From 771cc8b16e64b31092af1ee7dc99e657ebdc587e Mon Sep 17 00:00:00 2001 From: Laurent Pouilloux <laurent.pouilloux@inria.fr> Date: Fri, 21 Mar 2025 12:52:27 +0100 Subject: [PATCH] [sophia][mercantour7] add data from g5k-check-import + generate ib0 ip --- .../clusters/mercantour7/mercantour7.json | 2 +- .../mercantour7/nodes/mercantour7-1.json | 245 +++++++++++++++--- input/grid5000/ipv4.yaml | 1 + .../clusters/mercantour7/mercantour7.yaml | 19 +- .../clusters/mercantour7/nodes.yaml.erb | 84 ------ .../mercantour7/nodes/mercantour7-1.yaml | 171 ++++++++++++ 6 files changed, 390 insertions(+), 132 deletions(-) delete mode 100644 input/grid5000/sites/sophia/clusters/mercantour7/nodes.yaml.erb create mode 100644 input/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.yaml diff --git a/data/grid5000/sites/sophia/clusters/mercantour7/mercantour7.json b/data/grid5000/sites/sophia/clusters/mercantour7/mercantour7.json index 58ea8739b3f..b29076e492f 100644 --- a/data/grid5000/sites/sophia/clusters/mercantour7/mercantour7.json +++ b/data/grid5000/sites/sophia/clusters/mercantour7/mercantour7.json @@ -53,7 +53,7 @@ ], "model": "Dell PowerEdge R7525", "nodes_count": 1, - "nodes_description": "1 CPU Unknown vendor, 72 cores/CPU, 0GB RAM, 2x0GB SSD, 1 x 10Gb Ethernet", + "nodes_description": "2 CPUs AMD EPYC 7502, 32 cores/CPU, 384GB RAM, 893GB SSD, 44709GB HDD, 1 x 1Gb Ethernet, 1 x 40Gb InfiniBand", "priority": 197001, "queues": [ "admin", diff --git a/data/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.json b/data/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.json index 608e6b296dd..fba1dc4135b 100644 --- a/data/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.json +++ b/data/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.json @@ -1,26 +1,27 @@ { "architecture": { "cpu_core_numbering": "contiguous", - "nb_cores": 72, - "nb_procs": 1, - "nb_threads": 72, + "nb_cores": 64, + "nb_procs": 2, + "nb_threads": 128, "platform_type": "x86_64" }, "bios": { - "release_date": "01/01/2000", - "vendor": "Unknown", - "version": 1 + "release_date": "09/12/2023", + "vendor": "Dell Inc.", + "version": "2.13.3" }, - "bmc_version": "v1", + "bmc_version": "7.00.60.00", "chassis": { "manufactured_at": "1970-01-01", - "manufacturer": "Unknown", - "name": "Unknown", + "manufacturer": "Dell Inc.", + "name": "PowerEdge R7525", + "serial": "JN4LS73", "warranty_end": "1970-01-01" }, "exotic": false, "main_memory": { - "ram_size": 8 + "ram_size": 412316860416 }, "management_tools": { "bmc_vendor_tool": "racadm", @@ -30,29 +31,188 @@ }, "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_a5", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a6", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a7", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a8", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a9", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a10", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a11", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_a12", + "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" + }, + { + "device": "dimm_b5", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b6", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b7", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b8", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b9", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b10", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b11", + "size": 