Measures reported by ESThreadPoolTest
This test auto-discovers the thread pools in the Elasticsearch cluster, and sheds light on the load and thread pool usage. Using the metrics reported by this test, administrators can take crucial decisions on sizing the thread pools corresponding to the current and expected request load.
Output of the test: One set of results for each thread pool in the target Elasticsearch cluster.
The measures made by this test are as follows:
| Measurement |
Description |
Measurement Unit |
Interpretation |
| ThreadPool_Util |
Indicates the percentage of threads in this thread pool that are currently in use. |
Percent |
This measure is computed using the following formula:
(ThreadPool_Active / ThreadPool_Threads)X100
A high value is indicative of a busy pool. A value close to 100% indicates excessive utilization of threads in a pool. If the value of this measure grows closer to 100% over time, it indicates that the pool is rapidly running out of threads to service the search requests. In such a case, administrators may then have to fine-tune the thread pool size. |
| ThreadPool_Active |
Indicates the number of threads in this pool that are currently active in servicing the requests. |
Number |
This measure is a good indicator of cluster workload. |
| ThreadPool_Queue |
Indicates the number of threads queued for processing in this pool. |
Number |
|
| ThreadPool_Threads |
Indicates the maximum number of active threads that this pool can contain. |
Number |
|
| ThreadPool_Rejected |
Indicates the number of threads that were rejected from this pool. |
Number |
|
|