AI 용어집
인공지능 완전 사전
NoSQL
Category of database management systems that do not use the traditional SQL relational model, designed to handle large volumes of distributed data with high performance and schema flexibility.
Document-oriented database
Type of NoSQL that stores data in document formats similar to JSON or XML, allowing flexible hierarchical structure and complex queries on document contents.
Column-oriented database
NoSQL system optimized for massive data reads and writes, storing data by columns rather than rows to improve analysis performance and compression efficiency.
Graph-oriented database
NoSQL database using graph structures with nodes, edges, and properties to represent and store data, optimized for queries on relationships and complex connections.
Key-value database
Simplest NoSQL database model using key-value association for storage and retrieval, offering extremely fast performance for read and write operations.
Horizontal scalability
Ability of a NoSQL system to increase its processing power by adding additional servers to the cluster, allowing linear growth in performance and storage capacity.
CAP Theorem
Fundamental principle in distributed systems stating that a database can only simultaneously guarantee two of the three properties: Consistency, Availability, and Partition Tolerance.
Eventual Consistency
Consistency model in distributed systems where updates are gradually propagated throughout the system, ensuring that all copies will eventually converge to the same state if no new updates are made.
Sharding
Horizontal partitioning technique in NoSQL databases where data is distributed across multiple servers based on a partition key, enabling better performance and greater storage capacity.
Schema-less
Characteristic of NoSQL databases that do not impose a predefined schema, allowing maximum flexibility in data structure and rapid evolution of data models without complex migrations.
MongoDB
Open-source document-oriented NoSQL database system using BSON (Binary JSON) documents with full support for queries, indexing, and aggregation for modern web applications.
Cassandra
Distributed column-oriented NoSQL database designed to handle massive amounts of data on standard servers, offering high availability without single points of failure and excellent scalability.
Redis
Ultra-fast open-source in-memory key-value database supporting various data structures, used as cache, message broker, and primary database for applications requiring extreme performance.
Neo4j
Native ACID-compliant graph database platform optimized for complex graph traversal queries, using the Cypher query language to interact with relational data.
Couchbase
Distributed document-oriented NoSQL database combining JSON flexibility with in-memory performance, offering SQL-like queries via N1QL and specialized subsystems for different use cases.
DynamoDB
Fully managed NoSQL database service by AWS offering predictable performance and seamless scalability, using a key-value and document-oriented model with provisioning and on-demand capabilities.
HBase
Distributed column-oriented NoSQL database built on Hadoop, optimized for storing and fast accessing terabytes of structured data with strong consistency and real-time random read/write.
Riak
Distributed key-value NoSQL database designed for high availability and extreme fault tolerance, using eventual consistency and replication to ensure data persistence in critical environments.
CouchDB
Master-master synchronizable document-oriented NoSQL database using HTTP protocol and JSON format, designed to facilitate replication and work efficiently in offline-first mode.
MapReduce
Programming model for distributed processing of large datasets, used by NoSQL databases to execute aggregation and analysis operations across the cluster in a parallel and scalable manner.