# membench

> Memory bandwidth, Memory latency benchmark workloads: Memory copy bandwidth (sc-membench), Memory read bandwidth (sc-membench) and 2 more.

Memory bandwidth, Memory latency

## Memory copy bandwidth (sc-membench) {#membench-bandwidth_copy}

Measures aggregate memory copy bandwidth across all threads using OpenMP parallelization.

Benchmark details<dl><dt>Benchmark ID</dt><dd><code>membench:bandwidth_copy</code></dd><dt>Unit of measurement</dt><dd>MB/s</dd><dt>Server types evaluated</dt><dd>2,393</dd><dt>Total measurements</dt><dd>25,660</dd><dt>Range of measurements</dt><dd>940.8 – 17.1M <small>MB/s</small></dd><dt>Better</dt><dd>↑ Higher is better</dd></dl>

Configuration options

| Field | Description | Example values |
| --- | --- | --- |
| `scope` | RAM: working set exceeds total CPU cache (L3, else L2, else L1d). | `RAM` |
| `size_kb` | Per-thread buffer size in KiB. | `16` `32` `64` `128` `256` `512` `1024` `2048` `4096` `8192` `16384` `32768` `65536` `131072` `262144` `524288` `1048576` `2097152` |

Score distribution <small>(25,660 measurements)</small>

![Score distribution of Memory copy bandwidth (sc-membench)](/benchmarks/histograms/membench-bandwidth_copy.svg)

## Memory read bandwidth (sc-membench) {#membench-bandwidth_read}

Measures aggregate sequential read bandwidth across all threads using OpenMP parallelization.

Benchmark details<dl><dt>Benchmark ID</dt><dd><code>membench:bandwidth_read</code></dd><dt>Unit of measurement</dt><dd>MB/s</dd><dt>Server types evaluated</dt><dd>2,393</dd><dt>Total measurements</dt><dd>25,660</dd><dt>Range of measurements</dt><dd>1.9K – 37M <small>MB/s</small></dd><dt>Better</dt><dd>↑ Higher is better</dd></dl>

Configuration options

| Field | Description | Example values |
| --- | --- | --- |
| `scope` | RAM: working set exceeds total CPU cache (L3, else L2, else L1d). | `RAM` |
| `size_kb` | Per-thread buffer size in KiB. | `16` `32` `64` `128` `256` `512` `1024` `2048` `4096` `8192` `16384` `32768` `65536` `131072` `262144` `524288` `1048576` `2097152` |

Score distribution <small>(25,660 measurements)</small>

![Score distribution of Memory read bandwidth (sc-membench)](/benchmarks/histograms/membench-bandwidth_read.svg)

## Memory write bandwidth (sc-membench) {#membench-bandwidth_write}

Measures aggregate sequential write bandwidth across all threads using OpenMP parallelization.

Benchmark details<dl><dt>Benchmark ID</dt><dd><code>membench:bandwidth_write</code></dd><dt>Unit of measurement</dt><dd>MB/s</dd><dt>Server types evaluated</dt><dd>2,393</dd><dt>Total measurements</dt><dd>25,660</dd><dt>Range of measurements</dt><dd>1.7K – 40.6M <small>MB/s</small></dd><dt>Better</dt><dd>↑ Higher is better</dd></dl>

Configuration options

| Field | Description | Example values |
| --- | --- | --- |
| `scope` | RAM: working set exceeds total CPU cache (L3, else L2, else L1d). | `RAM` |
| `size_kb` | Per-thread buffer size in KiB. | `16` `32` `64` `128` `256` `512` `1024` `2048` `4096` `8192` `16384` `32768` `65536` `131072` `262144` `524288` `1048576` `2097152` |

Score distribution <small>(25,660 measurements)</small>

![Score distribution of Memory write bandwidth (sc-membench)](/benchmarks/histograms/membench-bandwidth_write.svg)

## Memory latency (sc-membench) {#membench-latency}

Measures median memory latency using pointer chasing with randomized access to defeat hardware prefetching.

Benchmark details<dl><dt>Benchmark ID</dt><dd><code>membench:latency</code></dd><dt>Unit of measurement</dt><dd>ns</dd><dt>Server types evaluated</dt><dd>2,393</dd><dt>Total measurements</dt><dd>25,660</dd><dt>Range of measurements</dt><dd>0.73 – 860.6 <small>ns</small></dd><dt>Better</dt><dd>↓ Lower is better</dd></dl>

Configuration options

| Field | Description | Example values |
| --- | --- | --- |
| `scope` | RAM: working set exceeds total CPU cache (L3, else L2, else L1d). | `RAM` |
| `size_kb` | Per-thread buffer size in KiB. | `16` `32` `64` `128` `256` `512` `1024` `2048` `4096` `8192` `16384` `32768` `65536` `131072` `262144` `524288` `1048576` `2097152` |

Score distribution <small>(25,660 measurements)</small>

![Score distribution of Memory latency (sc-membench)](/benchmarks/histograms/membench-latency.svg)
