Configuration

FalkorDB supports Redis configuration and multiple module configuration parameters. Some of these parameters can only be set at load-time, while other parameters can be set either on load-time or on run-time.

For example the following will run the server with global authentication password and 4 threads.

docker run -p 6379:6379 -p 3000:3000 -it -e REDIS_ARGS="--requirepass falkordb" -e FALKORDB_ARGS="THREAD_COUNT 4" --rm falkordb/falkordb:latest

Setting configuration parameters on module load

Setting configuration parameters at load-time is done by appending arguments after the --loadmodule argument when starting a server from the command line or after the loadmodule directive in a Redis config file. For example:

In redis.conf:

loadmodule ./falkordb.so [OPT VAL]...

From the Redis CLI, using the MODULE LOAD command:

127.0.0.6379> MODULE LOAD falkordb.so [OPT VAL]...

From the command line:

$ redis-server --loadmodule ./falkordb.so [OPT VAL]...

When running a docker container

docker run -p 6379:6379 -p 3000:3000 -it -e FALKORDB_ARGS="[OPT VAL]" --rm falkordb/falkordb:latest

Setting configuration parameters at run-time (for supported parameters)

FalkorDB exposes the GRAPH.CONFIG command to allowing for the setting and retrieval of configuration parameters at run-time.

To set the value of a configuration parameter at run-time (for supported parameters), simply run:

GRAPH.CONFIG SET OPT1 VAL1

Similarly, current configuration parameter values can be retrieved using:

GRAPH.CONFIG GET OPT1
GRAPH.CONFIG GET *

Values set using GRAPH.CONFIG SET are not persisted after server restart.

FalkorDB configuration parameters

The following table summarizes which configuration parameters can be set at module load-time and which can also be set at run-time:

Configuration Parameter	Load-time	Run-time
THREAD_COUNT	V	X
CACHE_SIZE	V	X
OMP_THREAD_COUNT	V	X
NODE_CREATION_BUFFER	V	X
BOLT_PORT	V	X
MAX_QUEUED_QUERIES	V	V
TIMEOUT (deprecated in v2.10)	V	V
TIMEOUT_MAX (since v2.10)	V	V
TIMEOUT_DEFAULT (since v2.10)	V	V
RESULTSET_SIZE	V	V
QUERY_MEM_CAPACITY	V	V
VKEY_MAX_ENTITY_COUNT	V	V
EFFECTS_THRESHOLD	V	V
CMD_INFO	V	V
MAX_INFO_QUERIES	V	V

THREAD_COUNT

The number of threads in FalkorDB’s thread pool. This is equivalent to the maximum number of queries that can be processed concurrently.

Default

THREAD_COUNT defaults to the system’s hardware threads (logical cores).

Example

$ redis-server --loadmodule ./falkordb.so THREAD_COUNT 4

CACHE_SIZE

The max number of queries for FalkorDB to cache. When a new query is encountered and the cache is full, meaning the cache has reached the size of CACHE_SIZE, it will evict the least recently used (LRU) entry.

Default

CACHE_SIZE default value is 25.

Example

$ redis-server --loadmodule ./falkordb.so CACHE_SIZE 10

OMP_THREAD_COUNT

The maximum number of threads that OpenMP may use for computation per query. These threads are used for parallelizing GraphBLAS computations, so may be considered to control concurrency within the execution of individual queries.

Default

OMP_THREAD_COUNT is defined by GraphBLAS.

Example

$ redis-server --loadmodule ./falkordb.so OMP_THREAD_COUNT 1

NODE_CREATION_BUFFER

The node creation buffer is the number of new nodes that can be created without resizing matrices. For example, when set to 16,384, the matrices will have extra space for 16,384 nodes upon creation. Whenever the extra space is depleted, the matrices’ size will increase by 16,384.

Reducing this value will reduce memory consumption, but cause performance degradation due to the increased frequency of matrix resizes.

Conversely, increasing it might improve performance for write-heavy workloads but will increase memory consumption.

If the passed argument was not a power of 2, it will be rounded to the next-greatest power of 2 to improve memory alignment.

Default

NODE_CREATION_BUFFER is 16,384.

Minimum

The minimum value for NODE_CREATION_BUFFER is 128. Values lower than this will be accepted as arguments, but will internally be converted to 128.

Example

$ redis-server --loadmodule ./falkordb.so NODE_CREATION_BUFFER 200

BOLT_PORT

The Bolt port configuration determines the port number on which FalkorDB handles the bolt protocol

Default

BOLT_PORT -1 (disabled).

Example

$ redis-server --loadmodule ./falkordb.so BOLT_PORT 7687

MAX_QUEUED_QUERIES

Setting the maximum number of queued queries allows the server to reject incoming queries with the error message Max pending queries exceeded. This reduces the memory overhead of pending queries on an overloaded server and avoids congestion when the server processes its backlog of queries.

Default

MAX_QUEUED_QUERIES is effectively unlimited (config value of UINT64_MAX).

Example

$ redis-server --loadmodule ./falkordb.so MAX_QUEUED_QUERIES 500

$ redis-cli GRAPH.CONFIG SET MAX_QUEUED_QUERIES 500

TIMEOUT

(Deprecated in FalkorDB v2.10 It is recommended to use TIMEOUT_MAX and TIMEOUT_DEFAULT instead)

The TIMEOUT configuration parameter specifies the default maximal execution time for read queries, in milliseconds. Write queries do not timeout.

When a read query execution time exceeds the maximal execution time, the query is aborted and the query reply is (error) Query timed out.

The TIMEOUT query parameter of the GRAPH.QUERY, GRAPH.RO_QUERY, and GRAPH.PROFILE commands can be used to override this value.

