--- title: Alibaba Cloud Function Compute description: >- Monitor Alibaba Cloud Function Compute invocations, errors, throttles, latency, and instance resource usage. breadcrumbs: Docs > Integrations > Alibaba Cloud Function Compute --- > For the complete documentation index, see [llms.txt](https://docs.datadoghq.com/llms.txt). # Alibaba Cloud Function Compute Integration version1.0.0 {% callout %} # Important note for users on the following Datadog sites: us2.ddog-gov.com {% alert level="info" %} To find out if this integration is available in your organization, see your [Datadog Integrations](https://app.datadoghq.com/integrations) page or ask your organization administrator. To initiate an exception request to enable this integration for your organization, email [support@ddog-gov.com](mailto:support@ddog-gov.com). {% /alert %} {% /callout %} ## Overview{% #overview %} [Alibaba Cloud Function Compute](https://www.alibabacloud.com/help/en/functioncompute/fc-3-0/product-overview/what-is-function-compute) is a fully managed, event-driven serverless compute service that runs code without requiring you to provision or manage servers. Use this integration to monitor function invocations, errors, throttles, and latency. You can also track instance and resource usage, asynchronous invocation backlog, and GPU inference workloads. ## Setup{% #setup %} ### Installation{% #installation %} After you set up the [Alibaba Cloud](https://docs.datadoghq.com/integrations/alibaba_cloud.md) integration, verify that the Function Compute service is included in your Alibaba Cloud monitoring configuration. ## Data Collected{% #data-collected %} ### Metrics{% #metrics %} | | | | | **alibabacloud.function_compute.async_events_backlog.value**(gauge) | Number of backlogged asynchronous invocations.*Shown as event* | | **alibabacloud.function_compute.async_events_expired_dropped.value**(gauge) | Number of asynchronous invocations dropped due to timeout.*Shown as event* | | **alibabacloud.function_compute.async_events_latency_avg.value**(gauge) | Average latency of backlogged asynchronous invocations.*Shown as millisecond* | | **alibabacloud.function_compute.async_events_latency_max.value**(gauge) | Maximum latency of backlogged asynchronous invocations.*Shown as millisecond* | | **alibabacloud.function_compute.async_events_received.value**(gauge) | Number of asynchronous invocations received.*Shown as event* | | **alibabacloud.function_compute.async_invocations.value**(gauge) | Number of asynchronous invocations.*Shown as invocation* | | **alibabacloud.function_compute.client_errors.value**(gauge) | Number of client errors.*Shown as error* | | **alibabacloud.function_compute.concurrency_throttles.value**(gauge) | Number of requests throttled because concurrent instances exceeded the limit.*Shown as request* | | **alibabacloud.function_compute.corp_function_cpu_usage_percent_avg.value**(gauge) | Average CPU usage of internal functions.*Shown as percent* | | **alibabacloud.function_compute.corp_function_cpu_usage_percent_max.value**(gauge) | Maximum CPU usage of internal functions.*Shown as percent* | | **alibabacloud.function_compute.corp_function_memory_usage_percent_avg.value**(gauge) | Average memory usage of internal functions.*Shown as percent* | | **alibabacloud.function_compute.corp_function_memory_usage_percent_max.value**(gauge) | Maximum memory usage of internal functions.*Shown as percent* | | **alibabacloud.function_compute.destination_delivery_failures.value**(gauge) | Number of failed asynchronous invocation destination deliveries.*Shown as event* | | **alibabacloud.function_compute.destination_delivery_successes.value**(gauge) | Number of successful asynchronous invocation destination deliveries.*Shown as event* | | **alibabacloud.function_compute.duration_avg.value**(gauge) | Average execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.duration_max.value**(gauge) | Maximum execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.duration_p90.value**(gauge) | 90th percentile execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.duration_p99.value**(gauge) | 99th percentile execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.function_async_event_expired_dropped.value**(gauge) | Number of function asynchronous invocation messages dropped due to timeout.*Shown as message* | | **alibabacloud.function_compute.function_async_message_latency_avg.value**(gauge) | Average backlog latency of function asynchronous invocation messages.*Shown as millisecond* | | **alibabacloud.function_compute.function_async_message_latency_max.value**(gauge) | Maximum backlog latency of function asynchronous invocation messages.*Shown as millisecond* | | **alibabacloud.function_compute.function_async_messages_backlog_v2.value**(gauge) | Number of backlogged function asynchronous messages.*Shown as message* | | **alibabacloud.function_compute.function_avg_duration.value**(gauge) | Average function execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.function_avg_memory_usage_mb.value**(gauge) | Average memory used by the function (instance metric).*Shown as mebibyte* | | **alibabacloud.function_compute.function_avg_memory_utilization.value**(gauge) | Average memory utilization of the function (instance metric).*Shown as percent* | | **alibabacloud.function_compute.function_avg_vcpu_cores.value**(gauge) | Average number of vCPU cores used by the function (instance metric).