StackOverflow2013

Note that there are some explanatory texts on larger screens.

plurals

PO
text
Body
copied!<p>Just tested latency from Java on my Corei5 2.8GHz, only single byte send/received, 2 Java processes just spawned, without assigning specific CPU cores with taskset:</p> <pre><code>TCP - 25 microseconds Named pipes - 15 microseconds </code></pre> <p>Now explicitly specifying core masks, like <strong>taskset 1 java Srv</strong> or <strong>taskset 2 java Cli</strong>:</p> <pre><code>TCP, same cores: 30 microseconds TCP, explicit different cores: 22 microseconds Named pipes, same core: 4-5 microseconds !!!! Named pipes, taskset different cores: 7-8 microseconds !!!! </code></pre> <p>so </p> <pre><code>TCP overhead is visible scheduling overhead (or core caches?) is also the culprit </code></pre> <p>At the same time Thread.sleep(0) (which as strace shows causes a single sched_yield() Linux kernel call to be executed) takes 0.3 microsecond - so named pipes scheduled to single core still have much overhead</p> <p>Some shared memory measurement: <em>September 14, 2009 – Solace Systems announced today that its Unified Messaging Platform API can achieve an average latency of less than 700 nanoseconds using a shared memory transport.</em> <a href="http://solacesystems.com/news/fastest-ipc-messaging/">http://solacesystems.com/news/fastest-ipc-messaging/</a></p> <p>P.S. - tried shared memory next day in the form of memory mapped files, if busy waiting is acceptable, we can reduce latency to 0.3 microsecond for passing a single byte with code like this:</p> <pre><code>MappedByteBuffer mem = new RandomAccessFile("/tmp/mapped.txt", "rw").getChannel() .map(FileChannel.MapMode.READ_WRITE, 0, 1); while(true){ while(mem.get(0)!=5) Thread.sleep(0); // waiting for client request mem.put(0, (byte)10); // sending the reply } </code></pre> <p>Notes: Thread.sleep(0) is needed so 2 processes can see each other's changes (I don't know of another way yet). If 2 processes forced to same core with taskset, the latency becomes 1.5 microseconds - that's a context switch delay</p> <p>P.P.S - and 0.3 microsecond is a good number! The following code takes exactly 0.1 microsecond, while doing a primitive string concatenation only:</p> <pre><code>int j=123456789; String ret = "my-record-key-" + j + "-in-db"; </code></pre> <p>P.P.P.S - hope this is not too much off-topic, but finally I tried replacing Thread.sleep(0) with incrementing a static volatile int variable (JVM happens to flush CPU caches when doing so) and obtained - record! - <strong>72 nanoseconds latency java-to-java process communication</strong>!</p> <p>When forced to same CPU Core, however, volatile-incrementing JVMs never yield control to each other, thus producing exactly 10 millisecond latency - Linux time quantum seems to be 5ms... So this should be used only if there is a spare core - otherwise sleep(0) is safer.</p>

Querying!

Guidance

An individual column

Larger individual text columns get their own page to allow for proper reading.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload