Follow the following procedures to setup an account for FAU clusters
We will ask you at some point to send us your 4K public ssh key, we will have one account per team.
$ ssh-keygen -t rsa -b 4096 ... $ cat .ssh/id_rsa_4k.pub
2. Once we will supply the access, you will need to login to the server with your given username.
Setup your ssh config file to make it easier:
$ cat .ssh/config
Host fau
HostName cshpc.rrze.fau.de
IdentityFile ~/.ssh/id_rsa_4k
User <user id>
3. Login to fau clusters
$ ssh fau .....
4. Setup public-key in your environment on FAU login node.
$ cd .ssh $ ssh-keygen ... $ cp id_rsa.pub authorized_keys
5. Login to Fritz or alex clusters
$ssh fritz =============================================================================== Welcome at "Fritz", FAU's NHR and Tier3 parallel computer .... =============================================================================== b154dc10@fritz4:~$ sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST singlenode up 1-00:00:00 4 drain* f[0402,0737,1083,1151] singlenode up 1-00:00:00 1 drng f1145 singlenode up 1-00:00:00 594 alloc f[0401,0403-0488,0505,0507-0508,0513,0543-0545,0547-0549,0551-0556,0558,0561-0562,0565-0588,0601-0619,0621-0631,0639-0640,0648,0650-0652,0658,0665-0688,0701-0736,0738-0788,0801-0804,0813-0850,0865-0867,0869-0888,0901-0933,0935-0943,0945-0988,1001-1075,1077-1082,1084-1088,1103,1106,1109-1111,1114,1120,1125-1126,1129-1132,1144,1146-1150,1152-1180,1182-1188,1201-1224] singlenode up 1-00:00:00 129 idle f[0501-0504,0506,0509-0512,0514-0542,0546,0550,0557,0559-0560,0563-0564,0620,0632-0638,0641-0647,0649,0653-0657,0659-0664,0805-0812,0851-0864,0868,0934,0944,1076,1101-1102,1104-1105,1107-1108,1112-1113,1115-1119,1121-1124,1127-1128,1133-1143,1181] multinode* up 1-00:00:00 5 drain* f[0243,0402,0737,1083,1151] multinode* up 1-00:00:00 1 drng f1145 multinode* up 1-00:00:00 853 alloc f[0101-0188,0201-0215,0217-0237,0239-0242,0244-0273,0276-0288,0301-0388,0401,0403-0488,0505,0507-0508,0513,0543-0545,0547-0549,0551-0556,0558,0561-0562,0565-0588,0601-0619,0621-0631,0639-0640,0648,0650-0652,0658,0665-0688,0701-0736,0738-0788,0801-0804,0813-0850,0865-0867,0869-0888,0901-0933,0935-0943,0945-0988,1001-1075,1077-1082,1084-1088,1103,1106,1109-1111,1114,1120,1125-1126,1129-1132,1144,1146-1150,1152-1180,1182-1188,1201-1224] multinode* up 1-00:00:00 133 idle f[0216,0238,0274-0275,0501-0504,0506,0509-0512,0514-0542,0546,0550,0557,0559-0560,0563-0564,0620,0632-0638,0641-0647,0649,0653-0657,0659-0664,0805-0812,0851-0864,0868,0934,0944,1076,1101-1102,1104-1105,1107-1108,1112-1113,1115-1119,1121-1124,1127-1128,1133-1143,1181] b154dc10@fritz4:~$
