Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

WIP: adding memory per core info to hardware spec pages #434

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

WIP: adding memory per core info to hardware spec pages #434

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
f8a6486
adding mem_per_core for Genius partitions
moravveji Oct 9, 2024
86f7d15
adding superdome and wice
moravveji Oct 9, 2024
84805e1
add more *_long partition names
moravveji Oct 10, 2024
296f35d
add superdome partition name
moravveji Oct 10, 2024
483c92e
isolate gpu_a100_debug partition
moravveji Oct 10, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 10 additions & 4 deletions source/leuven/tier2_hardware/genius_hardware.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,8 +20,9 @@ Hardware details
18 cores each :raw-html:`<br />`
(1 NUMA domain and 1 L3 cache per CPU)
- 192 GiB RAM (:ref:`memory bandwidth and latency measurements <memory bandwidth and latency cascadelake tier2>`)
- default memory per core is 5000 MiB
- 200 GB SSD local disk
- partition ``batch/batch_long``,
- partition ``batch|batch_long``,
specific Slurm :ref:`options <submit_genius_batch>` apply

- 12 interactive nodes
Expand All @@ -30,6 +31,7 @@ Hardware details
18 cores each :raw-html:`<br />`
(1 NUMA domain and 1 L3 cache per CPU)
- 192 GiB RAM (:ref:`memory bandwidth and latency measurements <memory bandwidth and latency skylake tier2>`)
- default memory per core is 5000 MiB
- 200 GB SSD local disk
- partition ``interactive``,
specific Slurm :ref:`options <submit_genius_interactive>` apply
Expand All @@ -40,8 +42,9 @@ Hardware details
18 cores each :raw-html:`<br />`
(1 NUMA domain and 1 L3 cache per CPU)
- 768 GiB RAM
- default memory per core is 5000 MiB
- 200 GB SSD local disk
- partition ``bigmem``, specific Slurm :ref:`options <submit_genius_bigmem>` apply
- partition ``bigmem|bigmem_long``, specific Slurm :ref:`options <submit_genius_bigmem>` apply

- 22 GPGPU nodes, 96 GPU devices

Expand All @@ -51,26 +54,29 @@ Hardware details
18 cores each :raw-html:`<br />`
(1 NUMA domain and 1 L3 cache per CPU)
- 192 GiB RAM
- default memory per core is 5000 MiB
- 4 NVIDIA P100 SXM2\@1.3 GHz, 16 GiB GDDR, connected with NVLink
- 200 GB SSD local disk
- partition ``gpu_p100``, specific Slurm :ref:`options <submit_genius_gpu>` apply
- partition ``gpu_p100|gpu_p100_long|gpu_p100_debug``, specific Slurm :ref:`options <submit_genius_gpu>` apply

- 2 V100 nodes

- 2 Xeon Gold 6240 CPUs\@2.6 GHz (Cascadelake),
18 cores each :raw-html:`<br />`
(1 NUMA domain and 1 L3 cache per CPU)
- 768 GiB RAM
- default memory per core is 21000 MiB
- 8 NVIDIA V100 SXM2\@1.5 GHz, 32 GiB GDDR, connected with NVLink
- 200 GB SSD local disk
- partition ``gpu_v100``, specific Slurm :ref:`options <submit_genius_gpu>` apply
- partition ``gpu_v100|gpu_v100_long``, specific Slurm :ref:`options <submit_genius_gpu>` apply

- 4 AMD nodes

- 2 EPYC 7501 CPUs\@2.0 GHz,
32 cores each :raw-html:`<br />`
(4 NUMA domains and 8 L3 caches per CPU)
- 256 GiB RAM
- default memory per core is 3800 MiB
- 200 GB SSD local disk
- partition ``amd``, specific Slurm :ref:`options <submit_genius_amd>` apply

Expand Down
2 changes: 2 additions & 0 deletions source/leuven/tier2_hardware/superdome_hardware.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,8 @@ Superdome consists of

- Xeon Gold 6132 CPU\@2.6 GHz (Skylake), 14 cores each
- 768 GiB RAM each
- default memory per core is 53500 MiB
- partition ``superdome|superdome_long``

