Application Scenarios

Migration of Big Data Services and Platforms to the Cloud

Scenario: You can smoothly migrate customers' on-premises/cloud big data platforms to CloudTable to quickly build their data service systems for rapid service growth.
Advantages
- Ease of use: Out-of-the-box availability frees you from dedicated deployment, reduces the O&M workload, and enables you to focus on service delivery, and facilitates big data application analysis.
- On-demand resource scalability: You can flexibly choose and scale compute and storage resources, adjust cluster nodes, specifications, and storage on demand.
- Open source compatibility: CloudTable is fully compatible with open source APIs, ensuring seamless service integration and zero code change.
- Superior performance: CloudTable experiences significantly faster real-time data queries compared to traditional data warehouses.
- Security and reliability: CloudTable enjoys independent cluster deployment, VPC network isolation, and encrypted data channels for top-tier security and reliability.
Figure 1 Migration of big data services and platforms to the cloud

Storage and Query of Message Logs

Scenario: CloudTable HBase is designed for the rapid querying of message and log data, efficiently delivering results to applications. Structured and semi-structured key-value data can be stored and queried, including messages, reports, recommendation data, risk control data, logs, and orders.
Advantages
- Mass storage: CloudTable offers robust support for both offline and online storage, accommodating vast amounts of key-value data with a scalable storage framework.
- High-performance read/write: CloudTable delivers tens of millions in write throughput and millisecond-level response time for queries, ideal for powering online applications and generating instantaneous report displays.
- Vibrant ecosystem: Based on a wide variety of components of the Hadoop ecosystem, CloudTable can enhance its capabilities through integration with cloud products.
Figure 2 Storage and query of message logs

Real-time Report Query and Analysis

Scenario: CloudTable Doris efficiently aggregates data from source service systems via real-time and batch processing. This enables downstream applications to conduct comprehensive multi-dimensional analyses, yielding report results in subseconds.
Advantages
- Real-time data write: Capable of collecting data from multiple data sources, CloudTable excels in writing millions of data lines per second, ensuring a continuous flow of real-time information.
- Subsecond-level query response: CloudTable distributes queries across different buckets for processing. It uses point query indexes (primary keys and inverted indexes) to reduce the amount of data to be read and to provide concurrent query. It uses materialized views and pre-aggregated results to provide sub-second aggregation and statistics analysis.
- High stability and availability: With a resilient infrastructure of multi-copy and multi-node cluster deployment, CloudTable offers flexible capacity scaling and automated load balancing to maintain optimal performance.
Figure 3 Real-time report query and analysis

User Behavior Analysis

Scenario: CloudTable builds a flat-wide table of user behavioral data. It consolidates diverse datasets, including web page visits, application usage patterns, and detailed user activity logs, into CloudTable ClickHouse in real-time or in batches. It swiftly processes over ten billion data entries in mere seconds. This rapid analysis empowers businesses to construct intricate user profiles and execute targeted, precision marketing strategies with unparalleled efficiency.
Advantages
- High performance: CloudTable leverages advanced kernel designs, including column vector engine technology, data compression, and multi-core parallel processing, to boost query performance exponentially over traditional databases. It supports non-blocking write operations, enabling continuous updates to vast datasets. It delivers fast query responses in seconds, even for wide tables with thousands of columns and data records in the tens of billions.
- High availability: CloudTable is engineered for high availability with a ready-to-use, dual-copy architecture. It offers flexible service-oriented features, such as on-demand scaling both horizontally and vertically, to meet evolving business needs.
- Monitoring and awareness: CloudTable provides a suite of monitoring tools to track critical metrics like CPU usage, memory consumption, partition count, and slow SQL queries, ensuring real-time visibility of cluster health.
- Low costs: You are only billed for the duration during which functions are processing files or data, and the storage capacity you use. Auto scaling enables you to avoid resource redundancy during off-peak hours.
Figure 4 User behavior analysis