*Shown as core* | | **alibabacloud.function_compute.function_avg_vcpu_utilization.value**(gauge) | Average vCPU utilization of the function (instance metric).*Shown as percent* | | **alibabacloud.function_compute.function_billable_invocations.value**(gauge) | Number of billable function invocations.*Shown as invocation* | | **alibabacloud.function_compute.function_billable_invocations_rate.value**(gauge) | Percentage of function invocations that are billable.*Shown as percent* | | **alibabacloud.function_compute.function_cpu_percent.value**(gauge) | CPU usage of the function (100% equals one core).*Shown as percent* | | **alibabacloud.function_compute.function_cpu_quota_percent.value**(gauge) | CPU quota of the function (100% equals one core).*Shown as percent* | | **alibabacloud.function_compute.function_client_errors.value**(gauge) | Number of function client errors.*Shown as error* | | **alibabacloud.function_compute.function_client_errors_rate.value**(gauge) | Percentage of function invocations that resulted in client errors.*Shown as percent* | | **alibabacloud.function_compute.function_concurrency_throttles.value**(gauge) | Number of function requests throttled with error code 429.*Shown as request* | | **alibabacloud.function_compute.function_concurrent_count.value**(gauge) | Number of concurrent function instances (deprecated).*Shown as instance* | | **alibabacloud.function_compute.function_concurrent_requests.value**(gauge) | Number of concurrent requests.*Shown as request* | | **alibabacloud.function_compute.function_cost.value**(gauge) | Function resource usage (memory multiplied by execution time).*Shown as unit* | | **alibabacloud.function_compute.function_dequeue_count.value**(gauge) | Number of function asynchronous requests processed.*Shown as request* | | **alibabacloud.function_compute.function_destination_errors.value**(gauge) | Number of failed function asynchronous invocation destination triggers.*Shown as event* | | **alibabacloud.function_compute.function_destination_succeeded.value**(gauge) | Number of successful function asynchronous invocation destination triggers.*Shown as event* | | **alibabacloud.function_compute.function_elastic_active_instances.average**(gauge) | Number of active elastic function instances.*Shown as instance* | | **alibabacloud.function_compute.function_elastic_idle_instances.average**(gauge) | Number of idle (light sleep) elastic function instances.*Shown as instance* | | **alibabacloud.function_compute.function_elastic_instances_quota.average**(gauge) | Elastic instance quota of the function.*Shown as instance* | | **alibabacloud.function_compute.function_enqueue_count.value**(gauge) | Number of function asynchronous invocation requests enqueued.*Shown as request* | | **alibabacloud.function_compute.function_errors.value**(gauge) | Total number of function errors.*Shown as error* | | **alibabacloud.function_compute.function_function_errors.value**(gauge) | Number of errors raised by function code.*Shown as error* | | **alibabacloud.function_compute.function_function_errors_rate.value**(gauge) | Percentage of function invocations that resulted in function errors.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_decoder_percent.maximum**(gauge) | Maximum GPU hardware decoder utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_decoder_percent.average**(gauge) | Average GPU hardware decoder utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_encoder_percent.maximum**(gauge) | Maximum GPU hardware encoder utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_encoder_percent.average**(gauge) | Average GPU hardware encoder utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_max_memory_usage.maximum**(gauge) | Maximum GPU memory used by the function.*Shown as mebibyte* | | **alibabacloud.function_compute.function_gpu_cards_max_memory_usage.average**(gauge) | Average GPU memory used by the function.*Shown as mebibyte* | | **alibabacloud.function_compute.function_gpu_cards_memory_usage_percent.maximum**(gauge) | Maximum GPU memory utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_memory_usage_percent.average**(gauge) | Average GPU memory utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_sm_percent.maximum**(gauge) | Maximum GPU streaming multiprocessor (SM) utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_cards_sm_percent.average**(gauge) | Average GPU streaming multiprocessor (SM) utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_decoder_percent.value**(gauge) | Hardware decoder utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_encoder_percent.value**(gauge) | Hardware encoder utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_max_memory_usage.value**(gauge) | Maximum GPU memory usage of the function.*Shown as mebibyte* | | **alibabacloud.function_compute.function_gpu_memory_limit_mb.value**(gauge) | GPU memory quota of the function.*Shown as mebibyte* | | **alibabacloud.function_compute.function_gpu_memory_usage_percent.value**(gauge) | GPU memory utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_gpu_sm_percent.value**(gauge) | GPU streaming multiprocessor (SM) utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_http_status2xx.value**(gauge) | Number of function requests with HTTP status code 2xx.*Shown as request* | | **alibabacloud.function_compute.function_http_status3xx.value**(gauge) | Number of function requests with HTTP status code 3xx.*Shown as request* | | **alibabacloud.function_compute.function_http_status4xx.value**(gauge) | Number of function requests with HTTP status code 4xx.