- a 6 TB local node disk

Expand Down
27 changes: 22 additions & 5 deletions source/leuven/tier2_hardware/wice_hardware.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,8 +20,9 @@ Hardware details
36 cores each :raw-html:`<br />`
(1 NUMA domain and 1 L3 cache per CPU)
- 256 GiB RAM
- default memory per core is 3400 MiB
- 960 GB SSD local disk
- partitions ``batch/batch_long``,
- partitions ``batch|batch_long|batch_icelake|batch_icelake_long``,
:ref:`submit options <submit to wice compute node>`

- 68 Sapphire Rapids nodes
Expand All @@ -31,8 +32,9 @@ Hardware details
(4 NUMA domains and 1 L3 cache per CPU) :raw-html:`<br />`
The base and max CPU frequencies are 2.1 GHz and 3.8 GHz, respectively.
- 256 GiB RAM
- default memory per core is 2500 MiB
- 960 GB SSD local disk
- partitions ``batch_sapphirerapids/batch_sapphirerapids_long``
- partitions ``batch_sapphirerapids|batch_sapphirerapids_long``
:ref:`submit options <submit to wice compute node>`

- 5 big memory nodes
Expand All @@ -41,6 +43,7 @@ Hardware details
36 cores each :raw-html:`<br />`
(2 NUMA domains and 1 L3 cache per CPU)
- 2048 GiB RAM
- default memory per core is 28000 MiB
- 960 GB SSD local disk
- partition ``bigmem``,
:ref:`submit options <submit to wice big memory node>`
Expand All @@ -52,6 +55,7 @@ Hardware details
(1 NUMA domain and 1 L3 cache per CPU) :raw-html:`<br />`
The base and max CPU frequencies are 2.4 GHz and 3.5 GHz, respectively.
- 8 TiB RAM
- default memory per core is 111900 MiB
- 960 GB SSD local disk
- partition ``hugemem``,
:ref:`submit options <submit to wice big memory node>`
Expand All @@ -64,9 +68,10 @@ Hardware details
36 cores each :raw-html:`<br />`
(2 NUMA domains and 1 L3 cache per CPU)
- 512 GiB RAM
- default memory per core is 7000 MiB
- 4 NVIDIA A100 SXM4, 80 GiB GDDR, connected with NVLink
- 960 GB SSD local disk
- partition ``gpu``,
- partition ``gpu|gpu_a100``,
:ref:`submit options <submit to wice GPU node>`

- 4 nodes with 16 H100 GPUs in total
Expand All @@ -76,23 +81,35 @@ Hardware details
(4 NUMA domains and 4 L3 caches per CPU) :raw-html:`<br />`
The base and max CPU frequencies are 2.7 GHz and 3.9 GHz, respectively.
- 768 GiB RAM
- default memory per core is 11700 MiB
- 4 NVIDIA H100 SXM5, 80 GiB HBM3, connected with NVLink
- 960 GB SSD local disk
- partition ``gpu_h100``,
:ref:`submit options <submit to wice GPU node>`

- 4 interactive nodes and 1 debug node
- 4 interactive nodes

- 2 Intel Xeon Gold 8358 CPUs\@2.6 GHz (Ice lake),
32 cores each :raw-html:`<br />`
(2 NUMA domains and 1 L3 cache per CPU)
- 512 GiB RAM
- default memory per core is 2000 MiB
- 1 NVIDIA A100, 80 GiB GDDR
- 960 GB SSD local disk
- partitions ``interactive``,
:ref:`submit options <submit to wice interactive node>`
:raw-html:`<br />`
and ``gpu_a100_debug``,

- 1 debug node

- 2 Intel Xeon Gold 8358 CPUs\@2.6 GHz (Ice lake),
32 cores each :raw-html:`<br />`
(2 NUMA domains and 1 L3 cache per CPU)
- 512 GiB RAM
- default memory per core is 7500 MiB
- 1 NVIDIA A100, 80 GiB GDDR
- 960 GB SSD local disk
- partitions ``gpu_a100_debug``,
:ref:`submit options <submit to wice GPU node>`

All nodes of the same type are interconnected using an Infiniband HDR-100
Expand Down
Loading