2017-04-11 21:52:47 -04:00
|
|
|
//lcawley Verified example output 2017-04-11
|
2017-04-04 18:26:39 -04:00
|
|
|
[[ml-snapshot-resource]]
|
|
|
|
==== Model Snapshot Resources
|
|
|
|
|
|
|
|
////
|
2017-04-11 16:25:38 -04:00
|
|
|
Model snapshots are saved to disk periodically.
|
|
|
|
By default, this is occurs approximately every 3 hours.
|
|
|
|
//TBD: Can you change this setting?
|
|
|
|
|
|
|
|
By default, model snapshots are retained for one day. You can change this
|
|
|
|
behavior with by updating the `model_snapshot_retention_days` for the job.
|
|
|
|
When choosing a new value, consider the following:
|
|
|
|
|
|
|
|
* Persistence enables resilience in the event of a system failure.
|
|
|
|
* Persistence allows for snapshots to be reverted.
|
|
|
|
* The time taken to persist a job is proportional to the size of the model in memory.
|
|
|
|
//* The smallest allowed value is 3600 (1 hour).
|
2017-04-04 18:26:39 -04:00
|
|
|
////
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
A model snapshot resource has the following properties:
|
|
|
|
|
|
|
|
`description`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(string) An optional description of the job.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`job_id`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(string) A numerical character string that uniquely identifies the job.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`latest_record_time_stamp`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 1455232663000.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`latest_result_time_stamp`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 1455229800000.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`model_size_stats`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(object) TBD. See <<ml-snapshot-stats,Model Size Statistics>>.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`retain`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(boolean) TBD. For example: false.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`snapshot_id`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(string) A numerical character string that uniquely identifies the model
|
2017-04-11 16:25:38 -04:00
|
|
|
snapshot. For example: "1491852978".
|
|
|
|
|
|
|
|
`snapshot_doc_count`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 1.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`timestamp`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(date) The creation timestamp for the snapshot, specified in ISO 8601 format.
|
2017-04-11 16:25:38 -04:00
|
|
|
For example: 1491852978000.
|
|
|
|
|
2017-04-11 21:52:47 -04:00
|
|
|
[float]
|
|
|
|
[[ml-snapshot-stats]]
|
2017-04-11 16:25:38 -04:00
|
|
|
===== Model Size Statistics
|
|
|
|
|
|
|
|
The `model_size_stats` object has the following properties:
|
|
|
|
|
|
|
|
`bucket_allocation_failures_count`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 0.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`job_id`::
|
2017-04-11 22:26:18 -04:00
|
|
|
(string) A numerical character string that uniquely identifies the job.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`log_time`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 1491852978000.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`memory_status`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: "ok".
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`model_bytes`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 100393.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`result_type`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: "model_size_stats".
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`timestamp`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 1455229800000.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`total_by_field_count`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 13.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`total_over_field_count`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 0.
|
2017-04-11 16:25:38 -04:00
|
|
|
|
|
|
|
`total_partition_field_count`::
|
2017-04-11 22:26:18 -04:00
|
|
|
() TBD. For example: 2.
|