*Shown as request* | | **alibabacloud.function_compute.function_http_status5xx.value**(gauge) | Number of function requests with HTTP status code 5xx.*Shown as request* | | **alibabacloud.function_compute.function_latency_avg.value**(gauge) | Average end-to-end latency of the function.*Shown as millisecond* | | **alibabacloud.function_compute.function_latency_max.value**(gauge) | Maximum end-to-end latency of the function.*Shown as millisecond* | | **alibabacloud.function_compute.function_max_concurrency_instances.value**(gauge) | Instance quota of the function.*Shown as instance* | | **alibabacloud.function_compute.function_max_duration.value**(gauge) | Maximum function execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.function_max_memory_usage.value**(gauge) | Maximum memory used by the function.*Shown as mebibyte* | | **alibabacloud.function_compute.function_max_memory_usage_mb.value**(gauge) | Maximum memory used by function instances, reported per instance.*Shown as mebibyte* | | **alibabacloud.function_compute.function_max_memory_utilization.value**(gauge) | Maximum memory utilization of the function (instance metric).*Shown as percent* | | **alibabacloud.function_compute.function_max_vcpu_cores.value**(gauge) | Maximum number of vCPU cores used by the function (instance metric).*Shown as core* | | **alibabacloud.function_compute.function_max_vcpu_utilization.value**(gauge) | Maximum vCPU utilization of the function (instance metric).*Shown as percent* | | **alibabacloud.function_compute.function_memory_limit_mb.value**(gauge) | Memory quota of the function.*Shown as mebibyte* | | **alibabacloud.function_compute.function_memory_usage_percent.value**(gauge) | Memory utilization of the function.*Shown as percent* | | **alibabacloud.function_compute.function_min_instances.average**(gauge) | Minimum number of elastic function instances.*Shown as instance* | | **alibabacloud.function_compute.function_ondemand_active_instance.value**(gauge) | Number of on-demand function instances in use.*Shown as instance* | | **alibabacloud.function_compute.function_ondemand_instance_quota.value**(gauge) | Maximum number of on-demand function instances.*Shown as instance* | | **alibabacloud.function_compute.function_p90_duration.value**(gauge) | 90th percentile function execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.function_p99_duration.value**(gauge) | 99th percentile function execution duration.*Shown as millisecond* | | **alibabacloud.function_compute.function_provision_invocations.value**(gauge) | Number of function invocations served by provisioned instances.*Shown as invocation* | | **alibabacloud.function_compute.function_provisioned_current_instance.value**(gauge) | Number of provisioned function instances.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_async_event_expired_dropped.value**(gauge) | Number of function asynchronous invocation requests dropped due to timeout for the specified version or alias.*Shown as message* | | **alibabacloud.function_compute.function_qualifier_async_message_latency_avg.value**(gauge) | Average processing latency of function asynchronous invocation messages for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_async_message_latency_avg_v2.value**(gauge) | Average processing latency of function asynchronous invocation messages for the specified alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_async_message_latency_max.value**(gauge) | Maximum processing latency of function asynchronous invocation messages for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_async_message_latency_max_v2.value**(gauge) | Maximum processing latency of function asynchronous invocation messages for the specified alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_avg_concurrent_requests.value**(gauge) | Average number of requests per function instance for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_avg_duration.value**(gauge) | Average function execution duration for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_avg_vcpu_cores.value**(gauge) | Average number of vCPU cores used by the function for the specified version or alias.*Shown as core* | | **alibabacloud.function_compute.function_qualifier_avg_vcpu_utilization.value**(gauge) | Average vCPU utilization of the function for the specified version or alias (normalized, maximum value is 1).*Shown as fraction* | | **alibabacloud.function_compute.function_qualifier_client_errors.value**(gauge) | Number of client errors for the specified version or alias.*Shown as error* | | **alibabacloud.function_compute.function_qualifier_concurrency_throttles.value**(gauge) | Number of function requests exceeding the concurrent instance limit for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_concurrent_count.value**(gauge) | Number of function instances for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_cost.value**(gauge) | Function resource usage for the specified version or alias.*Shown as unit* | | **alibabacloud.function_compute.function_qualifier_dequeue_count.value**(gauge) | Number of function asynchronous invocation requests dequeued for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_destination_errors.value**(gauge) | Number of failed asynchronous destination trigger requests for the specified version or alias.*Shown as event* | | **alibabacloud.function_compute.function_qualifier_destination_succeeded.value**(gauge) | Number of successful asynchronous destination trigger requests for the specified version or alias.