Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
grid5000
reference-repository
Commits
7d4149be
Commit
7d4149be
authored
Feb 01, 2021
by
BERARD Benjamin
Browse files
Installation du cluster avec GPU A100 à Nancy
parent
0c2adbfd
Changes
9
Hide whitespace changes
Inline
Side-by-side
data/grid5000/sites/nancy/clusters/grouille/grouille.json
0 → 100644
View file @
7d4149be
{
"created_at"
:
"Wed, 13 Jan 2021 00:00:00 GMT"
,
"exotic"
:
false
,
"kavlan"
:
false
,
"metrics"
:
[
],
"model"
:
"Dell PowerEdge R7525"
,
"queues"
:
[
"admin"
,
"testing"
],
"type"
:
"cluster"
,
"uid"
:
"grouille"
}
\ No newline at end of file
data/grid5000/sites/nancy/clusters/grouille/nodes/grouille-1.json
0 → 100644
View file @
7d4149be
{
"architecture"
:
{
"cpu_core_numbering"
:
"contiguous"
,
"nb_cores"
:
64
,
"nb_procs"
:
2
,
"nb_threads"
:
128
,
"platform_type"
:
"x86_64"
},
"bios"
:
{
"release_date"
:
"07/27/2020"
,
"vendor"
:
"Dell Inc."
,
"version"
:
"1.5.5"
},
"bmc_version"
:
"4.30.30.30"
,
"chassis"
:
{
"manufacturer"
:
"Dell Inc."
,
"name"
:
"PowerEdge R7525"
,
"serial"
:
"64WT393"
},
"exotic"
:
false
,
"gpu_devices"
:
{
"nvidia0"
:
{
"cpu_affinity"
:
0
,
"device"
:
"/dev/nvidia0"
,
"memory"
:
40537000000
,
"model"
:
"A100-PCIE-40GB"
,
"power_default_limit"
:
"250.00 W"
,
"vbios_version"
:
"92.00.25.00.08"
,
"vendor"
:
"Nvidia"
},
"nvidia1"
:
{
"cpu_affinity"
:
1
,
"device"
:
"/dev/nvidia1"
,
"memory"
:
40537000000
,
"model"
:
"A100-PCIE-40GB"
,
"power_default_limit"
:
"250.00 W"
,
"vbios_version"
:
"92.00.25.00.08"
,
"vendor"
:
"Nvidia"
}
},
"main_memory"
:
{
"ram_size"
:
137438953472
},
"monitoring"
:
{
"wattmeter"
:
"false"
},
"network_adapters"
:
[
{
"device"
:
"eth0"
,
"driver"
:
"tg3"
,
"enabled"
:
false
,
"firmware_version"
:
"FFV21.60.16 bc 5720-v1.39"
,
"interface"
:
"Ethernet"
,
"kavlan"
:
false
,
"mac"
:
"70:b5:e8:d0:b2:f8"
,
"management"
:
false
,
"model"
:
"NetXtreme BCM5720 Gigabit Ethernet PCIe"
,
"mountable"
:
false
,
"mounted"
:
false
,
"name"
:
"eno1"
,
"vendor"
:
"Broadcom"
},
{
"device"
:
"eth1"
,
"driver"
:
"tg3"
,
"enabled"
:
false
,
"firmware_version"
:
"FFV21.60.16 bc 5720-v1.39"
,
"interface"
:
"Ethernet"
,
"kavlan"
:
false
,
"mac"
:
"70:b5:e8:d0:b2:f9"
,
"management"
:
false
,
"model"
:
"NetXtreme BCM5720 Gigabit Ethernet PCIe"
,
"mountable"
:
false
,
"mounted"
:
false
,
"name"
:
"eno2"
,
"vendor"
:
"Broadcom"
},
{
"device"
:
"eth2"
,
"driver"
:
"mlx5_core"
,
"enabled"
:
true
,
"firmware_version"
:
"16.27.6106 (DEL0000000016)"
,
"interface"
:
"Ethernet"
,
"ip"
:
"172.16.79.1"
,
"ip6"
:
"2001:660:4406:500:10:0::1"
,
"kavlan"
:
false
,
"mac"
:
"04:3f:72:cd:c2:18"
,
"management"
:
false
,
"model"
:
"MT27800 Family [ConnectX-5]"
,
"mountable"
:
true
,
"mounted"
:
true
,
"name"
:
"eno33"
