Skip to content

Commit

Permalink
server: metrics: add llamacpp:prompt_seconds_total and llamacpp:token…
Browse files Browse the repository at this point in the history 
…s_predicted_seconds_total, reset bucket only on /metrics. Fix values cast to int. Add Process-Start-Time-Unix header. (ggerganov#5937)

Closes ggerganov#5850
  • Loading branch information
@phymbert @hodlen
phymbert authored and hodlen committed Apr 1, 2024
1 parent eba48a3 commit 6e7b656
Show file tree
Hide file tree
Showing 3 changed files with 46 additions and 13 deletions.
47 changes: 35 additions & 12 deletions examples/server/server.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -335,8 +335,12 @@ struct server_slot {
};

struct server_metrics {
const int64_t t_start = ggml_time_us();

uint64_t n_prompt_tokens_processed_total = 0;
uint64_t t_prompt_processing_total = 0;
uint64_t n_tokens_predicted_total = 0;
uint64_t t_tokens_generation_total = 0;

uint64_t n_prompt_tokens_processed = 0;
uint64_t t_prompt_processing = 0;
Expand All @@ -348,12 +352,14 @@ struct server_metrics {
n_prompt_tokens_processed_total += slot.n_prompt_tokens_processed;
n_prompt_tokens_processed += slot.n_prompt_tokens_processed;
t_prompt_processing += slot.t_prompt_processing;
t_prompt_processing_total += slot.t_prompt_processing;
}

void on_prediction(const server_slot &slot) {
n_tokens_predicted_total += slot.n_decoded;
n_tokens_predicted += slot.n_decoded;
t_tokens_generation += slot.t_token_generation;
n_tokens_predicted_total += slot.n_decoded;
n_tokens_predicted += slot.n_decoded;
t_tokens_generation += slot.t_token_generation;
t_tokens_generation_total += slot.t_token_generation;
}

void reset_bucket() {
Expand Down Expand Up @@ -1502,9 +1508,12 @@ struct server_context {
{ "idle", n_idle_slots },
{ "processing", n_processing_slots },
{ "deferred", queue_tasks.queue_tasks_deferred.size() },
{ "t_start", metrics.t_start},

{ "n_prompt_tokens_processed_total", metrics.n_prompt_tokens_processed_total},
{ "t_tokens_generation_total", metrics.t_tokens_generation_total},
{ "n_tokens_predicted_total", metrics.n_tokens_predicted_total},
{ "t_prompt_processing_total", metrics.t_prompt_processing_total},

{ "n_prompt_tokens_processed", metrics.n_prompt_tokens_processed},
{ "t_prompt_processing", metrics.t_prompt_processing},
Expand All @@ -1517,7 +1526,9 @@ struct server_context {
{ "slots", slots_data },
};

metrics.reset_bucket();
if (json_value(task.data, "reset_bucket", false)) {
metrics.reset_bucket();
}
queue_results.send(res);
} break;
}
Expand Down Expand Up @@ -2709,6 +2720,7 @@ int main(int argc, char ** argv) {
task.id_multi = -1;
task.id_target = -1;
task.type = SERVER_TASK_TYPE_METRICS;
task.data.push_back({{"reset_bucket", true}});

