Skip to content
logo
Percona Monitoring and Management
NUMA Details
Initializing search
    percona/pmm-doc
    percona/pmm-doc
    • Welcome
        • Network
        • Docker
        • Podman
        • Helm
        • Virtual Appliance
        • AWS Marketplace
        • Easy-install script
        • MySQL
        • MongoDB
        • PostgreSQL
        • ProxySQL
        • Amazon RDS
        • Microsoft Azure
        • Google Cloud Platform
        • Linux
        • External Services
        • HAProxy
        • Remote instances
      • User Interface
      • Percona Alerting
        • Prepare a storage location
          • MongoDB backup prerequisites
          • Create MongoDB on-demand and scheduled backups
          • Create MongoDB PITR backups
          • Restore a MongoDB backup
          • MongoDB Backup and Restore support matrix
          • MySQL backup prerequisites
          • Create a MySQL backup
          • Restore a MySQL backup
        • Edit a scheduled backup
        • Delete a backup
        • Configure access control
        • Labels for access control
        • Create access roles
        • Manage access roles
        • Assign roles to users
        • Use Case
      • Query Analytics
      • Working with Advisors
      • Configure
      • Manage users
      • Upgrade
      • Secure
      • Optimize
      • Annotate
      • Share dashboards and panels
      • Extend Metrics
      • Resolve issues
        • Check Percona Portal account information
      • Architecture
      • UI components
      • PMM components and versions
      • Data handling in PMM
      • Develop Advisor checks
      • PMM Inventory
        • Manage dashboards
            • Advanced Data Exploration
            • VictoriaMetrics
            • VictoriaMetrics Agents Overview
            • Environment Summary
            • Kubernetes Pods Status
            • Kubernetes Volumes
          • DBaas
            • Disk Details
            • Network Details
            • Memory Details
            • Node Temperature Details
            • Nodes Compare
            • Nodes Overview
            • Node Summary
            • NUMA Details
              • Memory Usage
              • Free Memory Percent
              • NUMA Memory Usage Types
              • NUMA Allocation Hits
              • NUMA Allocation Missed
              • Anonymous Memory
              • NUMA File (PageCache)
              • Shared Memory
              • HugePages Statistics
              • Local Processes
              • Remote Processes
              • Slab Memory
            • Processes Details
            • Prometheus Exporters Overview
            • MySQL Command/Handler Counters Compare
            • MySQL InnoDB Compression Details
            • MySQL InnoDB Details
            • MySQL MyISAM/Aria Details
            • MySQL MyRocks Details
            • MySQL Instance Summary
            • MySQL Instances Compare
            • MySQL Instances Overview
            • MySQL Wait Event Analyses Details
            • MySQL Performance Schema Details
            • MySQL Query Response Time Details
            • MySQL Replication Summary
            • MySQL Group Replication Summary
            • MySQL Table Details
            • MySQL User Details
            • MySQL TokuDB Details
            • Experimental MongoDB Collection Details
            • Experimental MongoDB Oplog Details
            • MongoDB Cluster Summary
            • MongoDB Instance Summary
            • MongoDB Instances Overview
            • MongoDB Instances Compare
            • MongoDB ReplSet Summary
            • MongoDB InMemory Details
            • MongoDB MMAPv1 Details
            • MongoDB WiredTiger Details
            • PostgreSQL Instance Summary
            • PostgreSQL Instances Compare
            • Experimental PostgreSQL Vacuum Monitoring
          • ProxySQL Dashboards
            • PXC/Galera Cluster Summary
            • Experimental PXC/Galera Cluster Summary
            • PXC/Galera Nodes Compare
            • HAProxy Instance Summary
        • pmm-admin - PMM Administration Tool
        • pmm-agent - PMM Client agent
      • API
      • VictoriaMetrics
      • ClickHouse
      • PostgreSQL
      • Glossary
      • Introduction
      • DBaaS architecture
        • Setting up DBaaS
        • Create a Kubernetes Cluster
        • Deleting Kubernetes clusters
        • Activating DBaaS
        • Add a Kubernetes cluster automatically
        • Add a Kubernetes cluster manually
        • Manage allowed component versions
          • OLM installation
          • Operators installation
        • Add a DB Cluster
        • Manage a DB Cluster
        • Delete a DB Cluster
        • Create a database cluster from a template
      • Backup and restore
    • FAQ
      • PMM 2.39.0
      • PMM 2.38.1
      • PMM 2.38.0
      • PMM 2.37.1
      • PMM 2.37.0
      • PMM 2.36.0
      • PMM 2.35.0
      • PMM 2.34.0
      • PMM 2.33.0
      • PMM 2.32.0
      • PMM 2.31.0
      • PMM 2.30.0
      • PMM 2.29.1
      • PMM 2.29.0
      • PMM 2.28.0
      • PMM 2.27.0
      • PMM 2.26.0
      • PMM 2.25.0
      • PMM 2.24.0
      • PMM 2.23.0
      • PMM 2.22.0
      • PMM 2.21.0
      • PMM 2.20.0
      • PMM 2.19.0
      • PMM 2.18.0
      • PMM 2.17.0
      • PMM 2.16.0
      • PMM 2.15.1
      • PMM 2.15.0
      • PMM 2.14.0
      • PMM 2.13.0
      • PMM 2.12.0
      • PMM 2.11.1
      • PMM 2.11.0
      • PMM 2.10.1
      • PMM 2.10.0
      • PMM 2.9.1
      • PMM 2.9.0
      • PMM 2.8.0
      • PMM 2.7.0
      • PMM 2.6.1
      • PMM 2.6.0
      • PMM 2.5.0
      • PMM 2.4.0
      • PMM 2.3.0
      • PMM 2.2.2
      • PMM 2.2.1
      • PMM 2.2.0
      • PMM 2.1.0
      • PMM 2.0.1
      • PMM 2.0.0
      • Copyright and licensing information
      • Trademark policy

