diff --git a/data/grid5000/sites/sophia/clusters/esterel4/esterel4.json b/data/grid5000/sites/sophia/clusters/esterel4/esterel4.json index 28296fbef437bdc0b8693fbe2794db1ae0a02711..a43f8d84eeffa37a6fc249b61cc3473f793f9c45 100644 --- a/data/grid5000/sites/sophia/clusters/esterel4/esterel4.json +++ b/data/grid5000/sites/sophia/clusters/esterel4/esterel4.json @@ -49,12 +49,39 @@ "port": 9100, "protocol": "prometheus" } + }, + { + "description": "Default subset of metrics from Prometheus Nvidia DCGM Exporter", + "name": "prom_nvgpu_default_metrics", + "optional_period": 15000, + "period": 0, + "source": { + "id": [ + "DCGM_FI_DEV_SM_CLOCK", + "DCGM_FI_DEV_MEM_CLOCK", + "DCGM_FI_DEV_GPU_TEMP", + "DCGM_FI_DEV_POWER_USAGE", + "DCGM_FI_DEV_MEM_COPY_UTIL" + ], + "port": 9400, + "protocol": "prometheus" + } + }, + { + "description": "All metrics from Prometheus Nvidia DCGM Exporter", + "name": "prom_nvgpu_all_metrics", + "optional_period": 15000, + "period": 0, + "source": { + "port": 9400, + "protocol": "prometheus" + } } ], "model": "Dell PowerEdge T630", "nodes_count": 1, - "nodes_description": "1 CPU Unknown vendor, 72 cores/CPU, 0GB RAM, 2x0GB SSD, 1 x 1Gb Ethernet", - "priority": 201606, + "nodes_description": "2 CPUs Intel Xeon E5-2630 v4, 10 cores/CPU, 4 GPUs GeForce GTX TITAN X, 128GB RAM, 1862GB HDD, 1489GB HDD, 1 x 1Gb Ethernet, 1 x 40Gb InfiniBand", + "priority": 201706, "queues": [ "admin", "testing" diff --git a/data/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.json b/data/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.json index afb5eee397b3927f1c06615a3088e505f99140ca..3d6f342b02950819e6dc846b82e5a8d5f1a92159 100644 --- a/data/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.json +++ b/data/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.json @@ -1,26 +1,97 @@ { "architecture": { - "cpu_core_numbering": "contiguous", - "nb_cores": 72, - "nb_procs": 1, - "nb_threads": 72, + "cpu_core_numbering": "round-robin", + "nb_cores": 20, + "nb_procs": 2, + "nb_threads": 40, "platform_type": "x86_64" }, "bios": { - "release_date": "01/01/2000", - "vendor": "Unknown", - "version": 1 + "release_date": "10/18/2023", + "vendor": "Dell Inc.", + "version": "2.18.2" }, - "bmc_version": "v1", + "bmc_version": "unknown", "chassis": { "manufactured_at": "2016-06-08", - "manufacturer": "Unknown", - "name": "Unknown", + "manufacturer": "Dell Inc.", + "name": "PowerEdge T630", + "serial": "HHTDDD2", "warranty_end": "2021-06-08" }, "exotic": false, + "gpu_devices": { + "nvidia0": { + "compute_capability": "5.2", + "cores": 3072, + "cpu_affinity": 0, + "device": "/dev/nvidia0", + "memory": 12884901888, + "microarchitecture": "Maxwell", + "model": "GeForce GTX TITAN X", + "performance": { + "fp-16": 0, + "fp-32": 66050000000000, + "fp-64": 206500000000 + }, + "power_default_limit": "250.00 W", + "vbios_version": "84.00.45.00.03", + "vendor": "Nvidia" + }, + "nvidia1": { + "compute_capability": "5.2", + "cores": 3072, + "cpu_affinity": 0, + "device": "/dev/nvidia1", + "memory": 12884901888, + "microarchitecture": "Maxwell", + "model": "GeForce GTX TITAN X", + "performance": { + "fp-16": 0, + "fp-32": 66050000000000, + "fp-64": 206500000000 + }, + "power_default_limit": "250.00 W", + "vbios_version": "84.00.45.00.03", + "vendor": "Nvidia" + }, + "nvidia2": { + "compute_capability": "5.2", + "cores": 3072, + "cpu_affinity": 1, + "device": "/dev/nvidia2", + "memory": 12884901888, + "microarchitecture": "Maxwell", + "model": "GeForce GTX TITAN X", + "performance": { + "fp-16": 0, + "fp-32": 66050000000000, + "fp-64": 206500000000 + }, + "power_default_limit": "250.00 W", + "vbios_version": "84.00.45.00.03", + "vendor": "Nvidia" + }, + "nvidia3": { + "compute_capability": "5.2", + "cores": 3072, + "cpu_affinity": 1, + "device": "/dev/nvidia3", + "memory": 12884901888, + "microarchitecture": "Maxwell", + "model": "GeForce GTX TITAN X", + "performance": { + "fp-16": 0, + "fp-32": 66050000000000, + "fp-64": 206500000000 + }, + "power_default_limit": "250.00 W", + "vbios_version": "84.00.45.00.03", + "vendor": "Nvidia" + } + }, "main_memory": { - "ram_size": 8 + "ram_size": 137438953472 }, "management_tools": { "bmc_vendor_tool": "racadm", @@ -30,29 +101,108 @@ }, "memory_devices": [ { - "device": "dimm_proc 1 dimm 1", - "size": 8, + "device": "dimm_a1", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a2", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a3", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a4", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b1", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b2", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b3", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b4", + "size": 17179869184, "technology": "dram" } ], "network_adapters": [ { "device": "eth0", - "driver": "mlx_core", + "driver": "igb", "enabled": true, + "firmware_version": "1.67, 0x80000fae, 19.5.12", "interface": "Ethernet", "ip": "172.16.131.4", "ip6": "2001:660:4406:800:4::4", "kavlan": false, "mac": "18:66:da:55:21:c8", "management": false, + "model": "I350 Gigabit Network Connection", "mountable": true, "mounted": true, "name": "enp1s0f0np0", "network_address": "esterel4-1.sophia.grid5000.fr", "rate": 1000000000, + "sriov": false, + "sriov_totalvfs": 0, "switch": "sw-6", - "switch_port": "1/1/3" + "switch_port": "1/1/3", + "vendor": "Intel" + }, + { + "device": "eth1", + "driver": "igb", + "enabled": false, + "firmware_version": "1.67, 0x80000fae, 19.5.12", + "interface": "Ethernet", + "kavlan": false, + "mac": "18:66:da:55:21:c9", + "management": false, + "model": "I350 Gigabit Network Connection", + "mountable": false, + "mounted": false, + "name": "eno2", + "sriov": false, + "sriov_totalvfs": 0, + "vendor": "Intel" + }, + { + "device": "ib0", + "driver": "mlx4_core", + "enabled": true, + "firmware_version": "2.35.5100", + "guid": "0xe41d2d030050b8c1", + "interface": "InfiniBand", + "ip": "172.18.131.4", + "kavlan": false, + "mac": "e4:1d:2d:03:00:50:b8:c1", + "management": false, + "model": "MT27500 Family [ConnectX-3]", + "mountable": true, + "mounted": true, + "name": "ibp130s0", + "netmask": "255.255.240.0", + "network_address": "esterel4-1-ib0.sophia.grid5000.fr", + "rate": 40000000000, + "sriov": false, + "sriov_totalvfs": 0, + "vendor": "Mellanox Technologies" }, { "device": "bmc", @@ -69,59 +219,61 @@ ], "nodeset": "esterel4", "operating_system": { - "cstate_driver": "unknown", - "cstate_governor": "unknown", + "cstate_driver": "intel_idle", + "cstate_governor": "menu", "ht_enabled": true, - "pstate_driver": "unknwon", - "pstate_governor": "unknown", + "pstate_driver": "intel_cpufreq", + "pstate_governor": "performance", "turboboost_enabled": true }, "performance": { "core_flops": 35200000000, - "node_flops": 2534400000000 + "node_flops": 704000000000 }, "processor": { "cache_l1": null, - "cache_l1d": 8, - "cache_l1i": 8, - "cache_l2": 8, - "cache_l3": 8, + "cache_l1d": 32768, + "cache_l1i": 32768, + "cache_l2": 262144, + "cache_l3": 26214400, "clock_speed": 2200000000, "ht_capable": true, "instruction_set": "x86-64", "microarchitecture": "Haswell", - "microcode": "0xd000001", - "model": "Unknown", - "other_description": "description", - "vendor": "vendor", - "version": "vendor" + "microcode": "0xb000040", + "model": "Intel Xeon", + "other_description": "Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz", + "vendor": "Intel", + "version": "E5-2630 v4" }, "redfish": true, "software": { "forced-deployment-timestamp": 202007300948, - "postinstall-version": "1.2025021810", + "postinstall-version": "1.2025032008", "standard-environment": "debian11-x64-std" }, "storage_devices": [ { - "by_id": "", + "by_id": "/dev/disk/by-id/wwn-0x61866da0578978001ef5b04c0f56fda4", "by_path": "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0", + "firmware_version": 4.3, "id": "disk0", "interface": "SAS", - "model": "unknown", - "size": 8, - "storage": "SSD", - "vendor": "Unknown" + "model": "PERC H730 Adp", + "size": 1999307276288, + "storage": "HDD", + "vendor": "Dell" }, { - "by_id": "", + "by_id": "/dev/disk/by-id/wwn-0x61866da0578978001efaccdb8e9c9a3e", "by_path": "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0", + "firmware_version": 4.3, "id": "disk1", "interface": "SATA", - "model": "unknown", - "size": 8, - "storage": "SSD", - "vendor": "Unknown" + "model": "PERC H730 Adp", + "size": 1599741100032, + "storage": "HDD", + "vendor": "Dell" } ], "supported_job_types": { @@ -131,7 +283,8 @@ "queues": [ "admin", "testing" - ] + ], + "virtual": "ivt" }, "type": "node", "uid": "esterel4-1" diff --git a/input/grid5000/dell-product-data.yaml b/input/grid5000/dell-product-data.yaml index 1d245811c150e0eb8e0c36236c9ef9fca6104e20..dff68ffa4c3e3d7639217b1e97220742fc186240 100644 --- a/input/grid5000/dell-product-data.yaml +++ b/input/grid5000/dell-product-data.yaml @@ -2238,6 +2238,12 @@ sites: chassis: manufactured_at: 2016-05-04 warranty_end: 2021-05-04 + esterel4: + nodes: + esterel4-1: + chassis: + manufactured_at: 2016-06-08 + warranty_end: 2021-06-08 esterel5: nodes: esterel5-1: diff --git a/input/grid5000/ipv4.yaml b/input/grid5000/ipv4.yaml index a4f41c9f7bbf4d7372757a731b48281bca3a5a36..4ccf9df6705ae8708bcad6cc314affa8d484edb9 100644 --- a/input/grid5000/ipv4.yaml +++ b/input/grid5000/ipv4.yaml @@ -160,6 +160,7 @@ ipv4: sophia esterel2 ib0 0 0 3 1 sophia esterel3 eth0 0 0 3 2 sophia esterel4 eth0 0 0 3 3 + sophia esterel4 ib0 0 0 3 3 sophia esterel5 eth0 0 0 3 4 sophia esterel5 ib0 0 0 3 4 sophia esterel6 eth0 0 0 3 6 diff --git a/input/grid5000/sites/sophia/clusters/esterel4/esterel4.yaml b/input/grid5000/sites/sophia/clusters/esterel4/esterel4.yaml index b28df4e29bfa45f5bc4715e6c162e48fe5a72e44..7e3b1ba2b260c46ac9d414e94c86fd928bb52ebd 100644 --- a/input/grid5000/sites/sophia/clusters/esterel4/esterel4.