wiki.snu.edu.in - User contributions [en]

Magus02

2026-10-12T06:47:17Z

Bhuvnesh:

====What is HPC?====
<div style="font-size: 0.875em;">High performance computing (HPC) is the ability to process data and perform complex calculations at high speeds. To put it into perspective, a laptop or desktop with a 3 GHz processor can perform around 3 billion calculations per second. While that is much faster than any human can achieve, it pales in comparison to HPC solutions that can perform quadrillions of calculations per second.</div>
HPC is specifically needed for these reasons:

* It paves the way for new innovations in science, technology, business and academia.
* It improves processing speeds, which can be critical for many kinds of computing operations, applications and workloads.
* It helps lay the foundation for a reliable, fast IT infrastructure that can store, process and analyze massive amounts of data for various applications.

'''Specification:'''

* Total Cores = 8064 (AMD)
* 238 Tflops
* 57 Compute Nodes
* 2 Login Nodes
* 2 Master Nodes
* 12 High memory nodes with local SSD
* Better performance of applications, which require high performance storage but do not scale well across nodes. E.g. Gaussian.

'''Memory'''

* 4 GB Per core for Compute Nodes, with a total memory = 29,184 GB
* 16 GB Per core for High Memory Nodes, with total memory = 12,288 GB

===='''Compute Specification:'''====
{| class=" wikitable" style="height: 49px;" width="1084"
|-
| width="115"|'''COMPUTE NODES'''
| width="162"|''' PROCESSOR'''
| width="86"|''' RAM'''
| width="79"|''' CORE'''
| width="86"|''' GPU '''
| width="91"|''' NO OF NODES'''
| width="81"|'''TOTAL RAM'''
| width="89"|'''TOTAL CORES'''
|}
{| class=" wikitable" style="height: 118px;" width="912"
|-
||Compute Node
||AMD EPYC 7763
||512 GB
||128
||-
||57
||29184GB
||7296
|-
||High Memory Node
||AMD EPYC 7543
||1024GB
||64
||-
||12
||12228GB
||768
|-
||GPU
||INTEL XEON PLATINUM 8358
||512GB
||64
||Tesla A100*4
||2
||1024GB
||128
|-
||Total
||
||
||
||
||
||42436
||8192
|-
||Storage
||750TB PFS
||
||
||
||
||
||
|}

===='''How to connect to Snu HPC ?'''====
 Enumerated here are the various types.

* '''SSH : '''Host magus02.snu.edu.in ( default port 22 )
* '''From Linux Machine : '''$ssh user@magus02.snu.edu.in
* '''From Windows Machine using ''''''putty.'''

===='''How to Reset Password on Magus02 ?'''====
 Step1. Login to magus02 account.

[[File:magus 02 pic 1.1716181911641.png|frameless|578x578px]]

Step2: Use Passwd Command.

[[File:magus 03 pic.1716182070345.png|frameless|558x558px]] 

==='''How to Connect to Magus from outside SNU Network ?'''===
To gain access to MEGUS from outside the SNU network, a VPN connection is mandatory. To acquire VPN access, kindly click on the [https://forms.office.com/Pages/ResponsePage.aspx?id=12YS_ftig0aFL0FyYUAqLN3tPd66H1xCnswYz1fby35UMkhBR0lMSlJLUlZGT0wwTzJTOUtLNEpZVy4u VPN Request Form] or utilize the option of accessing it by scanning the barcode provided below.

===='''How to Transfers data using WinSCP Windows (Client).'''====
Step1 : Open the WinSCP and fill all the details, After that click on the Login Button.

[[File:magus 04 pic.1716182129572.png|frameless|589x589px]]

Step 2 : Click Yes on Warning tab.

Step 3: Connected, Now you can transfer the data using Right click of key Panel.

[[File:magus o5 pic.1716182172254.png|left|frameless|530x530px]]

'''Applications on Megus'''

[[Vasp5.4|Vasp]], [[02HPC Quantum Espresso Magus02|Quantum espresso]], [[Gaussian16|Gaussian]], Siesta, [[Gromacs]], OpenFOAM.

==='''Queue Structure on Magus02.'''===
{| class="wikitable" style="background-color: #fafafa; height: 418px;" border="1" width="1523"
|-
| style="background-color:#bababa;" class="" width="60" |S.No.
| style="background-color:#bababa;" class="" width="115" |Processor Architecture
| style="background-color:#bababa;" class="" width="81" |Priority
| style="background-color:#bababa;" class="" width="147" |Queue Name
| style="background-color:#bababa;" class="" width="110" |Min no of Cores Required to submit job
| style="background-color:#bababa;" class="" width="99" |Max no of cores allowed per job
| style="background-color:#bababa;" class="" width="68" |No of nodes in the queue
| style="background-color:#bababa;" class="" width="92" |Wall Time
| style="background-color:#bababa;" class="" width="111" |MAX Cores User can Use in the Queue
| style="background-color:#bababa;" class="" width="111" |Max No. of Pending Jobs Per user
| style="background-color:#bababa;" class="" width="111" |Max No. of Running Jobs Per user
|-
| width="60" |'''1'''

