Configure secondary indexes

The secondary indexes of an Aerospike namespace can be stored in three different mediums: shared memory (SHMem) by default, Intel® Optane™ Persistent Memory (PMem), or Flash device (on NVMe SSDs). Separate namespaces within the same cluster can use different types of secondary index storage.

Note

Secondary index storage (sindex-type) is not configurable in Aerospike Database Community Edition (CE). Its secondary index storage is always volatile process memory, and the cluster node is not able to perform a fast restart.

To specify a secondary index (sindex) storage method, use the namespace configuration parameter sindex-type.

• The default sindex-type shmem stores secondary index metadata in shared memory segments.
• sindex-type pmem specifies PMem storage for namespace secondary indexes.
• sindex-type flash specifies NVMe SSD storage for namespace secondary indexes.

For sizing information, see secondary index capacity Planning.

Caution

On a systemd Linux OS you might need to increase TimeoutSec from the default of 15s. This setting is in /usr/lib/systemd/system/aerospike.service. This prevents systemd from killing the asd process prematurely while the service is being shutdown. The secondary index clean-up process during shutdown might take longer than the default 15s. In Aerospike, this default value has been increased to 10 minutes as of server version 4.6.0.2.

Secondary index in memory

By default the namespace secondary index storage type in Aerospike Database Enterprise Edition variants (EE, FE, SE) is shared memory, equivalent to explicitly setting sindex-type shmem. As of Database 6.1 the use of secondary indexes in memory is compatible with fast restarts.

Secondary index on Flash

When the namespace secondary index storage is configured to sindex-type flash, Aerospike writes index metadata in multiple files spread across the configured NVMe SSD devices. This requires device partitions to be set up with an appropriate filesystem and mounted.

Set kernel parameters

The following Linux kernel parameters are required in an index on Flash deployment. enforce-best-practices verifies that these kernel parameters have the expected values.

/proc/sys/vm/dirty_bytes = 16777216
/proc/sys/vm/dirty_background_bytes = 1
/proc/sys/vm/dirty_expire_centisecs = 1
/proc/sys/vm/dirty_writeback_centisecs = 10

• When running as non-root, you must prepare these values before running the Aerospike server.
• When running as root, the server configures them automatically.

Either way, if these parameters can’t be correctly set manually, or automatically by the server, the node will not be able to start up with a sindex-type flash configuration.

Caution

While it is advisable to adjust the kernel’s min_free_kbytes parameter in any configuration, it is especially important to do so when storing the primary index on Flash memory (All Flash). The Linux kernel will attempt to use all free space by caching disk writes. With All Flash configuration, this may result in an out of memory (OOM) kill if there isn’t enough free memory left for normal system operations. For this reason, Aerospike recommends setting min_free_kbytes=1153434 (1.1GB). For more information, see Best practices for Aerospike and Linux.

Prepare and mount Flash devices

Aerospike instantiates at least 4 different arena allocations (as files) for each device partition configured for use by the namespace. This helps reduce contention against the same arena, which improves performance during heavy insertion loads.

An XFS filesystem is recommended because it has been shown to provide better concurrent access to files than EXT4.

sudo mkfs.xfs -f /dev/nvme0
sudo mount /dev/nvme0 /mnt/nvme0

• Using more physical devices will improve performance through increased parallelism of disk IO.

• Assigning more partitions per physical device doesn’t necessarily improve performance.

Subcontext configuration

In the sindex-type subcontext define

• A mount point for each mounted Flash device. Mount points can be shared across multiple namespaces.

• A mounts-budget (or mounts-size-limit before Database 7.0) directive to indicate this namespace’s share of device storage space available across the given mount points.

  Ensure the budget is smaller or equal to the size of the filesystem. If mount points are not shared between namespaces, then simply specify the total available space.

• An optional eviction threshold as a percent of the budget can be defined through evict-mounts-pct (or mounts-high-water-pct before Database 7.0).

Example

Database 7.0 and later

namespace test {
    sindex-type flash {
        mount /mnt/nvme0
        mount /mnt/nvme1
        mount /mnt/nvme2
        mount /mnt/nvme3
        mounts-budget 1T
    }
}

Prior to Database 7.0

namespace test {
    sindex-type flash {
        mount /mnt/nvme0
        mount /mnt/nvme1
        mount /mnt/nvme2
        mount /mnt/nvme3
        mounts-size-limit 1T
    }
}

Secondary index in PMem

When the namespace secondary index storage is configured to sindex-type pmem, Aerospike writes index metadata in multiple files spread across the configured PMem devices. This requires device partitions to be set up with an appropriate filesystem and mounted.

Aerospike requires PMem to be accessible using DAX (Direct Access), that is, using block devices such as /dev/pmem0:

• The NVDIMM regions must be configured as AppDirect regions, as in the following example from a machine with a 750-GiB AppDirect region:

sudo ipmctl show -region
SocketID             ISetID    PersistentMemoryType  Capacity FreeCapacity HealthState
0 0x59727f4821b32ccc               AppDirect        750.0 GiB      0.0 GiB     Healthy

• The NVDIMM regions must be turned into fsdax namespaces, as in the following example from the same machine:

sudo ndctl list
[
  {
    "dev":"namespace0.0",
    "mode":"fsdax",
    "blockdev":"pmem0",
    ...
  }
]

Prepare and mount PMem devices

The PMem block device must contain a DAX-enabled filesystem, such as XFS or EXT4. On the machine in the above example, this could be accomplished by using mkfs:

Prepare an XFS filesystem

sudo mkfs.xfs -f -d su=2m,sw=1 /dev/pmem0

Prepare an EXT4 filesystem

sudo mkfs.ext4 /dev/pmem0

Mount the filesystem

Finally, the file system must be mounted with the dax mount option. The dax mount option is important. Without this option, the Linux page cache is involved in all I/O to and from persistent memory, which would drastically reduce performance.