,
"network_address"
:
"grouille-1.nancy.grid5000.fr"
,
"rate"
:
25000000000
,
"switch"
:
null
,
"switch_port"
:
null
,
"vendor"
:
"Mellanox Technologies"
},
{
"device"
:
"eth3"
,
"driver"
:
"mlx5_core"
,
"enabled"
:
false
,
"firmware_version"
:
"16.27.6106 (DEL0000000016)"
,
"interface"
:
"Ethernet"
,
"kavlan"
:
false
,
"mac"
:
"04:3f:72:cd:c2:19"
,
"management"
:
false
,
"model"
:
"MT27800 Family [ConnectX-5]"
,
"mountable"
:
false
,
"mounted"
:
false
,
"name"
:
"eno34"
,
"vendor"
:
"Mellanox Technologies"
},
{
"device"
:
"bmc"
,
"enabled"
:
true
,
"interface"
:
"Ethernet"
,
"ip"
:
"172.17.79.1"
,
"kavlan"
:
false
,
"mac"
:
"70:b5:e8:e7:3c:74"
,
"management"
:
true
,
"mountable"
:
false
,
"mounted"
:
false
,
"network_address"
:
"grouille-1-bmc.nancy.grid5000.fr"
}
],
"operating_system"
:
{
"cstate_driver"
:
"acpi_idle"
,
"cstate_governor"
:
"menu"
,
"ht_enabled"
:
true
,
"pstate_driver"
:
"acpi-cpufreq"
,
"pstate_governor"
:
"performance"
,
"turboboost_enabled"
:
true
},
"performance"
:
{
"core_flops"
:
16800000000
,
"node_flops"
:
1075200000000
},
"processor"
:
{
"cache_l1"
:
null
,
"cache_l1d"
:
32768
,
"cache_l1i"
:
32768
,
"cache_l2"
:
524288
,
"cache_l3"
:
16777216
,
"clock_speed"
:
2100000000
,
"ht_capable"
:
true
,
"instruction_set"
:
"x86-64"
,
"microarchitecture"
:
"Zen"
,
"microcode"
:
"0x8301038"
,
"model"
:
"AMD EPYC"
,
"other_description"
:
"AMD EPYC 7452 32-Core Processor"
,
"vendor"
:
"AMD"
,
"version"
:
7452
},
"sensors"
:
{
},
"software"
:
{
"forced-deployment-timestamp"
:
202007300948
,
"postinstall-version"
:
"1.2021012800"
,
"standard-environment"
:
"debian10-x64-std"
},
"storage_devices"
:
[
{
"by_id"
:
"/dev/disk/by-id/wwn-0x500a075129de046c"
,
"by_path"
:
"/dev/disk/by-path/pci-0000:01:00.0-scsi-0:0:0:0"
,
"device"
:
"sda"
,
"firmware_version"
:
"D3DJ004"
,
"interface"
:
"SAS"
,
"model"
:
"MTFDDAK960TDT"
,
"size"
:
960197124096
,
"storage"
:
"SSD"
,
"vendor"
:
"Micron"
},
{
"by_id"
:
"/dev/disk/by-id/wwn-0x58ce38ee20f6cc95"
,
"by_path"
:
"/dev/disk/by-path/pci-0000:01:00.0-scsi-0:0:1:0"
,
"device"
:
"sdb"
,
"firmware_version"
:
"B707"
,
"interface"
:
"SATA"
,
"model"
:
"KRM5XVUG1T92"
,
"size"
:
1920383410176
,
"storage"
:
"SSD"
,
"vendor"
:
"Toshiba"
}
],
"supported_job_types"
:
{
"besteffort"
:
true
,
"deploy"
:
true
,
"max_walltime"
:
0
,
"queues"
:
[
"admin"
,
"testing"
],
"virtual"
:
"amd-v"
},
"type"
:
"node"
,
"uid"
:
"grouille-1"
}
\ No newline at end of file
data/grid5000/sites/nancy/clusters/grouille/nodes/grouille-2.json
0 → 100644
View file @
7d4149be
{
"architecture"
:
{
"cpu_core_numbering"
:
"contiguous"
,
"nb_cores"
:
64
,
"nb_procs"
:
2
,
"nb_threads"
:
128
,
"platform_type"
:
"x86_64"
},
"bios"
:
{
"release_date"
:
"10/05/2020"
,
"vendor"
:
"Dell Inc."