*Shown as event* | | **alibabacloud.function_compute.function_qualifier_elastic_active_instances.average**(gauge) | Number of active elastic instances for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_elastic_idle_instances.average**(gauge) | Number of idle (light sleep) elastic instances for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_enqueue_count.value**(gauge) | Number of function asynchronous invocation requests enqueued for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_function_errors.value**(gauge) | Number of function errors for the specified version or alias.*Shown as error* | | **alibabacloud.function_compute.function_qualifier_http_status2xx.value**(gauge) | Number of function requests with HTTP status code 2xx for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_http_status3xx.value**(gauge) | Number of function requests with HTTP status code 3xx for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_http_status4xx.value**(gauge) | Number of function requests with HTTP status code 4xx for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_http_status5xx.value**(gauge) | Number of function requests with HTTP status code 5xx for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_instances_quota.average**(gauge) | Elastic instance quota for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_latency_avg.value**(gauge) | Average end-to-end latency of the function for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_latency_max.value**(gauge) | Maximum end-to-end latency of the function for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_max_concurrent_requests.value**(gauge) | Maximum number of requests per function instance for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_max_duration.value**(gauge) | Maximum function execution duration for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_max_memory_usage.value**(gauge) | Maximum memory usage of the function for the specified version or alias.*Shown as mebibyte* | | **alibabacloud.function_compute.function_qualifier_max_vcpu_cores.value**(gauge) | Maximum number of vCPU cores used by the function for the specified version or alias.*Shown as core* | | **alibabacloud.function_compute.function_qualifier_max_vcpu_utilization.value**(gauge) | Maximum vCPU utilization of the function for the specified version or alias (normalized, maximum value is 1).*Shown as fraction* | | **alibabacloud.function_compute.function_qualifier_memory_limit_mb.value**(gauge) | Memory quota of the function for the specified version or alias.*Shown as mebibyte* | | **alibabacloud.function_compute.function_qualifier_memory_usage_mb.value**(gauge) | Memory usage of the function for the specified version or alias.*Shown as mebibyte* | | **alibabacloud.function_compute.function_qualifier_memory_usage_percent.value**(gauge) | Memory utilization of the function for the specified version or alias.*Shown as percent* | | **alibabacloud.function_compute.function_qualifier_min_instances.average**(gauge) | Minimum number of elastic instances for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_ondemand_active_instance.value**(gauge) | Number of on-demand function instances for the specified version or alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_ondemand_instance_quota.value**(gauge) | Maximum number of on-demand function instances for the specified version or alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_p90_duration.value**(gauge) | 90th percentile function execution duration for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_p99_duration.value**(gauge) | 99th percentile function execution duration for the specified version or alias.*Shown as millisecond* | | **alibabacloud.function_compute.function_qualifier_provision_invocations.value**(gauge) | Number of function invocations served by provisioned instances for the specified version or alias.*Shown as invocation* | | **alibabacloud.function_compute.function_qualifier_provisioned_current_instance.value**(gauge) | Number of provisioned function instances for the specified version or alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_rx_bytes_per_sec.value**(rate) | Inbound network traffic of the function for the specified version or alias.*Shown as byte* | | **alibabacloud.function_compute.function_qualifier_resident_instances.average**(gauge) | Number of resident instances for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_resident_live_instances.average**(gauge) | Number of live resident instances for the specified alias.*Shown as instance* | | **alibabacloud.function_compute.function_qualifier_resource_throttles.value**(gauge) | Number of function requests exceeding the total instance limit (error code 503) for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_server_errors.value**(gauge) | Number of server errors for the specified version or alias.*Shown as error* | | **alibabacloud.function_compute.function_qualifier_sglang_cache_hit_rate.value**(gauge) | Key-value (KV) cache hit rate of the SGLang instance.*Shown as percent* | | **alibabacloud.function_compute.function_qualifier_sglang_e2e_request_latency_seconds_avg.value**(gauge) | Average end-to-end SGLang request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_e2e_request_latency_seconds_p100.value**(gauge) | Maximum end-to-end SGLang request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_e2e_request_latency_seconds_p95.value**(gauge) | 95th percentile end-to-end SGLang request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_e2e_request_latency_seconds_p99.value**(gauge) | 99th percentile end-to-end SGLang request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_generation_throughput.value**(rate) | Generation throughput of the SGLang instance.