Scenarios describe the behavior you want MZBench to emulate during the benchmark. If you're testing an online store, your scenario will probably include opening a product page and adding the product to cart. For a search service, the scenario may be searching for a random word. You get the idea.

In MZBench, scenarios are .erl files written in a simple DSL. It looks a lot like Erlang but is much simpler.

Read on to learn how to write test scenarios for MZBench.

*[DSL]: Domain-Specific Language

MZBench test scenarios consist of lists, tuples, and atoms:

A list is a comma-separated sequence enclosed in square brackets: [A, B, C]
A tuple is a comma-separated sequence enclosed in curly braces: {A, B, C}
Atoms are reserved words—function names, rate units, conditions. Basically, anything that's not a list, a tuple, a number, or a string is an atom: print, make_install, lte, rpm

The whole scenario is a list of tuples with a dot at the end:

[
    {function1, param1, param2},
    {function2, [{param_name, param_value}]},
    ...
].

Each of these tuples is called a statement. A statement represents a function call: the first item is the function name, the others are the params to pass to it. Each param in its turn can also be a list, a tuple, or an atom.

Some statements only appear at the top level of a scenario. They're called top-level statements. There're two kinds of top-level statements: directives and pools.

See live examples of MZBench scenarios on GitHub →

Directives

Directives prepare the system for the benchmark and clean it up after it. This includes installing an external worker on test nodes, registering resource files, checking conditions, and executing shell commands before and after the test.

Top-Level Directives

make_install

{make_install, [{git, "<URL>"}, {branch, "<Branch>"}, {dir, "<Dir>"}]}

Install an external worker from a remote git repository on the test nodes before running the benchmark.

MZBench downloads the worker and builds a .tgz archive, which is then distributed among the nodes and used in future provisions.

The following actions are executed during make_install:

$ git clone <URL> temp_dir
$ cd temp_dir
$ git checkout <Branch>
$ cd <Dir>
$ make generate_tgz

If branch is not specified, the default git branch is used.

If dir is not specified, . is used.

include_resource

{include_resource, <ResourceName>, "<FileName>", <Type>}
{include_resource, <ResourceName>, "<FileURL>", <Type>}`

Register a resource file as <ResourceName>.

If the file is on your local machine, put it in the same directory where you invoke mzbench run.

<Type> is one of the following atoms:

text : Plain text file, interpreted as a single string.

json : JSON file. Lists are interpreted as Erlang lists, objects are interpreted as Erlang maps.

tsv : File with tabulation separated values, interpreted as a list of lists.

erlang : Erlang source file, interpreted directly as an Erlang term.

binary : Custom binary (image, executable, archive, etc.), not interpreted.

pre_hook and post_hook

{pre_hook, <Actions>}
{post_hook, <Actions>}

Run actions before and after the benchmark. Two kinds of actions are supported: exec commands and worker calls:

Actions = [Action]
Action = {exec, Target, BashCommand}
    | {worker_call, WorkerMethod, WorkerModule}
    | {worker_call, WorkerMethod, WorkerModule, WorkerType}
Target = all | director

Exec commands let you to run any shell command on all nodes or only on the director node.

Worker calls are functions defined by the worker. They can be executed only on the director node. Worker calls are used to update the environment variables used in the benchmark.

assert

{assert, always, <Condition>}
{assert, <Time>, <Condition>}

Check if the condition <Condition> is satisfied throughout the entire benchmark or at least for the amount of time <Time>.

<Condition> is a comparison of two value and is defined as a tuple {<Operation>, <Operand1>, <Operand2>}.

<Operation> is one of four atoms:

lt : Less than.

gt : Greater than.

lte : Less than or equal to.

gte : Greater than or equal to.

<Operand1> and <Operand2> are the values to compare. They can be integers, floats, or metrics values.

Metrics are numerical values collected by the worker during the benchmark. To get the metric value, put its name between double quotation marks:

{gt, "http_ok", 20}

The http_ok metric is provided by the simple_http worker. This condition passes if the number of successful HTTP responses is greater than 20.

Pools

Pool represents a sequence of jobs—statements to run. The statements are defined by the worker and MZBench's standard library. The jobs are evenly distributed between nodes, so they can be executed in parallel.

