---
title: Thread pool
description: The following topic describes the MBeans tab and the recommended number of threads in the pool.
component: pingfederate
version: 13.1
page_id: pingfederate:pingfederate_monitoring_guide:pf_thread_pool
canonical_url: https://docs.pingidentity.com/pingfederate/13.1/pingfederate_monitoring_guide/pf_thread_pool.html
llms_txt: https://docs.pingidentity.com/pingfederate/llms.txt
docs_for_agents: https://developer.pingidentity.com/build-with-ai/docs-for-agents.md
revdate: July 5, 2022
---

# Thread pool

The following topic describes the MBeans tab and the recommended number of threads in the pool.

The **MBeans** tab provides access to the JMX Managed Beans and their available attributes and operations. Of particular interest are the `queuedthreadpool` instances that are available within the `org.eclipse.jetty.util.thread` bean. For example, instance 0 represents the thread pool that handles runtime requests. When you click the **Attributes** item for instance 0, the current state of the thread pool is displayed.

![Screenshot of the MBeans tab showing the current state of the thread pool.](_images/vcd1580332503811.png)

The number of threads in the pool (the `threads` attribute) can be compared to the number of threads that are not currently in use (the `idleThreads` attribute). Ideally, a sufficient number of threads is available to handle load spikes that occur during peak usage times, while also limiting the number of idle threads that are running. If the thread pool is too small, requests might be blocked. If the thread pool is too large, memory might be used unnecessarily, and CPU contention might increase, limiting the processing effectiveness.

We recommend allowing for 10 to 25% more threads than are typically active while the system executes a normal or expected load. Because most of your users will not be active at the same time, set the minimum threads to 10% above the average number of active threads that are observed during monitoring.

We also recommend setting the maximum number of threads to 25% above the average number of active threads that are observed while monitoring under expected load conditions. Make certain to weigh this recommendation against the observed CPU utilization metrics and the suggestions in the [About Performance Tuning](../performance_tuning_guide/pf_about_performance_tuning.html).