Table of Contents

Search

  1. Preface
  2. Analyst Service
  3. Content Management Service
  4. Data Integration Service
  5. Data Integration Service Architecture
  6. Data Integration Service Management
  7. Data Integration Service Grid
  8. Data Integration Service Applications
  9. Mass Ingestion Service
  10. Metadata Access Service
  11. Metadata Manager Service
  12. Model Repository Service
  13. PowerCenter Integration Service
  14. PowerCenter Integration Service Architecture
  15. High Availability for the PowerCenter Integration Service
  16. PowerCenter Repository Service
  17. PowerCenter Repository Management
  18. PowerExchange Listener Service
  19. PowerExchange Logger Service
  20. SAP BW Service
  21. Search Service
  22. System Services
  23. Test Data Manager Service
  24. Test Data Warehouse Service
  25. Web Services Hub
  26. Application Service Upgrade
  27. Application Service Databases
  28. Connecting to Databases from Windows
  29. Connecting to Databases from UNIX
  30. Updating the DynamicSections Parameter of a DB2 Database

Maximum Number of Concurrent Jobs on a Grid

Maximum Number of Concurrent Jobs on a Grid

You can set the number of deployed and on-demand jobs that the Data Integration Service process can run concurrently.
You can configure the following Data Integration Service properties:
Maximum Hadoop batch pool size
Dtermines the maximum number of deployed jobs that you can run concurrently in the Hadoop environment. The Data Integration Service moves deployed Hadoop jobs from the queue to the Hadoop batch pool when enough resources are available. The default value is 100.
Maximum native batch pool size
Determines the maximum number of jobs that you can run concurrently in the native environment. The Data Integration Service moves deployed native jobs from the queue to the native batch pool when enough resources are available. The default value is 10.
Maximum on-demand pool size
Determine the maximum number of on-demand jobs that you can run concurrently. On-demand jobs include data previews, profiling jobs, SQL queries, and web service requests. The Data Integration Service immediately runs on-demand jobs if enough resources are available. Otherwise, the Data Integration Service rejects the job. The default value is 10.
When the Data Integration Service runs on a grid, the maximum number of deployed and on-demand jobs that can run concurrently across the grid are calculated as follows:
Maximum Hadoop batch pool size * Number of running service processes
Maximum native batch pool size * Number of running service processes
Maximum on-demand pool size * Number of running service processes
For example, a Data Integration Service grid includes three running service processes. If you set the Hadoop batch pool size to 10, each Data Integration Service process can run up to 10 deployed Hadoop jobs concurrently. A total of 30 deployed Hadoop jobs can run concurrently on the grid. If you try to run more than 30 Hadoop jobs, the Data Integration Service queues the jobs until there is space in the pool.
When you increase the pool size values, the Data Integration Service uses more hardware resources such as CPU, memory, and system I/O. Set these values based on the resources available on the nodes in the grid. For example, consider the number of CPUs on the machines where Data Integration Service processes run and the amount of memory that is available to the Data Integration Service.
If the Data Integration Service grid runs jobs in separate remote processes, additional concurrent jobs might not run on compute nodes after you increase the value of these properties. You might need to override compute node attributes to increase the number of concurrent jobs on each compute node. For more information, see Override Compute Node Attributes to Increase Concurrent Jobs.