Jump to: navigation, search

Difference between revisions of "Rally/BenchmarkScenarios"

m (Configuration example)
m (Configuration example)
Line 42: Line 42:
 
     "NovaServers.boot_and_delete_server": [
 
     "NovaServers.boot_and_delete_server": [
 
         {
 
         {
            "args": {"flavor_id": 1, "image_id": "73257560-c59b-4275-a1ec-ab140e5b9979"},
+
            "args": {"flavor_id": 1, "image_id": "73257560-c59b-4275-a1ec-ab140e5b9979"},
            "execution": "continuous",
+
            "execution": "continuous",
            "config": {"duration": 30, "active_users": 10, "tenants": 3, "users_per_tenant": 5}
+
            "config": {"duration": 30, "active_users": 10, "tenants": 3, "users_per_tenant": 5}
 
         }
 
         }
 
     ]
 
     ]

Revision as of 15:04, 5 December 2013

Available Benchmark Scenarios

Nova

NovaServers

This scenario class contains procedures concentrated on calling the Nova servers API.


boot_server

Boots one server.

Configuration example
   "NovaServers.boot_server": [
       {
           "args": {"flavor_id": 1, "image_id": "73257560-c59b-4275-a1ec-ab140e5b9979"},
           "execution": "continuous",
           "config": {"times": 8, "active_users": 4, "tenants": 2, "users_per_tenant": 2}
       }
   ]

boot_and_bounce_server

Boots one server and then performs N "power operations" on it. The supported power operation actions are:

  • stop_start -- Stop the VM and once stopped, start it and wait for it to become active.
  • soft_reboot -- Perform a soft reboot of the VM and wait for it to come back up into the ACTIVE state.
  • hard_reboot -- Perform a hard reboot of the VM and wait for it to come back into the ACTIVE state.

Note that the configuration format for this benchmark supports a list of power operations. So for example you could specify to stop_start the VM 5 times, followed by soft_boot it 2 times, followed by... These operations are defined in the 'actions' argument which is a list/array.

Configuration example
   "NovaServers.boot_and_bounce_server": [
       {
           "args": {
               "flavor_id": 1,
               "image_id": "73257560-c59b-4275-a1ec-ab140e5b9979",
               "actions" : [{"soft_reboot": 4}, {"stop_start": 2}]
           },
           "execution": "continuous",
           "config": {"times": 31, "active_users": 8}
       }
   ]

boot_and_delete_server

Boots and deletes one server.

Configuration example
   "NovaServers.boot_and_delete_server": [
       {
           "args": {"flavor_id": 1, "image_id": "73257560-c59b-4275-a1ec-ab140e5b9979"},
           "execution": "continuous",
           "config": {"duration": 30, "active_users": 10, "tenants": 3, "users_per_tenant": 5}
       }
   ]

snapshot_server

Boots a server, makes its snapshot, then deletes the server and tries to boot it again from the saved snapshot. Finally deletes both the server and its snapshot.

Configuration example
   "NovaServers.snapshot_server": [
       {
           "args": {"flavor_id": 1, "image_id": "73257560-c59b-4275-a1ec-ab140e5b9979"},
           "execution": "continuous",
           "config": {"times": 50, "active_users": 10, "tenants": 3, "users_per_tenant": 5}
       }
   ]

Benchmark Scenario Wishlist

This section contains a listing of benchmark scenario candidates to add to rally. Other community members can suggest scenarios here with the intent that the suggested scenario(s) provide interesting/useful results.

Wishlist

  • Boot VM and suspend/resume N times -- test VM suspend/resume under load
  • Boot a VM
  • Suspend and resume the VM N times
  • Volume create and destroy -- tests volume creation/deletion under load
  • Create and destroy a volume N time
  • Extend volume N times -- test volume extend under load
  • Create a volume and extend it N times