,
"version"
:
"1.7.3"
},
"bmc_version"
:
"4.32.10.00"
,
"chassis"
:
{
"manufacturer"
:
"Dell Inc."
,
"name"
:
"PowerEdge R7525"
,
"serial"
:
"54WT393"
},
"exotic"
:
false
,
"gpu_devices"
:
{
"nvidia0"
:
{
"cpu_affinity"
:
0
,
"device"
:
"/dev/nvidia0"
,
"memory"
:
40537000000
,
"model"
:
"A100-PCIE-40GB"
,
"power_default_limit"
:
"250.00 W"
,
"vbios_version"
:
"92.00.25.00.08"
,
"vendor"
:
"Nvidia"
},
"nvidia1"
:
{
"cpu_affinity"
:
1
,
"device"
:
"/dev/nvidia1"
,
"memory"
:
40537000000
,
"model"
:
"A100-PCIE-40GB"
,
"power_default_limit"
:
"250.00 W"
,
"vbios_version"
:
"92.00.25.00.08"
,
"vendor"
:
"Nvidia"
}
},
"main_memory"
:
{
"ram_size"
:
137438953472
},
"monitoring"
:
{
"wattmeter"
:
"false"
},
"network_adapters"
:
[
{
"device"
:
"eth0"
,
"driver"
:
"tg3"
,
"enabled"
:
false
,
"firmware_version"
:
"FFV21.60.16 bc 5720-v1.39"
,
"interface"
:
"Ethernet"
,
"kavlan"
:
false
,
"mac"
:
"70:b5:e8:d0:b1:9c"
,
"management"
:
false
,
"model"
:
"NetXtreme BCM5720 Gigabit Ethernet PCIe"
,
"mountable"
:
false
,
"mounted"
:
false
,
"name"
:
"eno1"
,
"vendor"
:
"Broadcom"
},
{
"device"
:
"eth1"
,
"driver"
:
"tg3"
,
"enabled"
:
false
,
"firmware_version"
:
"FFV21.60.16 bc 5720-v1.39"
,
"interface"
:
"Ethernet"
,
"kavlan"
:
false
,
"mac"
:
"70:b5:e8:d0:b1:9d"
,
"management"
:
false
,
"model"
:
"NetXtreme BCM5720 Gigabit Ethernet PCIe"
,
"mountable"
:
false
,
"mounted"
:
false
,
"name"
:
"eno2"
,
"vendor"
:
"Broadcom"
},
{
"device"
:
"eth2"
,
"driver"
:
"mlx5_core"
,
"enabled"
:
true
,
"firmware_version"
:
"16.27.6106 (DEL0000000016)"
,
"interface"
:
"Ethernet"
,
"ip"
:
"172.16.79.2"
,
"ip6"
:
"2001:660:4406:500:10:0::2"
,
"kavlan"
:
false
,
"mac"
:
"04:3f:72:da:4f:20"
,
"management"
:
false
,
"model"
:
"MT27800 Family [ConnectX-5]"
,
"mountable"
:
true
,
"mounted"
:
true
,
"name"
:
"eno33"
,
"network_address"
:
"grouille-2.nancy.grid5000.fr"
,
"rate"
:
25000000000
,
"switch"
:
null
,
"switch_port"
:
null
,
"vendor"
:
"Mellanox Technologies"
},
{
"device"
:
"eth3"
,
"driver"
:
"mlx5_core"
,
"enabled"
:
false
,
"firmware_version"
:
"16.27.6106 (DEL0000000016)"
,
"interface"
:
"Ethernet"
,
"kavlan"
:
false
,
"mac"
:
"04:3f:72:da:4f:21"
,
"management"
:
false
,
"model"
:
"MT27800 Family [ConnectX-5]"
,
"mountable"
:
false
,
"mounted"
:
false
,
"name"
:
"eno34"
,
"vendor"
:
"Mellanox Technologies"
},
{
"device"
:
"bmc"
,
"enabled"
:
true
,
"interface"
:
"Ethernet"
,
"ip"
:
"172.17.79.2"
,
"kavlan"
:
