getDTthreads

C++ Function Reference

1 Signature

int getDTthreads(const int64_t n, const bool throttle)

2 Description

Determines number of threads for parallel operations.

3 Parameters

n (const int64_t): Number of iterations/items to process
throttle (const bool): Whether to apply thread throttling

4 Returns

int Number of threads to use

5 Details

nNumber of iterations/items to process throttleWhether to apply thread throttling int Number of threads to use Thread allocation strategy:With throttling: Incremental thread activation based on workloadWithout throttling: Direct allocation based on available threadsRespects global thread limits

6 Call Graph

7 Source Code

Implementation

File: inst/include/Utilities/openme-utils.hpp • Lines 173-183

inline int getDTthreads(const int64_t n, const bool throttle) {
    
    initDTthreads();
    
    // throttle==true  : a number of iterations per thread (DTthrottle) is applied before a second thread is utilized
    // throttle==false : parallel region is already pre-chunked such as in fread; e.g. two batches intended for two threads
    if (n<1) return 1; // 0 or negative could be deliberate in calling code for edge cases where loop is not intended to run at all
    int64_t ans = throttle ? 1+(n-1)/DTthrottle :  // 1 thread for n<=1024, 2 thread for n<=2048, etc
        n;                    // don't use 20 threads for just one or two batches
    return ans>=DTthreads ? DTthreads : (int)ans;  // apply limit in static local DTthreads saved there by initDTthreads() and setDTthreads()
}

8 Usage Example

#include "BigDataStatMeth.hpp"

// Example usage
auto result = getDTthreads(...);

---
title: "getDTthreads"
subtitle: "C++ Function Reference"
---

## Signature

```cpp
int getDTthreads(const int64_t n, const bool throttle)
```

## Description

Determines number of threads for parallel operations.

## Parameters

- `n` (`const int64_t`): Number of iterations/items to process
- `throttle` (`const bool`): Whether to apply thread throttling

## Returns

int Number of threads to use

## Details

nNumber of iterations/items to process throttleWhether to apply thread throttling int Number of threads to use
Thread allocation strategy:With throttling: Incremental thread activation based on workloadWithout throttling: Direct allocation based on available threadsRespects global thread limits

## Call Graph

::: {.diagram-section}
<object type="image/svg+xml" data="images/_big_data_stat_meth_8hpp_a7d6263da9da6b46b945746e3abdea4ca_cgraph.svg" class="class-diagram">
  <img src="images/_big_data_stat_meth_8hpp_a7d6263da9da6b46b945746e3abdea4ca_cgraph.svg" alt="Function dependencies"/>
</object>
:::

## Source Code

::: {.callout-note collapse="true"}
### Implementation

**File:** `inst/include/Utilities/openme-utils.hpp` • **Lines 173-183**

```cpp
inline int getDTthreads(const int64_t n, const bool throttle) {
    
    initDTthreads();
    
    // throttle==true  : a number of iterations per thread (DTthrottle) is applied before a second thread is utilized
    // throttle==false : parallel region is already pre-chunked such as in fread; e.g. two batches intended for two threads
    if (n<1) return 1; // 0 or negative could be deliberate in calling code for edge cases where loop is not intended to run at all
    int64_t ans = throttle ? 1+(n-1)/DTthrottle :  // 1 thread for n<=1024, 2 thread for n<=2048, etc
        n;                    // don't use 20 threads for just one or two batches
    return ans>=DTthreads ? DTthreads : (int)ans;  // apply limit in static local DTthreads saved there by initDTthreads() and setDTthreads()
}
```
:::

## Usage Example

```cpp
#include "BigDataStatMeth.hpp"

// Example usage
auto result = getDTthreads(...);
```