kepler/hack/config.yaml at main · sustainable-computing-io/kepler · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
log:
  level: debug # debug, info, warn, error (default: info)
  format: text # text or json (default: text)

monitor:
  #  Interval is the monitor's refresh interval. All process that
  #  have a life time (gets spawned and killed within one monitor interval)
  #  less than the interval will be ignored.
  #
  #  NOTE: setting the interval to 0s will disable the monitor refreshes
  interval: 5s

  # Staleness is the duration after the data that is computed by the monitor
  # in each refresh is considered stale and recomputed when requested again.
  #
  # This is especially useful if you have multiple prometheus instances
  # (in a HA) scrapping kepler and the data received by both instances will be
  # the same as long as the scrapes happens within the staleness duration.
  #
  # NOTE: Keep staleness shorter than the monitor interval.
  staleness: 1000ms

  # maximum number of terminated workloads (process, container, VM, pods)
  # to be kept in memory until the data is exported; 0 disables the limit
  maxTerminated: 500

  # minimum energy threshold (in joules) for terminated workloads
  # terminated workloads with energy consumption below this threshold will be filtered out
  minTerminatedEnergyThreshold: 10

host:
  sysfs: /sys # Path to sysfs filesystem (default: /sys)
  procfs: /proc # Path to procfs filesystem (default: /proc)

rapl:
  zones: [] # zones to be enabled, empty enables all default zones

exporter:
  stdout: # stdout exporter related config
    enabled: false # disabled by default

  prometheus: # prometheus exporter related config
    enabled: true
    debugCollectors:
      - go
      - process
    metricsLevel:
      - node
      - process
      - container
      - vm
      - pod

debug: # debug related config
  pprof: # pprof related config
    enabled: true

web:
  configFile: "" # Path to TLS server config file
  listenAddresses: # Web server listen addresses
    - :28282

kube: # kubernetes related config
  enabled: false # enable kubernetes monitoring (default: false)
  config: "" # path to kubeconfig file (optional if running in-cluster)
  nodeName: "" # name of the kubernetes node (required when enabled)
  podInformer:
    mode: kubelet          # "kubelet" (default) or "apiserver"
    pollInterval: 15s      # Poll interval for kubelet mode (default: 15s)

# WARN DO NOT ENABLE THIS IN PRODUCTION - for development / testing only
dev:
  fake-cpu-meter:
    enabled: false
    zones: [] # zones to be enabled, empty enables all default zones

# EXPERIMENTAL FEATURES - These features are experimental and may be unstable
# and are disabled by default
experimental:
  platform:
    redfish:
      enabled: false # Enable experimental Redfish BMC power monitoring
      configFile: hack/redfish.yaml # Path to Redfish BMC configuration file
      nodeName: "" # Node name to use (overrides Kubernetes node name and hostname fallback)
      httpTimeout: 5s # HTTP client timeout for BMC requests (default: 5s)
  hwmon:
    enabled: false # Enable experimental hwmon power monitoring
    zones: [] # List of zones to enable (default enable all)
    # chipRules allows overriding or adding chip pairing rules via configuration.
    # These rules take precedence over hardcoded defaults in the code.
    # Example:
    # chipRules:
    #   # Override an existing chip (replaces hardcoded rule)
    #   - name: "ina3221"
    #     useSameIndex: true
    #     skipVoltages: [4, 5, 6, 7]
    #   # Add a new chip not in the hardcoded list
    #   - name: "my_custom_chip"
    #     pairings:
    #       1: 1  # in1 pairs with curr1
    #       2: 2  # in2 pairs with curr2
    #     skipVoltages: [0]  # Skip shunt voltage
    chipRules: []
  gpu:
    enabled: false # Enable experimental GPU power monitoring
    idlePower: 0 # GPU idle power in Watts (0 = auto-detect)
    dcgmEndpoint: "" # dcgm-exporter metrics URL for MIG (auto-discovered if empty)