Default

Before v2.10: TIMEOUT is off (set to 0).
Since v2.10: TIMEOUT is not specified; TIMEOUT_MAX and TIMEOUT_DEFAULT are specified instead.

Example

$ redis-server --loadmodule ./falkordb.so TIMEOUT 1000

TIMEOUT_MAX

(Since v2.10)

The TIMEOUT_MAX configuration parameter specifies the maximum execution time for both read and write queries, in milliseconds.

The TIMEOUT query parameter value of the GRAPH.QUERY, GRAPH.RO_QUERY, and GRAPH.PROFILE commands cannot exceed the TIMEOUT_MAX value (the command would abort with a (error) The query TIMEOUT parameter value cannot exceed the TIMEOUT_MAX configuration parameter value reply). Similarly, the TIMEOUT_DEFAULT configuration parameter cannot exceed the TIMEOUT_MAX value.

When a query execution time exceeds the maximal execution time, the query is aborted and the query reply is (error) Query timed out. For a write query - any change to the graph is undone (which may take additional time).

Default

Before v2.10: unspecified and unsupported.
Since v2.10: TIMEOUT_MAX is off (set to 0).

Example

$ redis-server --loadmodule ./falkordb.so TIMEOUT_MAX 1000

TIMEOUT_DEFAULT

(Since v2.10)

The TIMEOUT_DEFAULT configuration parameter specifies the default maximal execution time for both read and write queries, in milliseconds.

For a given query, this default maximal execution time can be overridden by the TIMEOUT query parameter of the GRAPH.QUERY, GRAPH.RO_QUERY, and GRAPH.PROFILE commands. However, a query execution time cannot exceed TIMEOUT_MAX.

Default

Before v2.10: unspecified and unsupported.
Since v2.10: TIMEOUT_DEFAULT is equal to TIMEOUT_MAX (set to 0).

Example

$ redis-server --loadmodule ./falkordb.so TIMEOUT_MAX 2000 TIMEOUT_DEFAULT 1000

RESULTSET_SIZE

Result set size is a limit on the number of records that should be returned by any query. This can be a valuable safeguard against incurring a heavy IO load while running queries with unknown results.

Default

RESULTSET_SIZE is unlimited (negative config value).

Example

127.0.0.1:6379> GRAPH.CONFIG SET RESULTSET_SIZE 3
OK
127.0.0.1:6379> GRAPH.QUERY G "UNWIND range(1, 5) AS x RETURN x"
1) 1) "x"
2) 1) 1) (integer) 1
   2) 1) (integer) 2
   3) 1) (integer) 3
3) 1) "Cached execution: 0"
   2) "Query internal execution time: 0.445790 milliseconds"

QUERY_MEM_CAPACITY

Setting the memory capacity of a query allows the server to kill queries that are consuming too much memory and return with the error message Query's mem consumption exceeded capacity. This helps to avoid scenarios when the server becomes unresponsive due to an unbounded query exhausting system resources.

The configuration argument is the maximum number of bytes that can be allocated by any single query.

Default

QUERY_MEM_CAPACITY is unlimited; this default can be restored by setting QUERY_MEM_CAPACITY to zero or a negative value.

Example

$ redis-server --loadmodule ./falkordb.so QUERY_MEM_CAPACITY 1048576 // 1 megabyte limit

$ redis-cli GRAPH.CONFIG SET QUERY_MEM_CAPACITY 1048576

VKEY_MAX_ENTITY_COUNT

To lower the time Redis is blocked when replicating large graphs, FalkorDB serializes the graph in a number of virtual keys.

One virtual key is created for every N graph entities, where N is the value defined by this configuration.

This configuration can be set when the module loads or at runtime.

Default

VKEY_MAX_ENTITY_COUNT is 100,000.

CMD_INFO

An on/off toggle for the GRAPH.INFO command. Disabling this command may increase performance and lower the memory usage and these are the main reasons for it to be disabled.

It’s valid values are ‘yes’ and ‘no’ (i.e., on and off).

Default

CMD_INFO is yes.

MAX_INFO_QUERIES

A limit for the number of previously executed queries stored in the telemetry stream.

A number within the range [0, 1000]

Default

MAX_INFO_QUERIES is 1000.

Query Configurations

Query Timeout

Before v2.10, or if TIMEOUT_DEFAULT and TIMEOUT_MAX are not specified:

TIMEOUT allows overriding the TIMEOUT configuration parameter for a single read query. Write queries do not timeout.
Since v2.10, if either TIMEOUT_DEFAULT or TIMEOUT_MAX are specified:

TIMEOUT allows overriding the TIMEOUT_DEFAULT configuration parameter value for a single GRAPH.QUERY, GRAPH.RO_QUERY, or GRAPH.PROFILE command. The TIMEOUT value cannot exceed the TIMEOUT_MAX value (the command would abort with a (error) The query TIMEOUT parameter value cannot exceed the TIMEOUT_MAX configuration parameter value reply).

Example

Retrieve all paths in a graph with a timeout of 500 milliseconds.

GRAPH.QUERY wikipedia "MATCH p=()-[*]->() RETURN p" TIMEOUT 500

EFFECTS_THRESHOLD

Replicate modification via effect when average modification time > EFFECTS_THRESHOLD

Default

EFFECTS_THRESHOLD is 300 μs.

Example

Assume MATCH (n) WHERE n.id < 100 SET n.v = n.v + 1 updated 5 nodes and the query total execution time is 5ms, the average modification time is: total execution time / number of changes: 5ms / 5 = 1ms. if the average modification time is greater then EFFECTS_THRESHOLD the query will be replicated to both replicas and AOF as a graph effect otherwise the original query will be replicated.