Machine learning uses SSE4.2 instructions, so will only work on machines whose
CPUs support SSE4.2. If you
run Elasticsearch on older hardware you must disable machine learning (by setting
All of these settings can be added to the
elasticsearch.yml configuration file.
The dynamic settings can also be updated across a cluster with the
cluster update settings API.
Dynamic settings take precedence over settings in the
General machine learning settingsedit
[7.9.0] Deprecated in 7.9.0. Use node.roles instead. Set to
true(default) to identify the node as a machine learning node.
If set to
elasticsearch.yml, the node cannot run jobs. If set to
xpack.ml.enabledis set to
node.mlsetting is ignored and the node cannot run jobs. If you want to run jobs, there must be at least one machine learning node in your cluster.
On dedicated coordinating nodes or dedicated master nodes, disable the
true(default) to enable machine learning on the node.
If set to
false, the machine learning APIs are disabled on the node. Therefore the node cannot open jobs, start datafeeds, or receive transport (internal) communication requests related to machine learning APIs. If the node is a coordinating node, machine learning requests from clients (including Kibana) also fail. For more information about disabling machine learning in specific Kibana instances, see Kibana machine learning settings.
If you want to use machine learning features in your cluster, it is recommended that you set
trueon all nodes. This is the default behavior. At a minimum, it must be enabled on all master-eligible nodes. If you want to use machine learning features in clients or Kibana, it must also be enabled on all coordinating nodes.
The maximum inference cache size allowed. The inference cache exists in the JVM
heap on each ingest node. The cache affords faster processing times for the
inferenceprocessor. The value can be a static byte sized value (i.e. "2gb") or a percentage of total allocated heap. The default is "40%". See also Machine learning circuit breaker settings.
The time to live (TTL) for models in the inference model cache. The TTL is
calculated from last access. The
inferenceprocessor attempts to load the model from cache. If the
inferenceprocessor does not receive any documents for the duration of the TTL, the referenced model is flagged for eviction from the cache. If a document is processed later, the model is again loaded into the cache. Defaults to
The total number of
inferencetype processors allowed across all ingest pipelines. Once the limit is reached, adding an
inferenceprocessor to a pipeline is disallowed. Defaults to
The maximum percentage of the machine’s memory that machine learning may use for running
analytics processes. (These processes are separate to the Elasticsearch JVM.) Defaults to
30percent. The limit is based on the total memory of the machine, not current free memory. Jobs will not be allocated to a node if doing so would cause the estimated memory use of machine learning jobs to exceed the limit.
model_memory_limitproperty value that can be set for any job on this node. If you try to create a job with a
model_memory_limitproperty value that is greater than this setting value, an error occurs. Existing jobs are not affected when you update this setting. For more information about the
model_memory_limitproperty, see `analysis_limits`.
The maximum number of jobs that can run simultaneously on a node. Defaults to
20. In this context, jobs include both anomaly detection jobs and data frame analytics jobs. The maximum number of jobs is also constrained by memory usage. Thus if the estimated memory usage of the jobs would be higher than allowed, fewer jobs will run on a node. Prior to version 7.1, this setting was a per-node non-dynamic setting. It became a cluster-wide dynamic setting in version 7.1. As a result, changes to its value after node startup are used only after every node in the cluster is running version 7.1 or higher. The maximum permitted value is
The maximum number of jobs that can concurrently be in the
openingstate on each node. Typically, jobs spend a small amount of time in this state before they move to
openstate. Jobs that must restore large models when they are opening spend more time in the
openingstate. Defaults to
Advanced machine learning settingsedit
These settings are for advanced use cases; the default values are generally sufficient:
The maximum number of records that are output per bucket. The default value is
The number of lazily spun up Machine Learning nodes. Useful in situations where ML nodes are not desired until the first Machine Learning Job is opened. It defaults to
0and has a maximum acceptable value of
3. If the current number of ML nodes is
>=than this setting, then it is assumed that there are no more lazy nodes available as the desired number of nodes have already been provisioned. When a job is opened with this setting set at
>0and there are no nodes that can accept the job, then the job will stay in the
OPENINGstate until a new ML node is added to the cluster and the job is assigned to run on that node.
This setting assumes some external process is capable of adding ML nodes to the cluster. This setting is only useful when used in conjunction with such an external process.
The connection timeout for machine learning processes that run separately from the Elasticsearch JVM.
10s. Some machine learning processing is done by processes that run separately to the Elasticsearch JVM. When such processes are started they must connect to the Elasticsearch JVM. If such a process does not connect within the time period specified by this setting then the process is assumed to have failed. Defaults to
10s. The minimum value for this setting is
Machine learning circuit breaker settingsedit
- Limit for the model inference breaker, which defaults to 50% of the JVM heap. If the parent circuit breaker is less than 50% of the JVM heap, it is bound to that limit instead. See Circuit breaker settings.
- A constant that all accounting estimations are multiplied by to determine a final estimation. Defaults to 1. See Circuit breaker settings.
The underlying type of the circuit breaker. There are two valid options:
noopmeans the circuit breaker does nothing to prevent too much memory usage.
memorymeans the circuit breaker tracks the memory used by inference models and can potentially break and prevent OutOfMemory errors. The default is