17179869184, + "technology": "dram" + }, + { + "device": "dimm_b12", + "size": 17179869184, "technology": "dram" } ], "network_adapters": [ { "device": "eth0", - "driver": "mlx_core", + "driver": "tg3", "enabled": true, + "firmware_version": "FFV22.71.3 bc 5720-v1.39", "interface": "Ethernet", "ip": "172.16.130.47", "ip6": "2001:660:4406:800:3::2f", "kavlan": false, "mac": "f4:02:70:b9:60:34", "management": false, + "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", "mountable": true, "mounted": true, "name": "enp1s0f0np0", "network_address": "mercantour7-1.sophia.grid5000.fr", - "rate": 10000000000, + "rate": 1000000000, + "sriov": false, + "sriov_totalvfs": 0, "switch": "sw-3", - "switch_port": "1/1/19" + "switch_port": "1/1/19", + "vendor": "Broadcom Inc. and subsidiaries" + }, + { + "device": "eth1", + "driver": "tg3", + "enabled": false, + "firmware_version": "FFV22.71.3 bc 5720-v1.39", + "interface": "Ethernet", + "kavlan": false, + "mac": "f4:02:70:b9:60:35", + "management": false, + "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "mountable": false, + "mounted": false, + "name": "eno2", + "sriov": false, + "sriov_totalvfs": 0, + "vendor": "Broadcom Inc. and subsidiaries" + }, + { + "device": "ib0", + "driver": "ib_ipoib", + "enabled": true, + "firmware_version": "2.42.5000", + "guid": "0x1c34da0300e46d11", + "interface": "InfiniBand", + "ip": "172.18.130.47", + "kavlan": false, + "mac": "80:00:02:08:fe:80:00:00:00:00:00:00:1c:34:da:03:00:e4:6d:11", + "management": false, + "model": "MT27520 Family [ConnectX-3 Pro]", + "mountable": true, + "mounted": true, + "name": "ibp161s0", + "netmask": "255.255.240.0", + "network_address": "mercantour7-1-ib0.sophia.grid5000.fr", + "rate": 40000000000, + "sriov": true, + "sriov_totalvfs": 8, + "vendor": "Mellanox Technologies" }, { "device": "bmc", @@ -69,32 +229,32 @@ ], "nodeset": "mercantour7", "operating_system": { - "cstate_driver": "unknown", - "cstate_governor": "unknown", + "cstate_driver": "acpi_idle", + "cstate_governor": "menu", "ht_enabled": true, - "pstate_driver": "unknwon", - "pstate_governor": "unknown", + "pstate_driver": "acpi-cpufreq", + "pstate_governor": "performance", "turboboost_enabled": true }, "performance": { "core_flops": 40000000000, - "node_flops": 2880000000000 + "node_flops": 2560000000000 }, "processor": { "cache_l1": null, - "cache_l1d": 8, - "cache_l1i": 8, - "cache_l2": 8, - "cache_l3": 8, + "cache_l1d": 32768, + "cache_l1i": 32768, + "cache_l2": 524288, + "cache_l3": 16777216, "clock_speed": 2500000000, "ht_capable": true, "instruction_set": "x86-64", "microarchitecture": "Zen 2", - "microcode": "0xd000001", - "model": "Unknown", - "other_description": "description", - "vendor": "vendor", - "version": "vendor" + "microcode": "0x830107c", + "model": "AMD EPYC", + "other_description": "AMD EPYC 7502 32-Core Processor", + "vendor": "AMD", + "version": 7502 }, "redfish": true, "software": { @@ -104,24 +264,26 @@ }, "storage_devices": [ { - "by_id": "", + "by_id": "/dev/disk/by-id/wwn-0x670b5e80db81ca0027c136530350d932", "by_path": "/dev/disk/by-path/pci-0000:41:00.0-scsi-0:2:0:0", + "firmware_version": 5.16, "id": "disk0", - "interface": "SAS", - "model": "unknown", - "size": 8, + "interface": "RAID", + "model": "PERC H745 