diff --git a/_sass/libs/_vars.scss b/_sass/libs/_vars.scss index bca9d9b..be9bb56 100644 --- a/_sass/libs/_vars.scss +++ b/_sass/libs/_vars.scss @@ -47,10 +47,10 @@ border2-bg: rgba(144,144,144,0.5), accent1: ( - bg: #21b2a6, + bg: #1F79B7, fg-bold: #ffffff, - fg: mix(#21b2a6, #ffffff, 25%), - fg-light: mix(#21b2a6, #ffffff, 40%), + fg: mix(#1F79B7, #ffffff, 25%), + fg-light: mix(#1F79B7, #ffffff, 40%), border: rgba(0,0,0,0.125), border-bg: rgba(255,255,255,0.075), border2: rgba(0,0,0,0.25), @@ -58,10 +58,10 @@ ), accent2: ( - bg: #00ffcc, + bg: #c7deed, fg-bold: #ffffff, - fg: mix(#00ffcc, #ffffff, 25%), - fg-light: mix(#00ffcc, #ffffff, 40%), + fg: mix(#c7deed, #ffffff, 25%), + fg-light: mix(#c7deed, #ffffff, 40%), border: rgba(0,0,0,0.125), border-bg: rgba(255,255,255,0.075), border2: rgba(0,0,0,0.25), @@ -69,10 +69,10 @@ ), accent3: ( - bg: #00f0ff, + bg: #c7deed, fg-bold: #ffffff, - fg: mix(#00f0ff, #ffffff, 25%), - fg-light: mix(#00f0ff, #ffffff, 40%), + fg: mix(#c7deed, #ffffff, 25%), + fg-light: mix(#c7deed, #ffffff, 40%), border: rgba(0,0,0,0.125), border-bg: rgba(255,255,255,0.075), border2: rgba(0,0,0,0.25), @@ -80,10 +80,10 @@ ), accent4: ( - bg: #76ddff, + bg: #c7deed, fg-bold: #ffffff, - fg: mix(#76ddff, #ffffff, 25%), - fg-light: mix(#76ddff, #ffffff, 40%), + fg: mix(#c7deed, #ffffff, 25%), + fg-light: mix(#c7deed, #ffffff, 40%), border: rgba(0,0,0,0.125), border-bg: rgba(255,255,255,0.075), border2: rgba(0,0,0,0.25), @@ -102,10 +102,10 @@ ), accent6: ( - bg: #ed4933, + bg: #bfbfbf, fg-bold: #ffffff, - fg: mix(#ed4933, #ffffff, 25%), - fg-light: mix(#ed4933, #ffffff, 40%), + fg: mix(#bfbfbf, #ffffff, 25%), + fg-light: mix(#bfbfbf, #ffffff, 40%), border: rgba(0,0,0,0.125), border-bg: rgba(255,255,255,0.075), border2: rgba(0,0,0,0.25), @@ -122,4 +122,4 @@ border2: #bfbfbf, border2-bg: rgba(0,0,0,0.1) ) - ); \ No newline at end of file + ); diff --git a/getting_started.html b/getting_started.html deleted file mode 100644 index ee17782..0000000 --- a/getting_started.html +++ /dev/null @@ -1,52 +0,0 @@ ---- -layout: page -title: Getting started -subtitle: How to build and use the erlangpl script ---- -
-

Building

-

The Erlang Performance Lab tool (erlangpl for short) can be started using escript or as a regular Erlang release. First we need to build it.

- - 
$ git clone git@github.com:erlanglab/erlangpl.git
-$ cd erlangpl
-$ make rebar
-$ make
-$ ./bootstrap
-
-
-
-

Running erlangpl script

-

The erlangpl shell script is a self-contained escript, which can be started from a command line as long as you have Erlang/OTP installed.

- - 
$ ./erlangpl -h
-
-Usage: erlangpl [-n <node>] [-c <cookie>] [-p <plugin>] [-h]
-                [-v <verbose>] [-P <port>] [-V] [-s <sname>] [-l <name>]
-
-  -n, --node     Monitored node name
-  -c, --cookie   Overwrite ~/.erlang.cookie
-  -p, --plugin   Path to plugins
-  -h, --help     Show the program options
-  -v, --verbose  Verbosity level (-v, -vv, -vvv)
-  -P, --port     HTTP and WS port number
-  -V, --version  Show version information
-  -s, --sname    Start with a shortname
-  -l, --name     Start with a longname, default erlangpl@127.0.0.1
-
-$ ./erlangpl -n testnode@127.0.0.1 -c YOURCOOKIE
-
- -

Once started, try visiting http://localhost:8000/

-
-
-

Running development release

-

You can also start the tool as a regular Erlang release and connect to its console to debug the tool itself.

- - 
$ make
-$ rebar -f generate
-$ ./rel/erlangpl/bin/erlangpl console node=testnode@127.0.0.1 cookie=YOURCOOKIE
-
-
-
-

Have fun!

