# bw_mem

> Memory bandwidth benchmark workloads: Memory bandwidth (bw_mem).

Memory bandwidth

## Memory bandwidth (bw\_mem) {#bw_mem}

bw\_mem allocates twice the specified amount of memory, zeros it, and then times the copying of the first half to the second half. Results are reported in megabytes moved per second (MB/sec). bw\_mem is provided by lmbench. For more details, see the man pages.

Benchmark details<dl><dt>Benchmark ID</dt><dd><code>bw_mem</code></dd><dt>Unit of measurement</dt><dd>Megabytes per second (MB/sec)</dd><dt>Server types evaluated</dt><dd>3,654</dd><dt>Total measurements</dt><dd>120,494</dd><dt>Range of measurements</dt><dd>100.4 – 39.6M <small>MB/sec</small></dd><dt>Better</dt><dd>↑ Higher is better</dd></dl>

Configuration options

| Field | Description | Example values |
| --- | --- | --- |
| `size` | Amount of memory to be used in MB | `0.016384` `0.262144` `1` `2` `4` `8` `16` `32` `64` `256` `512` |
| `operation` | The type of measurement: 'rd' measures the time to read data into the processor, 'wr' measures the time to write data to memory, and 'rdwr' measures the time to read data into memory and then write data to the same memory location. | `rd` `rdwr` `wr` |

Score distribution <small>(120,494 measurements)</small>

![Score distribution of Memory bandwidth (bw\_mem)](/benchmarks/histograms/bw_mem.svg)
