Resource Specification

This topic introduces specification of standard resource instances that can be managed online and the available options for resource items.

Stream Analytics Computing Resource

Multiple resource specifications are available to meet the requirements of different data amounts and processing efficiency. Resource specifications and corresponding data processing capabilities are as follows:

Specification Computing Capability Reference
Standard 1,500 points / second With 10 measuring points / second, supporting about 150 devices
Standard X 2 3,000 points / second With 10 measuring points / second, supporting about 300 devices
Standard X 4 6,000 points / second With 10 measuring points / second, supporting about 600 devices

Note

The computing capability refers to the total amount of data that can be processed by running stream processing jobs in unit time. For a stream processing job, the number of data points is equal to the number of measure points multiplied by the number of devices.

TSDB Resource

The TSDB resources include write resource and storage resource. For write resource, only standard specification can be requested. Storage resource can be requested based on actual business needs. The system will measure and charge the actual use of resources. Detailed specification is as follows:

Storage Capacity Description
Write Resource 5,000 points / second
Storage Resource Available options are 10~1000 GB

Batch Computing Resource

Batch computing resources can be requested based on computing unit (CU), with two kinds of computing specifications, Computing-Intensive and Memory-Intensive. You can request appropriate computing specification according to the characteristics of offline data analytics jobs (1 ~ 20 CU).

Computing Specification Allocated Resources
Computing-Intensive 1CU = 1 core CPU + 2GB Memory
Memory-Intensive 1CU = 1 core CPU + 4GB Memory

Data Warehouse Resource

Data warehouse (EnOS Hive) is a subject-oriented and integrated data storage environment for data analysis. Select the appropriate storage size according to actual business requirements. Detailed specification is as follows:

Storage Capacity Description
Storage Resource Available options are 10~1000 GB

Data Explorer Resource

Data Explorer sandbox provides running environment for the data analytics applications developed by data scientists and data engineers. Select the appropriate CPU, memory, and storage according to actual business requirements. Detailed specification is as follows:

Computing Capacity Description
CPU Available options are 2~10 vcore
Memory Available options are 4~20 GB
Storage Resource Available options are 1~10 GB

HDFS File Storage Resource

HDFS is used for big data analysis and storage scenarios (including data archiving). Select the appropriate storage size according to actual business requirements. Detailed specification is as follows:

Storage Capacity Description
Storage Resource Available options are 10~1000 GB

Data Archiving Resource

Multiple resource specifications are available for archiving different data amount and data archiving efficiency. Resource specifications and corresponding data archiving capabilities are as follows:

Specification Archiving Capability
Standard 5,000 records / second
Standard X 2 10,000 records / second
Standard X 4 20,000 records / second
Standard X 8 40,000 records / second