Here's a pool that sends HTTP GET requests to two sites on 10 nodes in parallel:

    [ {pool,
        [ {size, 10}, {worker_type, simple_http_worker} ],
        [
            {get, "http://example.com"},
            {get, "http://foobar.com"} 
        ]
    } ].

The get statement is provided by the built-in simple_http worker.

The first param in the pool statement is a list of pool options.

Pool Options

size

required

{size, <NumberOfJobs>}`

How many times you want the pool executed.

If there's enough nodes and worker_start is not set, MZBench will start the jobs simultaneously and run them in parallel.

<NumberOfJobs> is any positive number.

worker_type

required

{worker_type, <WorkerName>}

The worker that provides statements for the jobs.

!!!hint A pool uses exactly one worker. If you need multiple workers in the benchmark, just write a pool for each one.

worker_start

{worker_start, {linear, <Rate>}}
{worker_start, {poisson, <Rate>}}
{worker_start, {exp, <Scale>, <Time>}}
{worker_start, {pow, <Exponent>, <Scale>, <Time>}}

Start the jobs with a given rate:

linear : Constant rate <Rate>, e.g. 10 per minute.

poisson : Rate defined by a Poisson process with λ = <Rate>.

exp : Start jobs with exponentially growing rate with the scale factor <Scale>:

*Scale × e<sup>Time</sup>*

pow : Start jobs with rate growing as a power function with the exponent <Exponent> and the scale factor <Scale>:

*Scale × Time<sup>Exponent</sup>*

You can customize and combine rates:

think_time

{think_time, <Time>, <Rate>}

Start jobs with rate <Rate> for a second, then sleep for <Time> and repeat.

ramp

{ramp, linear, <StartRate>, <EndRate>}

Linearly change the rate from <StartRate> at the beginning of the pool to <EndRate> at its end.

comb

{comb, <Rate1>, <Time1>, <Rate2>, <Time2>, ...}

Start jobs with rate <Rate1> for <Time1>, then switch to <Rate2> for <Time2>, etc.

Loops

Loop is a sequence of statements executed over and over for a given time.

A loop looks similar to a pool—it consists of a list of [options](#loop options) and a list statements to run:

{loop, [
        {time, <Time>},
        {rate, <Rate>},
        {parallel, <N>},
        {iterator, <Name>},
        {spawn, <Spawn>}
    ],
    [
        <Statement1>,
        <Statement2>,
        ...
    ]
}

Here's a loop that sends HTTP GET requests for 30 seconds with a growing rate of 1 → 5 rps:

{loop, [
        {time, {30, sec}},
        {rate, {ramp, linear, {1, rps}, {5, rps}}}
    ],
    [
        {get, "http://example.com"}
    ]
}

You can put loops inside loops. Here's a nested loop that sends HTTP GET requests for 30 seconds, increasing the rate by 1 rps every three seconds:

{loop, [
        {time, {30, sec}},
        {rate, {10, rpm}},
        {iterator, "i"}
    ],
    [
        {loop, [
                {time, {3, sec}}, 
                {rate, {{var, "i"}, rps}}
            ],
            [
                {get, "http://google.com"}
            ]
        }
    ]
}

The difference between these two examples is that in the first case the rate is growing smoothly and in the second one it's growing in steps.

Loop options

time

required

{time, <Time>}

Run the loop for <Time>.

rate

{rate, <Rate>}

Repeat the loop with the <Rate> rate.

parallel

{parallel, <N>}

Run <N> iterations of the loop in parallel.

iterator

{iterator, "<IterName>"}

Define a variable named <IterName> inside the loop that contains the current iteration number. It can be accessed with {var, <IterName>}.

spawn

{spawn, (true|false)}

If true, every iteration runs in a separate, spawned process. Default is false.

Resource Files

Resource file is an external data source for the benchmark.

To declare a resource file for the benchmark, use include_resource.

Once the resource file is registered, its content can be included at any place in the scenario using the resource statement: {resource, <ResourceName>}.

For example, suppose we have a file names.json:

[
    "Bob",
    "Alice",
    "Guido"
]

Here's how you can use this file in a scenario:

[
    {include_resource, names, "names.json", json},
    {pool, [
            {size, 3},
            {worker_type, dummy_worker}
        ],
        [
            {loop, [
                    {time, {5, sec}},
                    {rate, {1, rps}}
                ],
                [
                    {print, {choose, {resource, names}}} % print a random name from the file
                ]
            }
        ]
    }
].