false
,
"mac"
:
"70:b5:e8:e7:46:70"
,
"management"
:
true
,
"mountable"
:
false
,
"mounted"
:
false
,
"network_address"
:
"grouille-2-bmc.nancy.grid5000.fr"
}
],
"operating_system"
:
{
"cstate_driver"
:
"acpi_idle"
,
"cstate_governor"
:
"menu"
,
"ht_enabled"
:
true
,
"pstate_driver"
:
"acpi-cpufreq"
,
"pstate_governor"
:
"performance"
,
"turboboost_enabled"
:
true
},
"performance"
:
{
"core_flops"
:
16800000000
,
"node_flops"
:
1075200000000
},
"processor"
:
{
"cache_l1"
:
null
,
"cache_l1d"
:
32768
,
"cache_l1i"
:
32768
,
"cache_l2"
:
524288
,
"cache_l3"
:
16777216
,
"clock_speed"
:
2100000000
,
"ht_capable"
:
true
,
"instruction_set"
:
"x86-64"
,
"microarchitecture"
:
"Zen"
,
"microcode"
:
"0x830104d"
,
"model"
:
"AMD EPYC"
,
"other_description"
:
"AMD EPYC 7452 32-Core Processor"
,
"vendor"
:
"AMD"
,
"version"
:
7452
},
"sensors"
:
{
},
"software"
:
{
"forced-deployment-timestamp"
:
202007300948
,
"postinstall-version"
:
"1.2021012800"
,
"standard-environment"
:
"debian10-x64-std"
},
"storage_devices"
:
[
{
"by_id"
:
"/dev/disk/by-id/wwn-0x500a075129de049a"
,
"by_path"
:
"/dev/disk/by-path/pci-0000:01:00.0-scsi-0:0:0:0"
,
"device"
:
"sda"
,
"firmware_version"
:
"D3DJ004"
,
"interface"
:
"SAS"
,
"model"
:
"MTFDDAK960TDT"
,
"size"
:
960197124096
,
"storage"
:
"SSD"
,
"vendor"
:
"Micron"
},
{
"by_id"
:
"/dev/disk/by-id/wwn-0x58ce38ee20f6cc9d"
,
"by_path"
:
"/dev/disk/by-path/pci-0000:01:00.0-scsi-0:0:1:0"
,
"device"
:
"sdb"
,
"firmware_version"
:
"B707"
,
"interface"
:
"SATA"
,
"model"
:
"KRM5XVUG1T92"
,
"size"
:
1920383410176
,
"storage"
:
"SSD"
,
"vendor"
:
"Toshiba"
}
],
"supported_job_types"
:
{
"besteffort"
:
true
,
"deploy"
:
true
,
"max_walltime"
:
0
,
"queues"
:
[
"admin"
,
"testing"
],
"virtual"
:
"amd-v"
},
"type"
:
"node"
,
"uid"
:
"grouille-2"
}
\ No newline at end of file
input/grid5000/disks.yaml
View file @
7d4149be
...
...
@@ -50,6 +50,7 @@ disk_vendor_model_mapping:
-
TOSHIBA MG04ACA4
-
MG04SCA40ENY
-
AL15SEB060NY
-
KRM5XVUG1T92
Intel
:
-
INTEL SSDSC2BB30
-
SSDSC2KG480G7R
...
...
@@ -66,5 +67,6 @@ disk_vendor_model_mapping:
Micron
:
-
MTFDDAK480TDN
-
MTFDDAK960TDN
-
MTFDDAK960TDT
Unknown
:
-
unknown
input/grid5000/ipv4.yaml
View file @
7d4149be
...
...
@@ -53,6 +53,7 @@ ipv4:
nancy grvingt eth0 0 0 12 0
nancy grue eth0 0 0 13 0
nancy grappe eth0 0 0 14 0
nancy grouille eth2 0 0 15 0
nantes econome eth0 0 0 0 0
nantes ecotype eth0 0 0 1 0
nantes ecotype eth1 0 0 2 0
...