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel4/esterel4.yaml @@ -9,9 +9,6 @@ queues: - testing nodes: esterel4-1: - chassis: - manufactured_at: 2016-06-08 - warranty_end: 2021-06-08 supported_job_types: deploy: true besteffort: true @@ -29,15 +26,17 @@ nodes: enabled: true mountable: true mounted: true + eth1: + enabled: false + mountable: false + mounted: false storage_devices: - disk0: + pci-0000:03:00.0-scsi-0:2:0:0: id: disk0 interface: SAS - by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0" - disk1: + pci-0000:03:00.0-scsi-0:2:1:0: id: disk1 interface: SATA - by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0" software: standard-environment: debian11-x64-std management_tools: diff --git a/input/grid5000/sites/sophia/clusters/esterel4/esterel4_ib.yaml b/input/grid5000/sites/sophia/clusters/esterel4/esterel4_ib.yaml new file mode 100644 index 0000000000000000000000000000000000000000..e0f6c8f31765cf9892a32e3dfa6e79c6780173f3 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel4/esterel4_ib.yaml @@ -0,0 +1,9 @@ +--- +nodes: + esterel4-1: + network_adapters: + ib0: + enabled: true + mountable: true + mounted: true + netmask: 255.255.240.0 \ No newline at end of file diff --git a/input/grid5000/sites/sophia/clusters/esterel4/nodes.yaml.erb b/input/grid5000/sites/sophia/clusters/esterel4/nodes.yaml.erb deleted file mode 100644 index a737e315869d7673d82743eb068fdfc25b608d59..0000000000000000000000000000000000000000 --- a/input/grid5000/sites/sophia/clusters/esterel4/nodes.yaml.erb +++ /dev/null @@ -1,84 +0,0 @@ -<% - # File generated by 'rake mass:create SRC=/home/npaulin/Downloads/mon-csv - # If changes are needed, it might be better to edit the source data and regenerate using 'rake mass:create' - - cluster_name = "esterel4" - nodes_number = 1 # Size of the cluster (number of nodes) - # MAC addresses declaration - mac_eth0_list = %w( -18:66:da:55:21:c8 - ) - mac_bmc_list = %w( -18:66:da:55:21:ca - ) -%> ---- -nodes: -<% (1..nodes_number).each { |i| %> - <%= cluster_name %>-<%= i %>: - architecture: - nb_procs: 1 # Fake data, will be replaced by g5k-checks - nb_cores: 72 # Fake data, will be replaced by g5k-checks - nb_threads: 72 # Fake data, will be replaced by g5k-checks - platform_type: x86_64 # Fake data, will be replaced by g5k-checks - cpu_core_numbering: contiguous # Fake data, will be replaced by g5k-checks - bios: - release_date: 01/01/2000 # Fake date, will be replaced by g5k-checks - vendor: Unknown # Fake vendor, will be replaced by g5k-checks - version: 1 # Fake version, will be replaced by g5k-checks - bmc_version: v1 # Fake version, will be replaced by g5k-checks - chassis: - manufacturer: Unknown # Fake manufacturer, will be replaced by g5k-checks - name: Unknown # Fake name, will be replaced by g5k-checks - main_memory: - ram_size: 8 # Fake size, will be replaced by g5k-checks - memory_devices: - dimm: - size: 8 # Fake size, will be replaced by g5k-checks - technology: dram # Common memory technology, will be replaced by g5k-checks - processor: - model: Unknown # Fake model name, will be replaced by g5k-checks - other_description: description # Fake description, will be replaced by g5k-checks - vendor: vendor # Fake vendor, will be replaced by g5k-checks - version: vendor # Fake version, will be replaced by g5k-checks - cache_l1d: 8 # Fake cache, will be replaced by g5k-checks - cache_l1i: 8 # Fake cache, will be replaced by g5k-checks - cache_l2: 8 # Fake cache, will be replaced by g5k-checks - cache_l3: 8 # Fake cache, will be replaced by g5k-checks - instruction_set: x86-64 # Common instruction set, will be replaced by g5k-checks - microcode: "0xd000001" # Fake microcode, will be replaced by g5k-checks - ht_capable: true # Default ht capable value, will be replaced by g5k-checks - main_memory: - ram_size: 8 # Fake ram size, will be replaced by g5k-checks - memory_devices: - dimm_proc 1 dimm 1: # Fake dimm name, will be replaced by g5k-checks - size: 8 # Fake dimm size, will be replaced by g5k-checks - technology: dram # Default dimm technology, will be replaced by g5k-checks - operating_system: - cstate_driver: unknown # Fake driver, will be replaced by g5k-checks - cstate_governor: unknown # Fake governor, will be replaced by g5k-checks - ht_enabled: true # common value for hyper threading, will be replaced by g5k-checks - pstate_driver: unknwon # Fake driver, will be replaced by g5k-checks - pstate_governor: unknown # Fake driver, will be replaced by g5k-checks - turboboost_enabled: true # Default value for turboboost, will be replaced by g5k-checks - network_adapters: - bmc: - management: true - mac: <%= mac_bmc_list[i - 1] %> - eth0: - interface: Ethernet - management: false - driver: mlx_core # Fake data, will be replaced by g5k-checks - name: enp1s0f0np0 # Fake data, will be replaced by g5k-checks - rate: 1000000000 # Fake data, will be replaced by g5k-checks - mac: <%= mac_eth0_list[i - 1] %> - storage_devices: - disk0: - storage: SSD # Fake data, will be replaced by g5k-checks - model: unknown # Fake data, will be replaced by g5k-checks - size: 8 # Fake data, will be replaced by g5k-checks - disk1: - storage: SSD # Fake data, will be replaced by g5k-checks - model: unknown # Fake data, will be replaced by g5k-checks - size: 8 # Fake data, will be replaced by g5k-checks -<% } %> diff --git a/input/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.yaml b/input/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.yaml new file mode 100644 index 0000000000000000000000000000000000000000..1ef4ce76d42eadc3857085a99299e228359c9825 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel4/nodes/esterel4-1.yaml @@ -0,0 +1,156 @@ +# Generated by g5k-checks (g5k-checks -m api) +--- +esterel4-1: + architecture: + cpu_core_numbering: round-robin + nb_cores: 20 + nb_procs: 2 + nb_threads: 40 + platform_type: x86_64 + bios: + release_date: 10/18/2023 + vendor: Dell Inc. + version: 2.18.2 + bmc_version: unknown + chassis: + manufacturer: Dell Inc. + name: PowerEdge T630 + serial: HHTDDD2 + gpu_devices: + nvidia0: + cpu_affinity: 0 + device: "/dev/nvidia0" + memory: 12884901888 + model: GeForce GTX TITAN X + power_default_limit: 250.00 W + vbios_version: 84.00.45.00.03 + vendor: Nvidia + nvidia1: + cpu_affinity: 0 + device: "/dev/nvidia1" + memory: 12884901888 + model: GeForce GTX TITAN X + power_default_limit: 250.00 W + vbios_version: 84.00.45.00.03 + vendor: Nvidia + nvidia2: + cpu_affinity: 1 + device: "/dev/nvidia2" + memory: 12884901888 + model: GeForce GTX TITAN X + power_default_limit: 250.00 W + vbios_version: 84.00.45.00.03 + vendor: Nvidia + nvidia3: + cpu_affinity: 1 + device: "/dev/nvidia3" + memory: 12884901888 + model: GeForce GTX TITAN X + power_default_limit: 250.00 W + vbios_version: 84.00.45.00.03 + vendor: Nvidia + main_memory: + ram_size: 137438953472 + memory_devices: + dimm_a1: + size: 17179869184 + technology: dram + dimm_a2: + size: 17179869184 + technology: dram + dimm_a3: + size: 17179869184 + technology: dram + dimm_a4: + size: 17179869184 + technology: dram + dimm_b1: + size: 17179869184 + technology: dram + dimm_b2: + size: 17179869184 + technology: dram + dimm_b3: + size: 17179869184 + technology: dram + dimm_b4: + size: 17179869184 + technology: dram + network_adapters: + bmc: + ip: 172.17.131.4 + mac: 18:66:da:55:21:ca + management: true + eth0: + driver: igb + firmware_version: 1.67, 0x80000fae, 19.5.12 + interface: Ethernet + ip: 172.16.131.4 + mac: 18:66:da:55:21:c8 + management: false + model: I350 Gigabit Network Connection + name: enp1s0f0np0 + rate: 1000000000 + sriov: false + sriov_totalvfs: 0 + vendor: Intel + eth1: + driver: igb + firmware_version: 1.67, 0x80000fae, 19.5.12 + interface: Ethernet + mac: 18:66:da:55:21:c9 + management: false + model: I350 Gigabit Network Connection + name: eno2 + sriov: false + sriov_totalvfs: 0 + vendor: Intel + ib0: + driver: mlx4_core + firmware_version: 2.35.5100 + guid: '0xe41d2d030050b8c1' + interface: InfiniBand + mac: e4:1d:2d:03:00:50:b8:c1 + management: false + model: MT27500 Family [ConnectX-3] + name: ibp130s0 + rate: 40000000000 + sriov: false + sriov_totalvfs: 0 + vendor: Mellanox Technologies + operating_system: + cstate_driver: intel_idle + cstate_governor: menu + ht_enabled: true + pstate_driver: intel_cpufreq + pstate_governor: performance + turboboost_enabled: true + processor: + cache_l1d: 32768 + cache_l1i: 32768 + cache_l2: 262144 + cache_l3: 26214400 + ht_capable: true + instruction_set: x86-64 + microcode: '0xb000040' + model: Intel Xeon + other_description: Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz + vendor: Intel + version: E5-2630 v4 + storage_devices: + pci-0000:03:00.0-scsi-0:2:0:0: + by_id: "/dev/disk/by-id/wwn-0x61866da0578978001ef5b04c0f56fda4" + by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0" + firmware_version: 4.3 + model: PERC H730 Adp + size: 1999307276288 + storage: HDD + pci-0000:03:00.0-scsi-0:2:1:0: + by_id: "/dev/disk/by-id/wwn-0x61866da0578978001efaccdb8e9c9a3e" + by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0" + firmware_version: 4.3 + model: PERC H730 Adp + size: 1599741100032 + storage: HDD + supported_job_types: + virtual: ivt diff --git a/lib/refrepo/gpu_ref.rb b/lib/refrepo/gpu_ref.rb index ea667843943a645a6fd7dd90ec4b2ed04bb5f599..48874d4fc4bcd1b6b57153a4c5a5deb611e862a7 100644 --- a/lib/refrepo/gpu_ref.rb +++ b/lib/refrepo/gpu_ref.rb @@ -140,6 +140,18 @@ class GPURef 'fp-64' => 342900000000, }, }, + 'GeForce GTX TITAN X' => { + 'cores' => 3072, + 'compute_capability' => '5.2', + 'short_name' => 'GTX TITAN X', + 'alias' => 'gtxtitanx', + 'microarchitecture' => 'Maxwell', + 'performance' => { + 'fp-16' => 0, + 'fp-32' => 66050000000000, + 'fp-64' => 206500000000, + }, + }, 'Tesla K40m' => { 'cores' => 2880, 'compute_capability' => '3.5',