    • Memory Usage
    • Free Memory Percent
    • NUMA Memory Usage Types
    • NUMA Allocation Hits
    • NUMA Allocation Missed
    • Anonymous Memory
    • NUMA File (PageCache)
    • Shared Memory
    • HugePages Statistics
    • Local Processes
    • Remote Processes
    • Slab Memory

    NUMA Details¶

    !image

    For each node, this dashboard shows metrics related to Non-uniform memory access (NUMA).

    Memory Usage¶

    Remotes over time the total, used, and free memory.

    Free Memory Percent¶

    Shows the free memory as the ratio to the total available memory.

    NUMA Memory Usage Types¶

    Dirty
    Memory waiting to be written back to disk
    Bounce
    Memory used for block device bounce buffers
    Mapped
    Files which have been mapped, such as libraries
    KernelStack
    The memory the kernel stack uses. This is not reclaimable.

    NUMA Allocation Hits¶

    Memory successfully allocated on this node as intended.

    NUMA Allocation Missed¶

    Memory missed is allocated on a node despite the process preferring some different node.

    Memory foreign is intended for a node, but actually allocated on some different node.

    Anonymous Memory¶

    Active
    Anonymous memory that has been used more recently and usually not swapped out.
    Inactive
    Anonymous memory that has not been used recently and can be swapped out.

    NUMA File (PageCache)¶

    Active(file) Pagecache memory that has been used more recently and usually not reclaimed until needed.

    Inactive(file) Pagecache memory that can be reclaimed without huge performance impact.

    Shared Memory¶

    Shmem Total used shared memory (shared between several processes, thus including RAM disks, SYS-V-IPC and BSD like SHMEM).

    HugePages Statistics¶

    Total
    Number of hugepages being allocated by the kernel (Defined with vm.nr_hugepages).
    Free
    The number of hugepages not being allocated by a process
    Surp
    The number of hugepages in the pool above the value in vm.nr_hugepages. The maximum number of surplus hugepages is controlled by vm.nr_overcommit_hugepages.

    Local Processes¶

    Memory allocated on a node while a process was running on it.

    Remote Processes¶

    Memory allocated on a node while a process was running on some other node.

    Slab Memory¶

    Slab
    Allocation is a memory management mechanism intended for the efficient memory allocation of kernel objects.
    SReclaimable
    The part of the Slab that might be reclaimed (such as caches).
    SUnreclaim
    The part of the Slab that can’t be reclaimed under memory pressure

    Contact us

    For free technical help, visit the Percona Community Forum.

    To report bugs or submit feature requests, open a JIRA ticket.

    For paid support and managed or consulting services , contact Percona Sales.


    Last update: 2023-09-22
    Percona LLC, © 2023
    Made with Material for MkDocs

    Cookie consent

    We use cookies to recognize your repeated visits and preferences, as well as to measure the effectiveness of our documentation and whether users find what they're searching for. With your consent, you're helping us to make our documentation better. Read more about Percona Cookie Policy.