...
input/grid5000/sites/nancy/clusters/grouille/grouille.yaml
0 → 100644
View file @
7d4149be
model
:
Dell PowerEdge R7525
created_at
:
2021-01-13
kavlan
:
false
exotic
:
false
queues
:
-
admin
-
testing
nodes
:
grouille-[1-2]
:
supported_job_types
:
deploy
:
true
besteffort
:
true
max_walltime
:
0
processor
:
microarchitecture
:
Zen
clock_speed
:
2100000000
software
:
standard-environment
:
debian10-x64-std
network_adapters
:
bmc
:
interface
:
Ethernet
enabled
:
true
mountable
:
false
mounted
:
false
eth0
:
enabled
:
false
mountable
:
false
mounted
:
false
eth1
:
enabled
:
false
mountable
:
false
mounted
:
false
eth2
:
enabled
:
true
mountable
:
true
mounted
:
true
eth3
:
enabled
:
false
mountable
:
false
mounted
:
false
storage_devices
:
sda
:
interface
:
SAS
storage
:
SSD
sdb
:
interface
:
SATA
storage
:
SSD
input/grid5000/sites/nancy/clusters/grouille/nodes/grouille-1.yaml
0 → 100644
View file @
7d4149be
# Generated by g5k-checks (g5k-checks -m api)
---
grouille-1
:
architecture
:
cpu_core_numbering
:
contiguous
nb_cores
:
64
nb_procs
:
2
nb_threads
:
128
platform_type
:
x86_64
bios
:
release_date
:
07/27/2020
vendor
:
Dell Inc.
version
:
1.5.5
bmc_version
:
4.30.30.30
chassis
:
manufacturer
:
Dell Inc.
name
:
PowerEdge R7525
serial
:
64WT393
gpu_devices
:
nvidia0
:
cpu_affinity
:
0
device
:
"
/dev/nvidia0"
memory
:
40537000000
model
:
A100-PCIE-40GB
power_default_limit
:
250.00 W
vbios_version
:
92.00.25.00.08
vendor
:
Nvidia
nvidia1
:
cpu_affinity
:
1
device
:
"
/dev/nvidia1"
memory
:
40537000000
model
:
A100-PCIE-40GB
power_default_limit
:
250.00 W
vbios_version
:
92.00.25.00.08
vendor
:
Nvidia
main_memory
:
ram_size
:
137438953472
network_adapters
:
bmc
:
ip
:
172.17.79.1
mac
:
70:b5:e8:e7:3c:74
management
:
true
eth0
:
driver
:
tg3
firmware_version
:
FFV21.60.16 bc 5720-v1.39
interface
:
Ethernet
mac
:
70:b5:e8:d0:b2:f8
management
:
false
model
:
NetXtreme BCM5720 Gigabit Ethernet PCIe
name
:
eno1
vendor
:
Broadcom
eth1
:
driver
:
tg3
firmware_version
:
FFV21.60.16 bc 5720-v1.39
interface
:
Ethernet
mac
:
70:b5:e8:d0:b2:f9
management
:
false
model
:
NetXtreme BCM5720 Gigabit Ethernet PCIe
name
:
eno2
vendor
:
Broadcom
eth2
:
driver
:
mlx5_core
firmware_version
:
16.27.6106 (DEL0000000016)
interface
:
Ethernet
ip
:
172.16.79.1
mac
:
04:3f:72:cd:c2:18
management
:
false
model
:
MT27800 Family [ConnectX-5]
name
:
eno33
rate
:
25000000000
vendor
:
Mellanox Technologies
eth3
:
driver
:
mlx5_core
firmware_version
:
16.27.6106 (DEL0000000016)
interface
:
Ethernet
mac
:
04:3f:72:cd:c2:19
management
:
false
model
:
MT27800 Family [ConnectX-5]
name
:
eno34
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
:
'
0x8301038'
model
:
AMD EPYC
other_description
:
AMD EPYC 7452 32-Core Processor
vendor
:
AMD
version
:
7452