Standard Library

Environment Variables

Environment variables are global values that can be accessed at any point of the benchmark. They are useful to store the benchmark global state like its total duration, or global params like the execution speed.

To set an environment variable, call mzbench with the --env param:

$ ./bin/mzbench run --env foo=bar --env n=42

var

{var, "<VarName>"}
{var, "<VarName>", <DefaultValue>}

To get the value of a variable, refer to it by the name: {var, "<VarName>"}.

{var, "foo"} % returns "bar"
{var, "n"} % returns "42", a string

If you refer to an undefined variable, the benchmark crashes. You can avoid this by setting a default value for the variable: {var, "<VarName>", <DefaultValue>}:

{var, "anothervar", "Foo"} % returns "Foo" if anothervar is not set

numvar

{numvar, "<VarName>"}
{numvar, "<VarName>", <DefaultValue>}

By default, variable values are considered strings. To get a numerical value (integer or float), use {numvar, "VarName"}:

{numvar, "n"} % returns 42, an integer.

Parallelization and Syncing

parallel

{parallel, <Statement1>, <Statement2>, ...}

Execute multiple statements in parallel. Unlike executing statements in a pool, this way all statements are executed on the same node.

set_signal

{set_signal, <SignalName>}
{set_signal, <SignalName>, <Count>]}

Emit a global signal <SignalName>.

If <Count> is specified, the signal is emitted <Count> times.

<SignalName> is a string, atom, number, or, in fact, any Erlang term.

wait_signal

{wait_signal, <SignalName>}
{wait_signal, <SignalName>, <Count>}

Wait for the global signal <SignalName> to be emitted. If <Count> is specified, wait for the signal to be emitted <Count> times.

Errors Handling

ignore_failure

{ignore_failure, <Statement>}

Execute the statement <Statement> and continue with the benchmark even if it fails.

If the statement succeeds, its result is returned; otherwise, the failure reason is returned.

Randomization

random_number

{random_number, <Min>, <Max>}
{random_number, <Max>}

Return a random number between <Min> and <Max>, including <Min> and not including <Max>.

{random_number, <Max>} is equivalent to {random_number, 0, <Max>}

random_list

{random_list, <Size>}

Return a list of random integer of length <Size>.

random_binary

{random_binary, <Size>}

Return a binary sequence of <Size> random bytes.

choose

{choose, <N>, <List>}
{choose, <List>}

Return a list of <N> random elements of the list <List>.

{choose, <List>} is equivalent to {choose, 1, <List>}.

round_robin

{round_robin, <List>}

Pick the next element of the list. When the last one is picked, start over from the first one.

Logging

dump

{dump, "<Text>"}

Write <Text> to the benchmark log.

sprintf

{sprintf, "<Format>", [<Value1>, <Value2>, ...]}

Return formatted text with a given format and placeholder values.

Data Conversion

t

{t, <List>}

Convert <List> to a tuple.

term_to_binary

{term_to_binary, <term>}

Convert an Erlang term to a binary object. Learn more in the Erlang docs.

Pause

wait

{wait, <Time>}

Pause the current job for <Time>.

Conventions

Time

<Time> is a tuple {<Duration>, (ms|sec|min|h)}:

{1, sec}` % one second
{10, min} % 10 minutes
{0.5, h} % half hour

Rate

<Rate> is a tuple {<N>, (rps|rpm|rph)}:

{10, rps} % 10 jobs per second
{12, rpm} % 12 jobs per minute
{100, h} % 100 jobs per hour

Files

scenarios.md

Latest commit

History

scenarios.md

File metadata and controls

Directives

Top-Level Directives

make_install

include_resource

pre_hook and post_hook

assert

Pools

Pool Options

size

worker_type

worker_start

think_time

ramp

comb

Loops

Loop options

time

rate

parallel

iterator

spawn

Resource Files

Standard Library

Environment Variables

var

numvar

Parallelization and Syncing

parallel

set_signal

wait_signal

Errors Handling

ignore_failure

Randomization

random_number

random_list

random_binary

choose

round_robin

Logging

dump

sprintf

Data Conversion

t

term_to_binary

Pause

wait

Conventions

Time

Rate