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_sglang_num_queue_reqs.value**(gauge) | Number of requests waiting in the SGLang instance queue.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_sglang_num_running_reqs.value**(gauge) | Number of requests running on the SGLang instance.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_sglang_num_used_tokens.value**(gauge) | Key-value (KV) cache usage of the SGLang instance in tokens.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_sglang_prompt_throughput.value**(rate) | Prompt throughput of the SGLang instance.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_sglang_time_per_output_token_seconds_avg.value**(gauge) | Average SGLang time per output token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_per_output_token_seconds_p100.value**(gauge) | Maximum SGLang time per output token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_per_output_token_seconds_p95.value**(gauge) | 95th percentile SGLang time per output token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_per_output_token_seconds_p99.value**(gauge) | 99th percentile SGLang time per output token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_to_first_token_seconds_avg.value**(gauge) | Average SGLang time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_to_first_token_seconds_p100.value**(gauge) | Maximum SGLang time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_to_first_token_seconds_p95.value**(gauge) | 95th percentile SGLang time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_time_to_first_token_seconds_p99.value**(gauge) | 99th percentile SGLang time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_sglang_token_usage.value**(gauge) | Key-value (KV) cache token utilization of the SGLang instance.*Shown as percent* | | **alibabacloud.function_compute.function_qualifier_tx_bytes_per_sec.value**(rate) | Outbound network traffic of the function for the specified version or alias.*Shown as byte* | | **alibabacloud.function_compute.function_qualifier_total_invocations.value**(gauge) | Total number of function invocations for the specified version or alias.*Shown as invocation* | | **alibabacloud.function_compute.function_qualifier_vllm_e2e_request_latency_seconds_avg.value**(gauge) | Average end-to-end vLLM request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_e2e_request_latency_seconds_p100.value**(gauge) | Maximum end-to-end vLLM request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_e2e_request_latency_seconds_p95.value**(gauge) | 95th percentile end-to-end vLLM request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_e2e_request_latency_seconds_p99.value**(gauge) | 99th percentile end-to-end vLLM request latency.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_generation_tokens_per_second.value**(rate) | Generation throughput of the vLLM instance.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_gpu_cache_usage_perc.value**(gauge) | Key-value (KV) cache utilization of the vLLM instance.*Shown as percent* | | **alibabacloud.function_compute.function_qualifier_vllm_gpu_prefix_cache_hits_rate.value**(gauge) | GPU prefix cache hit rate of the vLLM instance.*Shown as percent* | | **alibabacloud.function_compute.function_qualifier_vllm_num_requests_running.value**(gauge) | Number of requests running on the vLLM instance.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_vllm_num_requests_waiting.value**(gauge) | Number of requests waiting on the vLLM instance.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_vllm_prompt_tokens_per_second.value**(rate) | Prompt throughput of the vLLM instance.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_decode_time_seconds_avg.value**(gauge) | Average vLLM request decode time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_decode_time_seconds_p95.value**(gauge) | 95th percentile vLLM request decode time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_decode_time_seconds_p99.value**(gauge) | 99th percentile vLLM request decode time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_generation_tokens_avg.value**(gauge) | Average number of generation tokens per vLLM request.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_generation_tokens_p95.value**(gauge) | 95th percentile number of generation tokens per vLLM request.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_generation_tokens_p99.value**(gauge) | 99th percentile number of generation tokens per vLLM request.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_inference_time_seconds_avg.value**(gauge) | Average vLLM request inference time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_inference_time_seconds_p95.value**(gauge) | 95th percentile vLLM request inference time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_inference_time_seconds_p99.value**(gauge) | 99th percentile vLLM request inference time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_params_max_tokens_avg.value**(gauge) | Average value of the max_tokens parameter in vLLM requests.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_params_max_tokens_p95.value**(gauge) | 95th percentile value of the max_tokens parameter in vLLM requests.