ClusterCockpit/cc-metric-collector
1
0
Fork 0
mirror of https://github.com/ClusterCockpit/cc-metric-collector.git synced 2025-10-31 09:05:05 +01:00
Code Issues Packages Projects Releases Wiki Activity

Add derived_values for numastats (#134)

Browse Source 
* Check creation of CCMessage in NATS receiver

* add derived_values for numastats

* change to ccMessage

* remove vim command artefact

---------

Co-authored-by: Thomas Roehl <thomas.roehl@fau.de>
Co-authored-by: exterr2f <Robert.Externbrink@rub.de>
Co-authored-by: Thomas Gruber <Thomas.Roehl@googlemail.com>
This commit is contained in:
brinkcoder
2025-02-19 11:35:32 +01:00
committed by GitHub
parent f2f38c81af
commit 0e57c8db1c
2 changed files with 62 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

71
collectors/numastatsMetric.go
View File

@@ -11,9 +11,14 @@ import (
"time"
cclog "github.com/ClusterCockpit/cc-metric-collector/pkg/ccLogger"
lp "github.com/ClusterCockpit/cc-energy-manager/pkg/cc-message"
lp "github.com/ClusterCockpit/cc-lib/ccMessage"
)
type NUMAStatsCollectorConfig struct {
SendAbsoluteValues bool `json:"send_abs_values"`
SendDerivedValues bool `json:"send_derived_values"`
}
// Non-Uniform Memory Access (NUMA) policy hit/miss statistics
//
// numa_hit:
@@ -47,13 +52,16 @@ import (
//
// See: https://www.kernel.org/doc/html/latest/admin-guide/numastat.html
type NUMAStatsCollectorTopolgy struct {
file string
tagSet map[string]string
file string
tagSet map[string]string
previousValues map[string]int64
}
type NUMAStatsCollector struct {
metricCollector
topology []NUMAStatsCollectorTopolgy
topology []NUMAStatsCollectorTopolgy
config NUMAStatsCollectorConfig
lastTimestamp time.Time
}
func (m *NUMAStatsCollector) Init(config json.RawMessage) error {
@@ -86,8 +94,9 @@ func (m *NUMAStatsCollector) Init(config json.RawMessage) error {
file := filepath.Join(dir, "numastat")
m.topology = append(m.topology,
NUMAStatsCollectorTopolgy{
file: file,
tagSet: map[string]string{"memoryDomain": node},
file: file,
tagSet: map[string]string{"memoryDomain": node},
previousValues: make(map[string]int64),
})
}
@@ -102,23 +111,27 @@ func (m *NUMAStatsCollector) Read(interval time.Duration, output chan lp.CCMessa
return
}
now := time.Now()
timeDiff := now.Sub(m.lastTimestamp).Seconds()
m.lastTimestamp = now
for i := range m.topology {
// Loop for all NUMA domains
t := &m.topology[i]
now := time.Now()
file, err := os.Open(t.file)
if err != nil {
cclog.ComponentError(
m.name,
fmt.Sprintf("Read(): Failed to open file '%s': %v", t.file, err))
return
continue
}
scanner := bufio.NewScanner(file)
// Read line by line
for scanner.Scan() {
split := strings.Fields(scanner.Text())
line := scanner.Text()
split := strings.Fields(line)
if len(split) != 2 {
continue
}
@@ -130,18 +143,38 @@ func (m *NUMAStatsCollector) Read(interval time.Duration, output chan lp.CCMessa
fmt.Sprintf("Read(): Failed to convert %s='%s' to int64: %v", key, split[1], err))
continue
}
y, err := lp.NewMessage(
"numastats_"+key,
t.tagSet,
m.meta,
map[string]interface{}{"value": value},
now,
)
if err == nil {
output <- y
if m.config.SendAbsoluteValues {
msg, err := lp.NewMessage(
"numastats_"+key,
t.tagSet,
m.meta,
map[string]interface{}{"value": value},
now,
)
if err == nil {
output <- msg
}
}
if m.config.SendDerivedValues {
prev, ok := t.previousValues[key]
if ok {
rate := float64(value-prev) / timeDiff
msg, err := lp.NewMessage(
"numastats_"+key+"_rate",
t.tagSet,
m.meta,
map[string]interface{}{"value": rate},
now,
)
if err == nil {
output <- msg
}
}
t.previousValues[key] = value
}
}
file.Close()
}
}

11
collectors/numastatsMetric.md
View File

@@ -2,7 +2,10 @@
## `numastat` collector
```json
"numastats": {}
"numastats": {
"send_abs_values" : true,
"send_derived_values" : true
}
```
The `numastat` collector reads data from `/sys/devices/system/node/node*/numastat` and outputs a handful **memoryDomain** metrics. See: <https://www.kernel.org/doc/html/latest/admin-guide/numastat.html>
@@ -15,3 +18,9 @@ Metrics:
* `numastats_local_node`: A process ran on this node's CPU, and got memory from this node.
* `numastats_other_node`: A process ran on a different node's CPU, and got memory from this node.
* `numastats_interleave_hit`: Interleaving wanted to allocate from this node and succeeded.
* `numastats_numa_hit_rate` (if `send_derived_values == true`): Derived rate value per second.
* `numastats_numa_miss_rate` (if `send_derived_values == true`): Derived rate value per second.
* `numastats_numa_foreign_rate` (if `send_derived_values == true`): Derived rate value per second.
* `numastats_local_node_rate` (if `send_derived_values == true`): Derived rate value per second.
* `numastats_other_node_rate` (if `send_derived_values == true`): Derived rate value per second.
* `numastats_interleave_hit_rate` (if `send_derived_values == true`): Derived rate value per second.