Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

chore: added new graph object for nim runtimes #334

Merged
merged 6 commits into from
Jan 10, 2025
Merged

chore: added new graph object for nim runtimes #334

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
89214f9
chore: added new graph object for nim runtimes
TomerFi Dec 11, 2024
ea7c973
chore: added REQUEST_OUTCOMES nim graph
TomerFi Dec 18, 2024
cd5035d
chore: added fixed typo in nim query object
TomerFi Dec 18, 2024
5e987a6
chore: fixed typo in nim query object
TomerFi Dec 18, 2024
2c0d933
chore: added initial query for nim gpu cache usage
TomerFi Dec 19, 2024
86217c9
chore: rewrite queries for nim new graphs
TomerFi Dec 20, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions internal/controller/constants/constants.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,6 +73,7 @@ const (
DefaultStorageConfig = "storage-config"
IntervalValue = "1m"
RequestRateInterval = "5m"
GPUKVCacheSamplingInterval = "24h"
OvmsImageName = "openvino_model_server"
TgisImageName = "text-generation-inference"
VllmImageName = "vllm"
Expand Down
80 changes: 78 additions & 2 deletions internal/controller/constants/runtime-metrics.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,13 +241,13 @@ const (
}
]
},
{
{
"title": "Average response time (ms)",
"type": "MEAN_LATENCY",
"queries": [
{
"title": "Average e2e latency",
"query": "sum by (model_name) (rate(e2e_request_latency_seconds_sum{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]) * 1000) / sum by (model_name) (rate(e2e_request_latency_seconds_count{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]) * 1000)"
"query": "(rate(e2e_request_latency_seconds_sum{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]) * 1000) / (rate(e2e_request_latency_seconds_count{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]) * 1000)"
}
]
},
Expand All @@ -270,6 +270,82 @@ const (
"query": "sum(container_memory_working_set_bytes{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'})/sum(kube_pod_resource_limit{resource='memory', pod=~'${MODEL_NAME}-predictor-.*', namespace='${NAMESPACE}'})"
}
]
},
{
"title": "GPU cache usage over time",
"type": "KV_CACHE",
"queries": [
{
"title": "GPU cache usage over time",
"query": "sum_over_time(gpu_cache_usage_perc{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${KV_CACHE_SAMPLING_RATE}])"
}
]
},
{
"title": "Current running, waiting, and max requests count",
"type": "CURRENT_REQUESTS",
"queries": [
{
"title": "Requests waiting",
"query": "num_requests_waiting{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}"
},
{
"title": "Requests running",
"query": "num_requests_running{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}"
},
{
"title": "Max requests",
"query": "num_request_max{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}"
}
]
},
{
"title": "Tokens count",
"type": "TOKENS_COUNT",
"queries": [
{
"title": "Total prompts token",
"query": "round(rate(prompt_tokens_total{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]))"
},
{
"title": "Total generation token",
"query": "round(rate(generation_tokens_total{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}]))"
}
]
},
{
"title": "Time to first token",
"type": "TIME_TO_FIRST_TOKEN",
"queries": [
{
"title": "Time to first token",
"query": "rate(time_to_first_token_seconds_sum{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}])"
}
]
},
{
"title": "Time per output token",
"type": "TIME_PER_OUTPUT_TOKEN",
"queries": [
{
"title": "Time per output token",
"query": "rate(time_per_output_token_seconds_sum{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${RATE_INTERVAL}])"
}
]
},
{
"title": "Requests outcomes",
"type": "REQUEST_OUTCOMES",
"queries": [
{
"title": "Number of successful incoming requests",
"query": "round(sum(increase(request_success_total{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${REQUEST_RATE_INTERVAL}])))"
},
{
"title": "Number of failed incoming requests",
"query": "round(sum(increase(request_failure_total{namespace='${NAMESPACE}', pod=~'${MODEL_NAME}-predictor-.*'}[${REQUEST_RATE_INTERVAL}])))"
}
]
}
]
}`
Expand Down
3 changes: 2 additions & 1 deletion internal/controller/utils/utils.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -428,6 +428,7 @@ func SubstituteVariablesInQueries(data string, namespace string, name string) st
"${NAMESPACE}", namespace,
"${MODEL_NAME}", name,
"${RATE_INTERVAL}", constants.IntervalValue,
"${REQUEST_RATE_INTERVAL}", constants.RequestRateInterval)
"${REQUEST_RATE_INTERVAL}", constants.RequestRateInterval,
"${KV_CACHE_SAMPLING_RATE}", constants.GPUKVCacheSamplingInterval)
return replacer.Replace(data)
}
Loading