6. User the multinode partition to allocate 4 fritz nodes for the application runs.
b154dc10@fritz4:~$ salloc -N 4 -p multinode
salloc: Granted job allocation 282366
salloc: Waiting for resource configuration
salloc: Nodes f[0216,0238,0274-0275] are ready for job
### Starting TaskPrologue of job 282366 on f0216 at Fri Dec 9 22:36:13 CET 2022
# SLURM_JOB_NODELIST=f[0216,0238,0274-0275]
# SLURM_JOB_NUM_NODES=4
# SLURM_NTASKS=
# SLURM_NPROCS=
# SLURM_TASKS_PER_NODE=72(x4)
# SLURM_JOB_CPUS_PER_NODE=72(x4)
# SLURM_EXPORT_ENV=
Running on cores 0-71 with governor powersave
### Finished TaskPrologue
b154dc10@f0216:~$ ibstat
CA 'mlx5_0'
CA type: MT4123
Number of ports: 1
Firmware version: 20.31.1014
Hardware version: 0
Node GUID: 0x1070fd030052c592
System image GUID: 0x1070fd030052c592
Port 1:
State: Active
Physical state: LinkUp
Rate: 100
Base lid: 386
LMC: 0
SM lid: 1
Capability mask: 0x2659e848
Port GUID: 0x1070fd030052c592
Link layer: InfiniBand
b154dc10@f0216:~$ lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 72
On-line CPU(s) list: 0-71
Thread(s) per core: 1
Core(s) per socket: 36
Socket(s): 2
NUMA node(s): 4
Vendor ID: GenuineIntel
CPU family: 6
Model: 106
Model name: Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz
Stepping: 6
CPU MHz: 2400.000
CPU max MHz: 3500.0000
CPU min MHz: 800.0000
BogoMIPS: 4800.00
Virtualization: VT-x
L1d cache: 48K
L1i cache: 32K
L2 cache: 1280K
L3 cache: 55296K
NUMA node0 CPU(s): 0-17
NUMA node1 CPU(s): 18-35
NUMA node2 CPU(s): 36-53
NUMA node3 CPU(s): 54-71
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 invpcid_single ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid fsrm md_clear pconfig flush_l1d arch_capabilities
b154dc10@f0216:~$
User the a100 partition to allocate 4 gpus on single node for the application runs, using the flag
--gres:a100:4
see below:
$ ssh alex =============================================================================== Welcome at "Alex", FAU's NHR and Tier3 GPGPU cluster .... =============================================================================== $ sinfo PARTITION AVAIL TIMELIMIT NODES STATE NODELIST a40* up 1-00:00:00 9 mix a[0221,0325,0328,0425-0429,0521] a40* up 1-00:00:00 24 alloc a[0121-0129,0222-0229,0321-0324,0326-0327,0522] a40* up 1-00:00:00 5 idle a[0329,0421-0424] a100 up 1-00:00:00 3 mix a[0701,0831,0903] a100 up 1-00:00:00 3 alloc a[0631,0901-0902] a100 up 1-00:00:00 28 idle a[0531-0537,0601-0605,0632,0702-0705,0801-0805,0904-0905,0931-0934] vis up 1-00:00:00 1 down atest01 $ salloc -p a100 --gres=gpu:a100:4 salloc: Granted job allocation 424852 salloc: Waiting for resource configuration salloc: Nodes a0904 are ready for job ### Starting TaskPrologue of job 424852 on a0904 at Fri Dec 9 22:43:35 CET 2022 Running on cores 0-63 with governor ondemand Fri Dec 9 22:43:36 2022 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 525.60.13 Driver Version: 525.60.13 CUDA Version: 12.0 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 NVIDIA A100-SXM... On | 00000000:0E:00.0 Off | 0 | | N/A 32C P0 53W / 400W | 0MiB / 40960MiB | 0% Default | | | | Disabled | +-------------------------------+----------------------+----------------------+ | 1 NVIDIA A100-SXM... On | 00000000:13:00.0 Off | 0 | | N/A 32C P0 54W / 400W | 0MiB / 40960MiB | 0% Default | | | | Disabled | +-------------------------------+----------------------+----------------------+ | 2 NVIDIA A100-SXM... On | 00000000:49:00.0 Off | 0 | | N/A 31C P0 56W / 400W | 0MiB / 40960MiB | 0% Default | | | | Disabled | +-------------------------------+----------------------+----------------------+ | 3 NVIDIA A100-SXM... On | 00000000:4F:00.0 Off | 0 | | N/A 34C P0 54W / 400W | 0MiB / 40960MiB | 0% Default | | | | Disabled | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | No running processes found | +-----------------------------------------------------------------------------+ ### Finished TaskPrologue b154dc10@a0904:~$
More info :
Getting started https://hpc.fau.de/systems-services/documentation-instructions/getting-started/
Fritz general info, how to access, its environment and usage https://hpc.fau.de/systems-services/documentation-instructions/clusters/fritz-cluster/
Alex general info, how to access, its environment and usage https://hpc.fau.de/systems-services/documentation-instructions/clusters/alex-cluster/
Add Comment