Adp", + "size": 959656755200, "storage": "SSD", - "vendor": "Unknown" + "vendor": "Dell" }, { - "by_id": "", + "by_id": "/dev/disk/by-id/wwn-0x670b5e80db81ca0027c13b3347a1dd0f", "by_path": "/dev/disk/by-path/pci-0000:41:00.0-scsi-0:2:1:0", + "firmware_version": 5.16, "id": "disk1", - "interface": "SAS", - "model": "unknown", - "size": 8, - "storage": "SSD", - "vendor": "Unknown" + "interface": "RAID", + "model": "PERC H745 Adp", + "size": 48005923209216, + "storage": "HDD", + "vendor": "Dell" } ], "supported_job_types": { @@ -131,7 +293,8 @@ "queues": [ "admin", "testing" - ] + ], + "virtual": "amd-v" }, "type": "node", "uid": "mercantour7-1" diff --git a/input/grid5000/ipv4.yaml b/input/grid5000/ipv4.yaml index a197eecf231..a4f41c9f7bb 100644 --- a/input/grid5000/ipv4.yaml +++ b/input/grid5000/ipv4.yaml @@ -153,6 +153,7 @@ ipv4: sophia mercantour6 eth0 0 0 2 45 sophia mercantour6 ib0 0 0 2 45 sophia mercantour7 eth0 0 0 2 46 + sophia mercantour7 ib0 0 0 2 46 sophia esterel1 eth0 0 0 3 0 sophia esterel1 ib0 0 0 3 0 sophia esterel2 eth0 0 0 3 1 diff --git a/input/grid5000/sites/sophia/clusters/mercantour7/mercantour7.yaml b/input/grid5000/sites/sophia/clusters/mercantour7/mercantour7.yaml index 095c6ba41b4..c11229b7e2d 100644 --- a/input/grid5000/sites/sophia/clusters/mercantour7/mercantour7.yaml +++ b/input/grid5000/sites/sophia/clusters/mercantour7/mercantour7.yaml @@ -29,15 +29,22 @@ nodes: enabled: true mountable: true mounted: true + eth1: + enabled: false + mountable: false + mounted: false + ib0: + mounted: true + enabled: true + mountable: true + netmask: 255.255.240.0 storage_devices: - disk0: + pci-0000:41:00.0-scsi-0:2:0:0: id: disk0 - interface: SAS - by_path: "/dev/disk/by-path/pci-0000:41:00.0-scsi-0:2:0:0" - disk1: + interface: RAID + pci-0000:41:00.0-scsi-0:2:1:0: id: disk1 - interface: SAS - by_path: "/dev/disk/by-path/pci-0000:41:00.0-scsi-0:2:1:0" + interface: RAID software: standard-environment: debian11-x64-std management_tools: diff --git a/input/grid5000/sites/sophia/clusters/mercantour7/nodes.yaml.erb b/input/grid5000/sites/sophia/clusters/mercantour7/nodes.yaml.erb deleted file mode 100644 index e8dac50541b..00000000000 --- a/input/grid5000/sites/sophia/clusters/mercantour7/nodes.yaml.erb +++ /dev/null @@ -1,84 +0,0 @@ -<% - # File generated by 'rake mass:create SRC=doc/sophia-clusters-mass-create-tmp.csv - # If changes are needed, it might be better to edit the source data and regenerate using 'rake mass:create' - - cluster_name = "mercantour7" - nodes_number = 1 # Size of the cluster (number of nodes) - # MAC addresses declaration - mac_eth0_list = %w( -F4:02:70:B9:60:34 - ) - mac_bmc_list = %w( -70:b5:e8:de:e8:76 - ) -%> ---- -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: 10000000000 # 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/mercantour7/nodes/mercantour7-1.yaml b/input/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.yaml new file mode 100644 index 00000000000..8bb40834222 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/mercantour7/nodes/mercantour7-1.yaml @@ -0,0 +1,171 @@ +# Generated by g5k-checks (g5k-checks -m api) +--- +mercantour7-1: + architecture: + cpu_core_numbering: contiguous + nb_cores: 64 + nb_procs: 2 + nb_threads: 128 + platform_type: x86_64 + bios: + release_date: '09/12/2023' + vendor: Dell Inc. + version: 2.13.3 + bmc_version: 7.00.60.00 + chassis: + manufacturer: Dell Inc. + name: PowerEdge R7525 + serial: JN4LS73 + main_memory: + ram_size: 412316860416 + memory_devices: + dimm_a1: + size: 17179869184 + technology: dram + dimm_a10: + size: 17179869184 + technology: dram + dimm_a11: + size: 17179869184 + technology: dram + dimm_a12: + size: 17179869184 + technology: dram + dimm_a2: + size: 17179869184 + technology: dram + dimm_a3: + size: 17179869184 + technology: dram + dimm_a4: + size: 17179869184 + technology: dram + dimm_a5: + size: 17179869184 + technology: dram + dimm_a6: + size: 17179869184 + technology: dram + dimm_a7: + size: 17179869184 + technology: dram + dimm_a8: + size: 17179869184 + technology: dram + dimm_a9: + size: 17179869184 + technology: dram + dimm_b1: + size: 17179869184 + technology: dram + dimm_b10: + size: 17179869184 + technology: dram + dimm_b11: + size: 17179869184 + technology: dram + dimm_b12: + size: 17179869184 + technology: dram + dimm_b2: + size: 17179869184 + technology: dram + dimm_b3: + size: 17179869184 + technology: dram + dimm_b4: + size: 17179869184 + technology: dram + dimm_b5: + size: 17179869184 + technology: dram + dimm_b6: + size: 17179869184 + technology: dram + dimm_b7: + size: 17179869184 + technology: dram + dimm_b8: + size: 17179869184 + technology: dram + dimm_b9: + size: 17179869184 + technology: dram + network_adapters: + bmc: + ip: 172.17.130.47 + mac: 70:b5:e8:de:e8:76 + management: true + eth0: + driver: tg3 + firmware_version: FFV22.71.3 bc 5720-v1.39 + interface: Ethernet + ip: 172.16.130.47 + mac: f4:02:70:b9:60:34 + management: false + model: NetXtreme BCM5720 2-port Gigabit Ethernet PCIe + name: enp1s0f0np0 + rate: 1000000000 + sriov: false + sriov_totalvfs: 0 + vendor: Broadcom Inc. and subsidiaries + eth1: + driver: tg3 + firmware_version: FFV22.71.3 bc 5720-v1.39 + interface: Ethernet + mac: f4:02:70:b9:60:35 + management: false + model: NetXtreme BCM5720 2-port Gigabit Ethernet PCIe + name: eno2 + sriov: false + sriov_totalvfs: 0 + vendor: Broadcom Inc. and subsidiaries + ib0: + driver: ib_ipoib + firmware_version: 2.42.5000 + guid: '0x1c34da0300e46d11' + interface: InfiniBand + mac: 80:00:02:08:fe:80:00:00:00:00:00:00:1c:34:da:03:00:e4:6d:11 + management: false + model: MT27520 Family [ConnectX-3 Pro] + name: ibp161s0 + rate: 40000000000 + sriov: true + sriov_totalvfs: 8 + vendor: Mellanox Technologies + operating_system: + cstate_driver: acpi_idle + cstate_governor: menu + ht_enabled: true + pstate_driver: acpi-cpufreq + pstate_governor: performance + turboboost_enabled: true + processor: + cache_l1d: 32768 + cache_l1i: 32768 + cache_l2: 524288 + cache_l3: 16777216 + ht_capable: true + instruction_set: x86-64 + microcode: '0x830107c' + model: AMD EPYC + other_description: AMD EPYC 7502 32-Core Processor + vendor: AMD + version: 7502 + storage_devices: + pci-0000:41:00.0-scsi-0:2:0:0: + by_id: "/dev/disk/by-id/wwn-0x670b5e80db81ca0027c136530350d932" + by_path: "/dev/disk/by-path/pci-0000:41:00.0-scsi-0:2:0:0" + firmware_version: 5.16 + model: PERC H745 Adp + size: 959656755200 + storage: SSD + pci-0000:41:00.0-scsi-0:2:1:0: + by_id: "/dev/disk/by-id/wwn-0x670b5e80db81ca0027c13b3347a1dd0f" + by_path: "/dev/disk/by-path/pci-0000:41:00.0-scsi-0:2:1:0" + firmware_version: 5.16 + model: PERC H745 Adp + size: 48005923209216 + storage: HDD + supported_job_types: + virtual: amd-v -- GitLab