MongoDB provides AI-based database management solutions for developers.

On the 13th, MongoDB held its annual developer conference 'MongoDB.local Seoul' and a press conference at the E-Tower in Seoul, where it presented MongoDB's SW technology and strategies for developers.

In his welcoming speech, Jae-Sung Shin, CEO of MongoDB Korea, said, “MongoDB is the No. 1 open source database platform preferred by developers, and our goal is to support developers in developing the most modern applications.”

Sahir Azam, CPO of MongoDB, introduced new DB solutions in his keynote speech, saying, "The ultimate goal of MongoDB is to reduce the time it takes developers to work with data," and "It is different in that it was developed by developers, for developers."

MongoDB has secured 45,000 customers in various industries, and has been recording meaningful growth, increasing from 450 customers last year to 680 customers this year, an increase of 230 customers. Its clients include Samsung Electronics, Riot Games, Flow, Hyundai Motors, Naver Shopping, SSG, and Shinhan Financial Investment.

Modern applications today are developed on a cloud-based basis, and management and scaling across data centers, clouds, edge, and mobile have become increasingly complex to meet diverse needs.

To support this, MongoDB launched the industry’s first distributed cloud platform. MongoDB’s Atlas platform is a developer data solution that accelerates data creation. It maximizes the potential of data by applying document models, providing high availability and powerful performance so that it can be modeled as you code.

Recently, MongoDB has been integrating AI capabilities into its developer data platform Atlas. In June, MongoDB announced MongoDB Atlas ‘Vector Search’, which enables AI-based search and personalization capabilities to be integrated into applications running on Atlas.

Vector search is a function that vectorizes the similarity of data on the MongoDB platform to make it easier to find information. In simple terms, it helps to easily search for information on unstructured data such as audio, video, and images in queries.

For example, when working with insurance-related data at an insurance company, it vectorizes documents and connects vector queries to help easily understand and manage complex information. When problems occur in various industries, solutions can be obtained in the form of a chat based on LLM linked to existing registered information. This is expected to reduce development time and costs.

MongoDB is leading the way in building a leading developer ecosystem by supporting the adoption of AI technologies by companies, such as through the AI Innovator Program. Sahir Azam, CTO, said, “MongoDB has the broadest coverage and maintains partnerships with all major cloud companies.”

■ Samsung Electronics Smart Home Migration Case Introduction At the event that day, Samsung Electronics presented its adoption of MongoDB for IoT smart home data processing. Samsung Electronics DA Business Division's Pro Seongbin Lim presented a case study of migration to MongoDB Atlas.

According to Samsung Electronics, the number of smart home devices registered in 2023 will rapidly increase to 16 million, resulting in increased traffic and issues such as the need for specialized personnel and technical support.

Samsung Electronics' smart home platform SmartThings is a service that controls home appliances, lighting, security, etc. by connecting them through an application, and announced that it has recently introduced MongoDB to control and manage the increasing traffic.

Samsung Electronics first introduced MongoDB in 2015, introduced the Atlas platform in 2022, and performed a shard transition in 2023.

'Database shard' refers to the horizontal division of data in a database. In the existing monolithic architecture, there is a limit to the capacity for scaling, but if it is built in a distributed manner, there is an advantage in that HW can be used horizontally at a much lower cost. With the sharding operation, you can quickly find the information you want through queries in the distributed data. This is explained as increasing scalability and reducing costs.

“Through Atlas migration, the average response speed was reduced from 8ms to 3ms, and the latency, which indicates availability, was also reduced from a maximum of 3s to a maximum of 18ms,” said Pro Lim Seong-bin. He also announced that the number of outage issues was reduced from 19 to 0 due to the shard strategy introduced to ensure the summer traffic surge.