-
diff --git a/getting_started.md b/getting_started.md new file mode 100644 index 0000000..c54789a --- /dev/null +++ b/getting_started.md @@ -0,0 +1,107 @@ +--- +layout: page +title: Getting started +subtitle: How to build and use the erlangpl script +--- + +### Download prebuilt script + +The easiest way to get started is to download a prebuilt `erlangpl` script +([download link](https://erlang.pl/erlangpl.gz)). + +### Build it manually + +#### Prerequisites + +For building UI you need to have following dependencies installed: +* [node](https://nodejs.org/en/) +* [yarn](https://yarnpkg.com/lang/en/) +* [elm-lang](https://guide.elm-lang.org/install.html) + +Be aware that building UI can take some time. It takes around 1 minute on stock MacBook 2015 plus dependencies +download for the first time. Second time dependencies will be cached. + +``` +$ git clone --recursive git@github.com:erlanglab/erlangpl.git +$ cd erlangpl +$ make rebar +$ make build-ui +$ make +$ ./bootstrap +``` + +## Running erlangpl script + +The erlangpl shell script is a self-contained escript, which can be started from a command line as long as you have Erlang/OTP installed. + +``` +$ ./erlangpl -h + +Usage: erlangpl [-n ] [-c ] [-p ] [-h] + [-v ] [-P ] [-V] [-s ] [-l ] + + -n, --node Monitored node name + -c, --cookie Overwrite ~/.erlang.cookie + -p, --plugin Path to plugins + -h, --help Show the program options + -v, --verbose Verbosity level (-v, -vv, -vvv) + -P, --port HTTP and WS port number + -V, --version Show version information + -s, --sname Start with a shortname + -l, --name Start with a longname, default erlangpl@127.0.0.1 + +$ ./erlangpl -n testnode@127.0.0.1 -c YOURCOOKIE +``` + +Once started, try visiting http://localhost:8000/ + +## Examples + +### Connecting to an Elixir iex session + +``` +$ iex --name foo@127.0.0.1 -S mix +``` + +``` +$ ./erlangpl --node foo@127.0.0.1 +``` + +### Mnesia cluster +You can generate messages between nodes by querying a distributed database Mnesia. + +To setup a Mnesia cluster, start several Erlang nodes with unique names e.g. `a@`, `b@`, `c@`, etc. and start the database on all of them: +``` +erl -name a@127.0.0.1 +(a@127.0.0.1)1> mnesia:start(). +``` +Then create a `test_table` and configure it to be replicated on all nodes: +``` +(a@127.0.0.1)2> mnesia:change_config(extra_db_nodes, ['b@127.0.0.1']). +(a@127.0.0.1)3> mnesia:change_config(extra_db_nodes, ['c@127.0.0.1']). +(a@127.0.0.1)4> mnesia:change_config(extra_db_nodes, ['d@127.0.0.1']). +(a@127.0.0.1)5> mnesia:create_table(test_table, []). +(a@127.0.0.1)6> [mnesia:add_table_copy(test_table, Node, ram_copies) || Node <- nodes()]. +``` + +Here are some behaviours you can test: +``` +[begin mnesia:transaction(fun() -> mnesia:write({test_table, Key, "value"}) end), timer:sleep(10) end || Key <- lists:seq(1,2000)]. +[begin mnesia:sync_dirty(fun() -> mnesia:write({test_table, Key, "value"}) end), timer:sleep(10) end || Key <- lists:seq(1,2000)]. +[begin mnesia:dirty_write({test_table, Key, "value"}), timer:sleep(10) end || Key <- lists:seq(1,2000)]. +``` + +Videos from those experiments were posted on [YouTube](https://www.youtube.com/channel/UCGkcbu799cC1rtMaQtAajpg) + + +## Running development release + +You can also start the tool as a regular Erlang release and connect to its console to debug the tool itself. + +``` +$ make +$ rebar -f generate +$ ./rel/erlangpl/bin/erlangpl console node=testnode@127.0.0.1 cookie=YOURCOOKIE +``` + +Have fun! diff --git a/images/erlanglab_message_passing.png b/images/erlanglab_message_passing.png new file mode 100644 index 0000000..2c6e5d3 Binary files /dev/null and b/images/erlanglab_message_passing.png differ diff --git a/index.html b/index.html index 5d0666d..0ef96d2 100644 --- a/index.html +++ b/index.html @@ -7,7 +7,9 @@

{{ site.title }}

{{ site.description | markdownify }}

Learn More @@ -17,9 +19,9 @@

{{ site.title }}

-

What is Erlang Performance Lab ?

-

It is a tool for developers working with the Erlang VM (BEAM).
- It helps with observing a system and its performance analysis.

+

What is Erlang Performance Lab?

+

Tool for developers working with the Erlang VM (BEAM).
+ It helps with observing a system and its performance analysis.

  • GitHub
    • @@ -31,7 +33,8 @@

    What is Erlang Performance Lab ?

    -
    + +

    Dashboard

    Gives high level overview of the current system throughput, @@ -43,25 +46,42 @@

    Dashboard

    -
    -
    + + +
    +

    Message passing graph

    In Erlang-based systems processes share data through message passing. If certain processes receive more messages than they can handle, their inbox queue starts growing. Over time it can lead to bottlenecks. The graph visualises which processes communicate extensively - and highlights them in red, so you know which ones to click and inspect. - You can even put on a pair of 3D glasses - and admire the beauty of complex systems visualisations. + allowing you to easily spot key places slowing down your application.

    + +

    Traffic graph

    -

    - Gaining visibility into network traffic in distributed systems is of great importance as it helps to understand the behaviour of the entire system, which can consist of more than one Erlang node. - Graphical representation of clustering helps grasp the inter-node traffic patterns and allows displaying traffic data on a graph of nodes.

    +

    Gaining visibility into network traffic in distributed systems is of great + importance as it helps to understand the behaviour of the entire system, + which can consist of more than one Erlang node. + Graphical representation of clustering helps grasp the inter-node traffic + patterns and allows displaying traffic data on a graph of nodes. +

    +
    +
    + + +
    +
    +

    Supervision tree

    +

    May help you understand internal structure of your application. + This view allows you to monitor spawns and exits of your processes. + Visualizing failure and recovery can help you to understand relations + between workers and supervisors. +

    @@ -76,7 +96,7 @@

    Interested in erlang.pl?

    Don't forget to send us feedback!