Skip to content

Threading Functions

Functions for configuring the global thread pool used for parallel computation.

Thread Pool Configuration

set_num_threads builtin 

set_num_threads(n: int) -> Any

Set the number of threads for parallel computation.

Configures the global thread pool used by Brahe for parallel operations such as access computations. This function can be called multiple times to dynamically change the thread pool configuration - each call will reinitialize the pool with the new thread count.

Parameters:

Name Type Description Default
n int

Number of threads to use. Must be at least 1.

 required

Raises:

Type Description
ValueError

If n < 1.
RuntimeError

If thread pool fails to build.
Example 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
import brahe as bh

# Set to 4 threads initially
bh.set_num_threads(4)
print(f"Threads: {bh.get_max_threads()}")  # Output: 4

# Reinitialize with 8 threads - no error!
bh.set_num_threads(8)
print(f"Threads: {bh.get_max_threads()}")  # Output: 8

# All parallel operations (e.g., location_accesses) will now use
# 8 threads unless overridden with AccessSearchConfig.num_threads
Note

Unlike earlier versions, this function no longer raises an error if the thread pool has already been initialized. You can safely call it at any time to reconfigure the thread pool.

set_max_threads builtin 

set_max_threads() -> Any

Set the thread pool to use all available CPU cores.

This is a convenience function that sets the number of threads to 100% of available CPU cores. Can be called multiple times to reinitialize the thread pool dynamically.

Raises:

Type Description
RuntimeError

If thread pool fails to build.
Example 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
import brahe as bh

# Use all available CPU cores
bh.set_max_threads()
print(f"Using all {bh.get_max_threads()} cores")

# Switch to 2 threads
bh.set_num_threads(2)

# Switch back to max - no error!
bh.set_max_threads()
print(f"Back to {bh.get_max_threads()} cores")
Note

This function can be called at any time, even after the thread pool has been initialized with a different configuration.

set_ludicrous_speed builtin 

set_ludicrous_speed() -> Any

LUDICROUS SPEED! GO!

Set the thread pool to use all available CPU cores (alias for set_max_threads).

This is a fun alias for set_max_threads() that sets the number of threads to 100% of available CPU cores for maximum performance. Can be called multiple times to dynamically reinitialize the thread pool.

Raises:

Type Description
RuntimeError

If thread pool fails to build.
Example 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
import brahe as bh

# MAXIMUM POWER! Use all available CPU cores
bh.set_ludicrous_speed()
print(f"Going ludicrous with {bh.get_max_threads()} threads!")

# Throttle down for testing
bh.set_num_threads(1)

# ENGAGE LUDICROUS SPEED again - no error!
bh.set_ludicrous_speed()
Note

This function can be called at any time to reconfigure the thread pool to use maximum available cores, regardless of previous configuration.

get_max_threads builtin 

get_max_threads() -> int

Get the current maximum number of threads for parallel computation.

Returns the number of threads configured for the global thread pool. If the thread pool hasn't been initialized yet, this initializes it with the default (90% of available cores) and returns that value.

Returns:

Name Type Description
int int

Number of threads currently configured.
Example 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
import brahe as bh

# Get default thread count (90% of cores, initialized on first call)
threads = bh.get_max_threads()
print(f"Default: {threads} threads")

# Set to specific value and verify
bh.set_num_threads(4)
assert bh.get_max_threads() == 4

# Reconfigure and verify again
bh.set_num_threads(8)
assert bh.get_max_threads() == 8

# Switch to max cores
bh.set_max_threads()
print(f"Max cores: {bh.get_max_threads()}")
Note

Calling this function will initialize the thread pool with default settings (90% of cores) if it hasn't been configured yet. After initialization, you can still reconfigure it using set_num_threads() or set_max_threads().