In the following example, we use /mnt/pmem0 as the mount point.

sudo mount -o dax /dev/pmem0 /mnt/pmem0

Remember to make the mount persistent to survive system reboots by adding it to /etc/fstab. The mount point configuration line can be copied from /etc/mtab to /etc/fstab.

Mount point configuration

In the sindex-type subcontext define

• A mount point for each PMem device. Secondary index metadata will be evenly distributed across all of them. Mount points can be shared across multiple namespaces.

• A mounts-budget (or mounts-size-limit before Database 7.0) directive to indicate this namespace’s share of PMem storage space available across the given mount points.

  Ensure the budget is lower than or equal to the size of the filesystem. If mount points are not shared between namespaces, then simply specify the total available space.

• An optional eviction threshold as a percent of the budget can be defined through evict-mounts-pct (or mounts-high-water-pct before Database 7.0).

Example

The following configuration snippet extends the earlier example and makes all of /mnt/pmem0 memory (for example, 750 GiB) available to the namespace:

Database 7.0 and later

namespace test {
    sindex-type pmem {
        mount /mnt/pmem0
        mounts-budget 750G
    }
}

Prior to Database 7.0

namespace test {
    sindex-type pmem {
        mount /mnt/pmem0
        mounts-size-limit 750G
    }
}

Manage your secondary indexes

This section describes using asadm for managing secondary indexes.

For information on storage consumption of secondary indexes (memory, PMem or SSD, depending on the sindex-type) see the capacity planning guide.

Use the indexes-memory-budget configuration parameter to limit memory usage by all indexes (primary, set, secondary). If indexes-memory-budget is exceeded, the namespace goes into stop-writes mode and refuses new write commands until memory usage falls below the configured threshold.

Note

Index management commands are run synchronously. When the command is complete, verify the state of the index before using it.

Avoid creating several indexes at one time, or creating indexes while nodes are migrating.

Create a secondary index

Use manage sindex create to create an index. For example:

Admin> enable
Admin+> manage sindex create numeric ageidx ns test set employees bin age

Drop a secondary index

Use manage sindex delete to drop a secondary index. For example:

Admin> enable
Admin+> manage sindex delete ageidx ns test set employees

List the secondary indexes

Use show sindex to list the secondary indexes:

Admin> show sindex
~~~~~~~~Secondary Indexes (2025-05-17 19:18:08 UTC)~~~~~~~~~
    Index|Namespace|      Set|     Bin|    Bin|  Index|State
     Name|         |         |        |   Type|   Type|
ageidx   |test     |employees|     age|numeric|default|RW
officeidx|test     |employees|  office|string |default|RW
phoneidx |test     |employees|phonenum|numeric|default|RW
Number of rows: 3

Admin> show sindex like age
~~~~~~~~Secondary Indexes (2025-05-17 19:18:12 UTC)~~~~~~~~~
    Index|Namespace|      Set|     Bin|    Bin|  Index|State
     Name|         |         |        |   Type|   Type|
ageidx   |test     |employees|     age|numeric|default|RW
Number of rows: 1

Get secondary index statistics

Use show statistics to display secondary index statistics:

Admin> show statistics sindex like bval
~test employees ageidx SIndex Statistics (2025-05-17 20:13:16 UTC)~
Node            |1.0.0.127.in-addr.arpa:3100
entries_per_bval|167
Number of rows: 2

~test employees officeidx SIndex Statistics (2025-05-17 20:13:16 UTC)~
Node            |1.0.0.127.in-addr.arpa:3100
entries_per_bval|1429
Number of rows: 2

~test employees phoneidx SIndex Statistics (2025-05-17 20:13:16 UTC)~
Node            |1.0.0.127.in-addr.arpa:3100
entries_per_bval|1
Number of rows: 2

Alternatively, use the sindex-stat info command directly:

Admin+> asinfo -l -v sindex-stat:namespace=test;indexname=officeidx
1.0.0.127.in-addr.arpa:3100 (127.0.0.1) returned:
entries=10000
used_bytes=16777216
entries_per_bval=1429
entries_per_rec=1
load_pct=100
load_time=30
stat_gc_recs=0

Performance histograms

Note

Deprecated in Database 6.0 and later.

To enable writing secondary index performance histograms to the log:

asadm -e 'enable; "asinfo -v "sindex-histogram:ns=NAMESPACE;[set=demo];indexname=INDEX_NAME;enable=true"'

To disable writing secondary index performance histograms to the log:

asadm -e 'enable; asinfo -v "sindex-histogram:ns=NAMESPACE;[set=demo];indexname=INDEX_NAME;enable=false"'

Related documentation

• Queries
• Secondary index system limits