Matrix Labs Provides Full Technical Support to Multiple NFT Projects on AWS

The largest challenge for Matrix Labs to provide Web 3.0 services is how to provide convenient user access around the world, with low latency and high efficiency comparable to local services, especially for NFT projects. As AWS runs 96 availability zones in 30 regions around the world, it can perfectly match such needs for wide geographical coverage of Matrix Labs.

Drawing on AWS’s best practices and solution architectures and following a well-documented guidance, Matrix Labs has built a globally accessible Web 3.0 platform. With Amazon Route 53 at the front end, Matrix Labs’ technical team can define policies for hosted domain names, which can intelligently analyse access requests from users around the world, and direct them to the nearest low-latency and healthy back-end services. At the back end, the combined use of Amazon Elastic Kubernetes Service (EKS) and Amazon Elastic Compute Cloud (Amazon EC2) help Matrix Labs create a managed Kubernetes service environment to grant applications high availability and automatic scalability, and better meet the growing needs of projects. By fully deploying its business on AWS, Matrix Labs has strengthened the ties with partners around the world and improved business efficiency, enabling its technical team to build a more comprehensive framework for business collaboration.

To support various NFT projects, it is necessary for Matrix Labs' technical team to find a more reliable way to store on-chain data, metadata, user information and other raw data. Amazon Simple Storage Service (Amazon S3) provides nearly limitless object storage capability that is particularly suitable for NFT works. With a data durability as high as 99.99999999999 percent (11 nines), Amazon S3 service provides Matrix Labs with unparalleled storage for all types of NFT work data. The integration of Amazon S3 with Amazon CloudFront also allows users to quickly load the content they need. Besides, Amazon S3 service offers a range of storage classes for different data access needs, namely flexible choices based on data access, resiliency and cost requirements according to the workload, which allows Matrix Labs to reduce the operation costs of each NFT project while meet the needs of developers.

As NoSQL database is required for the NFT and other Web 3.0 applications, Matrix Labs technical team selected Amazon DynamoDB, a fully managed, serverless NoSQL key-value database, to store metadata and automatically scale out for growing Web 3.0 business. Meanwhile, the serverless architecture of Amazon DynamoDB also frees Matrix Labs from needing to patch or optimize the performance of the underlying operating system and database software.

"By developing Web 3.0 applications and supporting NFT projects on AWS, we have brought the availability to 99.99 percent for most of our NFT services. Thanks to Amazon DynamoDB, we have reduced the query latency of our NFT projects from seconds to less than 200ms now, far exceeding our expectations," says Alvin Sun, CSO of Matrix Labs.

Matrix Labs adopts a managed workflow orchestration solution of Amazon Managed Workflows for Apache Airflow (MWAA), allowing its technical team to continue using their familiar Apache Airflow platform. Integrating MWAA with Amazon CloudWatch monitoring service further reduces the costs of operation and engineering designs. With the fully managed Apache Kafka real-time stream data ingestion and processing capacity of Amazon Managed Streaming for Apache Kafka (MSK), Matrix Labs makes it much easier to ingest and process log and event streaming in its Web 3.0 applications and NFT projects.

By leveraging AWS managed services in system building, Matrix Labs dramatically improves the overall system performance and reduces platform operating costs of its NFT solutions. For example, with the incremental backups of Amazon Elastic Block Store (Amazon EBS), Matrix Labs shortens the time for backup, reduces the impact on current applications, and improves the availability. "The managed services and billing models of AWS are much helpful in relieving our pressures from operation and maintenance, and saving costs as well. As a result, we saved 20 percent on total infrastructure operation costs, by matching the actual needs of each NFT project with more price-competitive reserved instances," says Sun.

Looking ahead, Matrix Labs plans to continue expanding its global presence throughout AWS global network, and has already established subsidiaries in Singapore and the Middle East. The company will also use AWS to deliver reliable Web 3.0 services to more users, especially builders of NFT projects.