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_params_max_tokens_p99.value**(gauge) | 99th percentile value of the max_tokens parameter in vLLM requests.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_params_n_avg.value**(gauge) | Average value of the n parameter (completions per request) in vLLM requests.*Shown as unit* | | **alibabacloud.function_compute.function_qualifier_vllm_request_params_n_p95.value**(gauge) | 95th percentile value of the n parameter (completions per request) in vLLM requests.*Shown as unit* | | **alibabacloud.function_compute.function_qualifier_vllm_request_params_n_p99.value**(gauge) | 99th percentile value of the n parameter (completions per request) in vLLM requests.*Shown as unit* | | **alibabacloud.function_compute.function_qualifier_vllm_request_prefill_time_seconds_avg.value**(gauge) | Average vLLM request prefill time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_prefill_time_seconds_p95.value**(gauge) | 95th percentile vLLM request prefill time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_prefill_time_seconds_p99.value**(gauge) | 99th percentile vLLM request prefill time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_prompt_tokens_avg.value**(gauge) | Average number of prompt tokens per vLLM request.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_prompt_tokens_p95.value**(gauge) | 95th percentile number of prompt tokens per vLLM request.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_prompt_tokens_p99.value**(gauge) | 99th percentile number of prompt tokens per vLLM request.*Shown as token* | | **alibabacloud.function_compute.function_qualifier_vllm_request_queue_time_seconds_avg.value**(gauge) | Average vLLM request queue time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_queue_time_seconds_p95.value**(gauge) | 95th percentile vLLM request queue time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_queue_time_seconds_p99.value**(gauge) | 99th percentile vLLM request queue time.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_request_success_per_second.value**(rate) | Number of vLLM requests successfully processed per second, grouped by finish reason.*Shown as request* | | **alibabacloud.function_compute.function_qualifier_vllm_time_per_output_token_seconds_avg.value**(gauge) | Average vLLM latency per generated token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_per_output_token_seconds_p100.value**(gauge) | Maximum vLLM latency per generated token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_per_output_token_seconds_p95.value**(gauge) | 95th percentile vLLM latency per generated token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_per_output_token_seconds_p99.value**(gauge) | 99th percentile vLLM latency per generated token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_to_first_token_seconds_avg.value**(gauge) | Average vLLM time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_to_first_token_seconds_p100.value**(gauge) | Maximum vLLM time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_to_first_token_seconds_p95.value**(gauge) | 95th percentile vLLM time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vllm_time_to_first_token_seconds_p99.value**(gauge) | 99th percentile vLLM time to first token.*Shown as second* | | **alibabacloud.function_compute.function_qualifier_vcpu_quota_cores.value**(gauge) | vCPU quota of the function for the specified version or alias.*Shown as core* | | **alibabacloud.function_compute.function_rx_bytes_per_sec.value**(rate) | Inbound network traffic of the function.*Shown as byte* | | **alibabacloud.function_compute.function_resident_live_instances.average**(gauge) | Number of live resident function instances.*Shown as instance* | | **alibabacloud.function_compute.function_resource_throttles.value**(gauge) | Number of function requests exceeding the total instance limit (error code 503).*Shown as request* | | **alibabacloud.function_compute.function_server_errors.value**(gauge) | Number of function server errors.*Shown as error* | | **alibabacloud.function_compute.function_server_errors_rate.value**(gauge) | Percentage of function invocations that resulted in server errors.*Shown as percent* | | **alibabacloud.function_compute.function_tx_bytes_per_sec.value**(rate) | Outbound network traffic of the function.*Shown as byte* | | **alibabacloud.function_compute.function_throttles.value**(gauge) | Number of function requests exceeding the concurrent instance limit.*Shown as request* | | **alibabacloud.function_compute.function_throttles_rate.value**(gauge) | Percentage of function requests exceeding the concurrent instance limit.*Shown as percent* | | **alibabacloud.function_compute.function_total_invocations.value**(gauge) | Total number of function invocations.*Shown as invocation* | | **alibabacloud.function_compute.instance_avg_vcpu_cores.value**(gauge) | Average number of vCPU cores used by the instance.*Shown as core* | | **alibabacloud.function_compute.instance_concurrent_requests.value**(gauge) | Number of concurrent requests on the instance.*Shown as request* | | **alibabacloud.function_compute.instance_gpu_cards_decoder_percent.maximum**(gauge) | Maximum GPU hardware decoder utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_decoder_percent.average**(gauge) | Average GPU hardware decoder utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_encoder_percent.maximum**(gauge) | Maximum GPU hardware encoder utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_encoder_percent.average**(gauge) | Average GPU hardware encoder utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_max_memory_usage.maximum**(gauge) | Maximum GPU memory used by the instance.*Shown as mebibyte* | | **alibabacloud.function_compute.instance_gpu_cards_max_memory_usage.average**(gauge) | Average GPU memory used by the instance.*Shown as mebibyte* | | **alibabacloud.function_compute.instance_gpu_cards_memory_limit_mb.maximum**(gauge) | Maximum GPU memory quota of the instance.*Shown as mebibyte* | | **alibabacloud.function_compute.instance_gpu_cards_memory_limit_mb.average**(gauge) | Average GPU memory quota of the instance.*Shown as mebibyte* | | **alibabacloud.function_compute.instance_gpu_cards_memory_usage_percent.maximum**(gauge) | Maximum GPU memory utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_memory_usage_percent.average**(gauge) | Average GPU memory utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_sm_percent.maximum**(gauge) | Maximum GPU streaming multiprocessor (SM) utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_gpu_cards_sm_percent.average**(gauge) | Average GPU streaming multiprocessor (SM) utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_max_vcpu_cores.value**(gauge) | Maximum number of vCPU cores used by the instance.*Shown as core* | | **alibabacloud.function_compute.instance_memory_limit_mb.value**(gauge) | Memory quota of the instance.*Shown as mebibyte* | | **alibabacloud.function_compute.instance_memory_usage_mb.value**(gauge) | Memory usage of the instance.*Shown as mebibyte* | | **alibabacloud.function_compute.instance_memory_usage_percent.value**(gauge) | Memory utilization of the instance.*Shown as percent* | | **alibabacloud.function_compute.instance_rx_bytes_per_sec.value**(rate) | Inbound network traffic of the instance.*Shown as byte* | | **alibabacloud.function_compute.instance_tx_bytes_per_sec.value**(rate) | Outbound network traffic of the instance.*Shown as byte* | | **alibabacloud.function_compute.instance_throttles.value**(gauge) | Number of requests throttled because total instances exceeded the limit.*Shown as request* | | **alibabacloud.function_compute.instance_vcpu_quota_cores.value**(gauge) | vCPU quota of the instance.*Shown as core* | | **alibabacloud.function_compute.invocations.value**(gauge) | Number of invocations.*Shown as invocation* | | **alibabacloud.function_compute.invocations_http_status2xx.value**(gauge) | Number of invocations with HTTP status 2xx.*Shown as invocation* | | **alibabacloud.function_compute.invocations_http_status3xx.value**(gauge) | Number of invocations with HTTP status 3xx.*Shown as invocation* | | **alibabacloud.function_compute.invocations_http_status4xx.value**(gauge) | Number of invocations with HTTP status 4xx.*Shown as invocation* | | **alibabacloud.function_compute.invocations_http_status5xx.value**(gauge) | Number of invocations with HTTP status 5xx.*Shown as invocation* | | **alibabacloud.function_compute.latency_avg.value**(gauge) | Average end-to-end latency (synchronous invocations).*Shown as millisecond* | | **alibabacloud.function_compute.latency_max.value**(gauge) | Maximum end-to-end latency (synchronous invocations).*Shown as millisecond* | | **alibabacloud.function_compute.memory_usage_max.value**(gauge) | Maximum memory used.*Shown as mebibyte* | | **alibabacloud.function_compute.ondemand_active_instances.value**(gauge) | Number of instances in standard mode.*Shown as instance* | | **alibabacloud.function_compute.provisioned_active_instances.value**(gauge) | Number of instances in fast mode.*Shown as instance* | | **alibabacloud.function_compute.provisioned_instances.value**(gauge) | Number of snapshot startup instances.*Shown as instance* | | **alibabacloud.function_compute.provisioned_invocations.value**(gauge) | Number of invocations served by provisioned instances.*Shown as invocation* | | **alibabacloud.function_compute.region_billable_invocations.value**(gauge) | Number of billable invocations in the region.*Shown as invocation* | | **alibabacloud.function_compute.region_billable_invocations_rate.value**(gauge) | Percentage of invocations in the region that are billable.*Shown as percent* | | **alibabacloud.function_compute.region_client_errors.value**(gauge) | Number of client errors in the region.*Shown as error* | | **alibabacloud.function_compute.region_client_errors_rate.value**(gauge) | Percentage of invocations in the region that resulted in client errors.*Shown as percent* | | **alibabacloud.function_compute.region_concurrency_limit.value**(gauge) | Maximum number of on-demand instances in the region.*Shown as instance* | | **alibabacloud.function_compute.region_concurrent_count.value**(gauge) | Number of on-demand instances in use in the region.*Shown as instance* | | **alibabacloud.function_compute.region_elastic_active_instances.average**(gauge) | Number of active elastic instances in the region.*Shown as instance* | | **alibabacloud.function_compute.region_elastic_idle_instances.average**(gauge) | Number of idle (light sleep) elastic instances in the region.*Shown as instance* | | **alibabacloud.function_compute.region_elastic_instances_quota.average**(gauge) | Elastic instance quota in the region.*Shown as instance* | | **alibabacloud.function_compute.region_function_errors.value**(gauge) | Number of function errors in the region.*Shown as error* | | **alibabacloud.function_compute.region_function_errors_rate.value**(gauge) | Percentage of invocations in the region that resulted in function errors.*Shown as percent* | | **alibabacloud.function_compute.region_max_concurrency_instances.value**(gauge) | Instance quota in the region.*Shown as instance* | | **alibabacloud.function_compute.region_min_instances.average**(gauge) | Minimum number of elastic instances in the region.*Shown as instance* | | **alibabacloud.function_compute.region_provisioned_current_instance.value**(gauge) | Number of provisioned instances in the region.*Shown as instance* | | **alibabacloud.function_compute.region_resident_live_instances.average**(gauge) | Number of live resident instances in the region.*Shown as instance* | | **alibabacloud.function_compute.region_resource_throttles.value**(gauge) | Number of requests in the region exceeding the total instance limit.*Shown as request* | | **alibabacloud.function_compute.region_server_errors.value**(gauge) | Number of server errors in the region.*Shown as error* | | **alibabacloud.function_compute.region_server_errors_rate.value**(gauge) | Percentage of invocations in the region that resulted in server errors.*Shown as percent* | | **alibabacloud.function_compute.region_throttles.value**(gauge) | Number of requests in the region exceeding the concurrent instance limit.*Shown as request* | | **alibabacloud.function_compute.region_throttles_rate.value**(gauge) | Percentage of requests in the region exceeding the concurrent instance limit.*Shown as percent* | | **alibabacloud.function_compute.region_total_invocations.value**(gauge) | Total number of invocations in the region.*Shown as invocation* | | **alibabacloud.function_compute.server_errors.value**(gauge) | Number of server errors.*Shown as error* | | **alibabacloud.function_compute.service_billable_invocations.value**(gauge) | Number of billable invocations for the service.*Shown as invocation* | | **alibabacloud.function_compute.service_billable_invocations_rate.value**(gauge) | Percentage of invocations for the service that are billable.*Shown as percent* | | **alibabacloud.function_compute.service_client_errors.value**(gauge) | Number of client errors for the service.*Shown as error* | | **alibabacloud.function_compute.service_client_errors_rate.value**(gauge) | Percentage of invocations for the service that resulted in client errors.*Shown as percent* | | **alibabacloud.function_compute.service_concurrent_count.value**(gauge) | Number of on-demand instances in use for the service.*Shown as instance* | | **alibabacloud.function_compute.service_function_errors.value**(gauge) | Number of function errors for the service.*Shown as error* | | **alibabacloud.function_compute.service_provisioned_current_instance.value**(gauge) | Number of provisioned instances for the service.*Shown as instance* | | **alibabacloud.function_compute.service_qualifier_client_errors.value**(gauge) | Number of client errors for the service for the specified version or alias.*Shown as error* | | **alibabacloud.function_compute.service_qualifier_function_errors.value**(gauge) | Number of function errors for the service for the specified version or alias.*Shown as error* | | **alibabacloud.function_compute.service_qualifier_provisioned_current_instance.value**(gauge) | Number of provisioned instances for the service for the specified version or alias.*Shown as instance* | | **alibabacloud.function_compute.service_qualifier_resource_throttles.value**(gauge) | Number of service requests exceeding the total instance limit for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.service_qualifier_server_errors.value**(gauge) | Number of server errors for the service for the specified version or alias.*Shown as error* | | **alibabacloud.function_compute.service_qualifier_throttles.value**(gauge) | Number of service requests exceeding the concurrent instance limit for the specified version or alias.*Shown as request* | | **alibabacloud.function_compute.service_qualifier_total_invocations.value**(gauge) | Total number of invocations for the service for the specified version or alias.*Shown as invocation* | | **alibabacloud.function_compute.service_resource_throttles.value**(gauge) | Number of service requests exceeding the total instance limit.*Shown as request* | | **alibabacloud.function_compute.service_server_errors.value**(gauge) | Number of server errors for the service.*Shown as error* | | **alibabacloud.function_compute.service_server_errors_rate.value**(gauge) | Percentage of invocations for the service that resulted in server errors.*Shown as percent* | | **alibabacloud.function_compute.service_throttles.value**(gauge) | Number of service requests exceeding the concurrent instance limit.*Shown as request* | | **alibabacloud.function_compute.service_throttles_rate.value**(gauge) | Percentage of service requests exceeding the concurrent instance limit.*Shown as percent* | | **alibabacloud.function_compute.service_total_invocations.value**(gauge) | Total number of invocations for the service.*Shown as invocation* | | **alibabacloud.function_compute.session_instance_evicted_count.sum**(gauge) | Number of evicted session instances.*Shown as instance* | ### Events{% #events %} Alibaba Cloud Function Compute does not include any events. ### Service Checks{% #service-checks %} Alibaba Cloud Function Compute does not include any service checks. ## Troubleshooting{% #troubleshooting %} Need help? Contact [Datadog support](https://docs.datadoghq.com/help/).