Versions Compared

Old Version 4

changes.mady.by.user Ankit Gandhi

Saved on

compared with

New Version Current

changes.mady.by.user Nitin Beri

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Corrected links that should have been relative instead of absolute.

Adeptia checks available After user has enabled Resource-Based-Round-Robin allocation of jobs, Adeptia would check for the available resources on every node of the cluster before allocating a job in round-robin method. In other words, Adeptia system will first check the Whereas, in a normal round-robin method, system allocates the job to the next node in sequence without checking for availability of the resources on every node before randomly allocating the job to the relevant node. If that node. After enabling this feature, if resources are not available on that node, it will move to the next node and check again before without allocating the job (after logging a notification in kernel logs file), and check the next node in the sequence for allocating the job. To enable resource-based allocation in Clustering:

  1. Go to ...\AdeptiaSuite-6.9\AdeptiaServer\ServerKernel\etc\cluster.
  2. Open cluster properties file.
  3. Change the value of cluster.load.balance.policy from Round-Robin to to Administer > Setup> Application Settings
  4. From Application Settings, go to Update System Properties > Load Management > Cluster and Queue Processor Configuration.
  5. Write Resource-Based-Round-Robin. 
    That is, change clusterRobin in the value box for cluster.load.balance.policy=Round-Robin to
    cluster.load.balance.policy=Resource-Based-Round-Robin
     Image RemovedSave the file property.

    Image Added

  6. Click Save.
  7. Login to Adeptia Suite.
  8. Restart Kernel and & WebRunner.

After you have enabled resource-based allocation of jobs in the Cluster Properties file, you should give threshold value of each of the resource. Note that these values should be same for all the nodes of a cluster. To set the values:


  2. Go toIn case there is no resource available in any of the node of the cluster, then system will wait for the time set in abpm.node.resourceavailability.check.waiting.period property. Users can view the notification of unsuccessful allocation of the job to a node in the kernel logs file.

  3. To set the threshold and waiting time go to Administer > Setup> Application Settings.
  4. From Application Settings, go to to Update System Properties > Load Management > System Throttling.
  5. In Load management set the threshold in relevant property.
    Image Removed
  6. Click Save.

...

  1. Image Added