ctx_server.queue_results.add_waiting_task_id(task.id);
ctx_server.queue_tasks.post(task);
Expand All @@ -2732,36 +2744,44 @@ int main(int argc, char ** argv) {
{"counter", {{
{"name", "prompt_tokens_total"},
{"help", "Number of prompt tokens processed."},
{"value", data["n_prompt_tokens_processed_total"]}
{"value", (uint64_t) data["n_prompt_tokens_processed_total"]}
}, {
{"name", "prompt_seconds_total"},
{"help", "Prompt process time"},
{"value", (uint64_t) data["t_prompt_processing_total"] / 1.e3}
}, {
{"name", "tokens_predicted_total"},
{"help", "Number of generation tokens processed."},
{"value", data["n_tokens_predicted_total"]}
{"value", (uint64_t) data["n_tokens_predicted_total"]}
}, {
{"name", "tokens_predicted_seconds_total"},
{"help", "Predict process time"},
{"value", (uint64_t) data["t_tokens_generation_total"] / 1.e3}
}}},
{"gauge", {{
{"name", "prompt_tokens_seconds"},
{"help", "Average prompt throughput in tokens/s."},
{"value", n_prompt_tokens_processed ? 1e3 / t_prompt_processing * n_prompt_tokens_processed : 0}
{"value", n_prompt_tokens_processed ? 1.e3 / t_prompt_processing * n_prompt_tokens_processed : 0.}
},{
{"name", "predicted_tokens_seconds"},
{"help", "Average generation throughput in tokens/s."},
{"value", n_tokens_predicted ? 1e3 / t_tokens_generation * n_tokens_predicted : 0}
{"value", n_tokens_predicted ? 1.e3 / t_tokens_generation * n_tokens_predicted : 0.}
},{
{"name", "kv_cache_usage_ratio"},
{"help", "KV-cache usage. 1 means 100 percent usage."},
{"value", 1. * kv_cache_used_cells / params.n_ctx}
},{
{"name", "kv_cache_tokens"},
{"help", "KV-cache tokens."},
{"value", data["kv_cache_tokens_count"]}
{"value", (uint64_t) data["kv_cache_tokens_count"]}
},{
{"name", "requests_processing"},
{"help", "Number of request processing."},
{"value", data["processing"]}
{"value", (uint64_t) data["processing"]}
},{
{"name", "requests_deferred"},
{"help", "Number of request deferred."},
{"value", data["deferred"]}
{"value", (uint64_t) data["deferred"]}
}}}
};

Expand All @@ -2775,13 +2795,16 @@ int main(int argc, char ** argv) {
const std::string name = metric_def["name"];
const std::string help = metric_def["help"];

auto value = json_value(metric_def, "value", 0);
auto value = json_value(metric_def, "value", 0.);
prometheus << "# HELP llamacpp:" << name << " " << help << "\n"
<< "# TYPE llamacpp:" << name << " " << type << "\n"
<< "llamacpp:" << name << " " << value << "\n";
}
}

const int64_t t_start = data["t_start"];
res.set_header("Process-Start-Time-Unix", std::to_string(t_start));

res.set_content(prometheus.str(), "text/plain; version=0.0.4");
res.status = 200; // HTTP OK
});
Expand Down
1 change: 1 addition & 0 deletions examples/server/tests/features/server.feature
View file
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,7 @@ Feature: llama.cpp server
And a completion request with no api error
Then <n_predicted> tokens are predicted matching <re_content>
And prometheus metrics are exposed
And metric llamacpp:tokens_predicted is <n_predicted>

Examples: Prompts
| prompt | n_predict | re_content | n_predicted |
Expand Down
11 changes: 10 additions & 1 deletion examples/server/tests/features/steps/steps.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -586,14 +586,24 @@ async def step_prometheus_metrics_exported(context):
metric_exported = False
if context.debug:
print(f"/metrics answer:\n{metrics_raw}\n")
context.metrics = {}
for metric in parser.text_string_to_metric_families(metrics_raw):
match metric.name:
case "llamacpp:kv_cache_usage_ratio":
assert len(metric.samples) > 0
metric_exported = True
context.metrics[metric.name] = metric
assert int(metrics_response.headers["Process-Start-Time-Unix"]) > 0, "no header process start time"
assert metric_exported, "No metrics exported"


@step(u'metric {metric_name} is {metric_value:d}')
def step_assert_metric_value(context, metric_name, metric_value):
if metric_name not in context.metrics:
assert False, f"no metric {metric_name} in {context.metrics.keys()}"
assert context.metrics[metric_name].samples[0].value == metric_value, f"metric: {context.metrics[metric_name]}"


@step(u'available models')
def step_available_models(context):
# openai client always expects an api_key
Expand Down Expand Up @@ -879,7 +889,6 @@ def assert_n_tokens_predicted(completion_response, expected_predicted_n=None, re
f' {n_predicted} <> {expected_predicted_n}')



async def gather_tasks_results(context):
n_tasks = len(context.concurrent_tasks)
if context.debug:
Expand Down

0 comments on commit 6e7b656

Please sign in to comment.