processing and a more responsive user experience.
AWS Wavelength Zones are deployed in collaboration with major telecommunications providers, including Verizon, Vodafone, and KDDI, enabling 5G applications to run at their full potential with improved reliability and lower latency.
Outposts: Extending AWS Services to On-Premises Environments
AWS Outposts allow organizations to extend AWS infrastructure into their on-premises data centers. By deploying AWS-designed hardware on-site, organizations can run services like EC2, RDS, S3, and more, just as they would in the AWS cloud.
AWS Outposts are ideal for businesses that require hybrid cloud setups, where workloads are spread across both on-premises and cloud environments. Outposts offer seamless integration with AWS’s cloud services, ensuring consistency in management and automation regardless of the location of the resources.
Outposts are fully managed by AWS, meaning that businesses do not need to worry about maintenance or patching. AWS handles all hardware updates and software upgrades, providing businesses with a reliable and hassle-free hybrid cloud solution.
Conclusion:
The AWS global infrastructure is a comprehensive network designed to support businesses’ growing need for scalable, secure, and low-latency cloud services. By understanding the roles of Availability Zones, Regions, Edge Locations, Regional Edge Caches, Local Zones, Wavelength Zones, and Outposts, organizations can design highly available and resilient cloud environments.
Whether you need to improve the performance of your website with CloudFront, reduce latency for 5G applications with Wavelength, or extend AWS services into your on-premises data center with Outposts, AWS provides the flexibility and scalability necessary for today’s digital world. By leveraging the right combination of these components, businesses can ensure that their applications are always available, highly performant, and compliant with local data regulations.
As AWS continues to expand its infrastructure globally, the future promises even more robust solutions for enterprises of all sizes. Whether you’re building a global application or a local service, AWS’s expansive infrastructure provides the foundation for innovation and success in the cloud.
In conclusion, AWS regions form the foundation of AWS’s global infrastructure, providing geographic diversity, scalability, and resiliency for businesses worldwide. By strategically deploying regions around the globe, AWS ensures that customers can access services with low-latency, meet regulatory requirements, and create disaster recovery strategies that span multiple regions. For businesses, choosing the right AWS region is critical to optimizing performance, maintaining compliance, and offering a seamless user experience across global markets. As AWS continues to expand its network of regions, businesses can rely on the platform to provide the flexibility and reliability needed to succeed in the cloud.