Static Service Resources
Overview
The resources allocated to each service in a cluster are static service resources. Such services include Flume, HBase, HDFS and Yarn. The total volume of computing resources allocated to each service is fixed, and they are static. A tenant can exclusively use or share a service to obtain the resources required for running this service.
Static Service Pool
Static service pools are used to specify service resource configurations.
Static service pools centrally manage resources applicable to each service.
- They restrict the total volume of resources used for services and dynamically configure the total CPU, I/O, and memory resources applicable to nodes running Flume, HBase, HDFS and Yarn.
- The resources of services are isolated. The services in a cluster are isolated from each other, and the workload of one service has limited impact on other services.
Scheduling Mechanism
The time-based dynamic resource scheduling mechanism enables different volumes of static resources to be configured for services at different time, optimizing service running environments and improving the cluster efficiency.
In a complex cluster environment, multiple services share resources in the cluster, but the resource service period of each service may be different.
The following use a bank customer as an example:
- The HBase query service is heavy in the daytime.
- The query service is light, but the Hive analysis service is heavy at night.
If fixed resources are allocated to each service, the following problems may occur:
- The query service cannot obtain sufficient resources while the resources for the analysis service are idle in the daytime.
- The analysis service cannot obtain sufficient resources while the resources for the query service are idle at night.
As a result, the cluster resource utilization is low and the service capability is weak. Resolve the problem in the following ways:
- Sufficient resources need to be configured for HBase in the daytime.
- Sufficient resources need to be configured for Hive at night.
The time-based dynamic scheduling mechanism can efficiently utilize resources and run tasks.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot