PMDANVIDIA(1)

💖 pmdanvidia — the nvidia GPU metrics domain agent (PMDA) 💖

okay bestie, welcome to PMDANVIDIA(1), straight outta the General Commands Manual section. this is the whole vibe, no skips, no skimming, we're doing all of it 🫶

---

✨ SYNOPSIS ✨

$PCP_PMDAS_DIR/nvidia/pmdanvidia [-D debug] [-d domain] [-l logfile] [-t interval]

---

🌸 DESCRIPTION 🌸

so pmdanvidia is a Performance Metrics Domain Agent, aka a PMDA (iykyk), and what it does is pull out performance metrics that describe the metrics available on NVIDIA GPU cards — by way of the NVML library. she's a little connector, that's her whole thing 💅

the nvidia PMDA exports metrics that measure gpu activity, memory utilization, fan speed, etc, on NVIDIA Tesla and Quadro cards. quick reality check though: metrics are unlikely to be available for consumer class cards. no cap, don't come for me.

here's a lil rundown of the pmdanvidia command line options, ok:

• -d — this one is absolutely crucial, like genuinely the most important thing here. the performance metrics domain number you specify has to be unique and consistent. meaning: domain should be different for every PMDA on the one host, and the same domain number should be used for the same PMDA on all hosts. don't be chaotic about it 🙈
• -l — location of the log file. by default, a log file named nvidia.log gets written in the current directory of pmcd(1) when pmdanvidia is started, i.e. $PCP_LOG_DIR/pmcd. if the log file cannot be created or is not writable, output is written to the standard error instead. she'll find a way ✨
• -t — enables and sets a sampling interval for automatic refreshing of metric values. this functionality is disabled by default, but this option lets you specify a time interval on which all values are sampled — which has the effect of constantly updating the accumulating metrics, with the goal of assisting client tools such as pcp-atop(1) and pmlogger(1) to observe sub-sample time changes in GPU and process state. because typically these tools have longer sampling intervals, and can thus 'miss' activity happening during their sampling interval. so -t is basically the "bestie keep up" button 🏃♀️💨

---

🎀 INSTALLATION 🎀

the nvidia PMDA is not installed and available by default. she's exclusive, she's invite-only.

if you want to undo the installation, do the following as root:

# cd $PCP_PMDAS_DIR/nvidia
# ./Remove

and if you want to establish access to the names, help text and values for the nvidia performance metrics once more, after removal, do the following as root:

# cd $PCP_PMDAS_DIR/nvidia
# ./Install

pmdanvidia is launched by pmcd(1) and should never be executed directly. the Install and Remove scripts notify pmcd(1) when the agent is installed or removed. do not freelance this, bestie 😭

---

🦋 FILES 🦋

• $PCP_PMCDCONF_PATH — command line options used to launch pmdanvidia
• $PCP_PMDAS_DIR/nvidia/help — default help text file for the nvidia metrics
• $PCP_PMDAS_DIR/nvidia/Install — installation script for the pmdanvidia agent
• $PCP_PMDAS_DIR/nvidia/Remove — undo installation script for the pmdanvidia agent
• $PCP_LOG_DIR/pmcd/nvidia.log — default log file for error messages and other information from pmdanvidia

---

🌈 PCP ENVIRONMENT 🌈

environment variables with the prefix PCP_ are used to parameterize the file and directory names used by PCP. on each installation, the file /etc/pcp.conf contains the local values for these variables. the $PCP_CONF variable may be used to specify an alternative configuration file, as described in pcp.conf(5). the way that config just *works*… obsessed 💜

---

🔥 DEBUGGING OPTIONS 🔥

the -D or --debug option enables the output of additional diagnostics on stderr to help triage problems, although ngl the information is sometimes cryptic and primarily intended to provide guidance for developers rather than for end-users. so if it reads like nonsense, it's not you, it's the vibes 💀

debug is a comma separated list of debugging options; use pmdbg(1) with the -l option to obtain a list of the available debugging options and their meaning.

debugging options specific to pmdanvidia are as follows:

• appl0 — interaction with pmcd(1)
• appl1 — loading the libnvidia-ml DSO
• appl2 — trace select(2) calls

---

👑 SEE ALSO 👑

PCPIntro(1), pcp-atop(1), pmcd(1), pmlogger(1), pcp.conf(5) and pcp.env(5). go read them, they're iconic, periodt 📚✨

---

and that's the whole page, bestie! this one came from Performance Co-Pilot, tagged PCP, page PMDANVIDIA(1) — sending you all the love and all the sub-sample time changes 🩷🧚💖