| width="115" |
AMD EPYC 7763

| width="81" |
50

| width="147" |
Medium

| width="110" |
32

| width="99" |
128

| width="68" |
17

| width="92" |
1 week

| width="111" |
512

| width="111" | 2
| width="111" | 2
|-
| width="60" |'''2'''

| width="115" |
AMD EPYC 7543

| width="81" |
50

| width="147" |
Small

| width="110" |
8

| width="99" |
32

| width="68" |
16

| width="92" |
1 Month

| width="111" |
128

| width="111" |2
| width="111" |4
|-
| width="60" |'''3'''

| width="115" |
AMD EPYC 7543

| width="81" |
50

| width="147" |
High_mem

| width="110" |
32

| width="99" |
256

| width="68" |
12

| width="92" |
2 weeks

| width="111" |
1024

| width="111" |2
| width="111" |2
|-
| width="60" |'''4'''

| width="115" |
AMD EPYC 7763

| width="81" |
50

| width="147" |
Large

| width="110" |
64

| width="99" |
256

| width="68" |
24

| width="92" |
2 weeks

| width="111" |
1024

| width="111" |2
| width="111" |2
|-
| width="60" |'''5'''

| width="115" |
Intel Xeon Platinum 8358

| width="81" |
50

| width="147" |
GPU

| width="110" |
4

| width="99" |
16

| width="68" |
2

| width="92" |
1 week

| width="111" |
32

| width="111" |
| width="111" |2
|-
| width="60" |'''6'''

| width="115" |
AMD EPYC 7763

| width="81" |
Low

| width="147" |
Long

| width="110" |
32

| width="99" |
1024

| width="68" |
All lm nodes

| width="92" |
1 Month

| width="111" |
1024

| width="111" |
| width="111" |8
|}
====='''[[Slurm Command Reference:]]'''=====
'''[[index.php?title=Media:Linux Command Reference.pdf|Linux_Command_Reference.pdf]]'''

'''[Https://wiki.snu.edu.in/index.php?title=GPU MIG SLICE&preload= How to Use GPU MIG Slice]'''

GPU MIG SLICE

2026-10-12T05:39:05Z

Bhuvnesh:

=== Why MIG is Essential for Magus ===
Our core HPC environment, '''Magus02''', houses a total of '''8 A100 GPU cards'''. Because our physical GPU footprint is highly concentrated, treating each A100 as an "all-or-nothing" resource creates severe scheduling bottlenecks and immense resource wastage.

Prior to this upgrade, a user running a small script that only required 5GB of memory would lock an entire 80GB card, effectively wasting 75GB of enterprise VRAM and blocking other researchers from working. With only 8 cards total on Magus02, this under-utilization is no longer sustainable. MIG directly solves this problem.

=== Why Use MIG? (And How It Benefits You) ===
MIG allows a single A100 physical GPU to be securely partitioned into 7 separate, isolated instances. This directly addresses the cluster bottlenecks discussed in our policy meeting by delivering:

* '''Drastically Reduced Resource Wastage:''' Instead of only 8 users locking down the entire cluster, Magus02 can now dynamically host up to 56 individual GPU jobs simultaneously, ensuring no VRAM sits idle.
* '''Faster Queue Times:''' You can now run small-to-medium jobs immediately without waiting days in the queue for a full 80GB GPU to become free.
* '''Hardware-Level Isolation:''' Each MIG slice has its own dedicated memory, cache, and compute cores. Your job's performance will remain entirely unaffected by other users sharing the same physical card.
* '''Support for Multi-Process Frameworks:''' Because each MIG instance operates as a fully distinct GPU device, your multi-process frameworks can seamlessly bind to individual slices. This allows for clean, parallel data-processing tasks without complex internal scheduling.

=== Updated GPU & CPU Configurations ===
To align with this shift toward efficiency, we have also '''lowered the minimum CPU requirement for GPU jobs from 16 cores to 4 cores''', allowing you to run much more compact and efficient workloads without wasting CPU cycles.
{| class="wikitable" style="color: rgb(34, 34, 34); font-size: small; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(255, 255, 255); text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; margin-bottom: 32px; font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;"
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Resource Type'''
| style="margin:0px;border:1px solid;line-height:1.15;" |'''VRAM'''
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Compute (SM)'''
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Slurm Directive'''
|- style="line-height: 1.15; margin-top: 0px;"
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Full GPU'''
| style="margin:0px;border:1px solid;line-height:1.15;" |80GB
| style="margin:0px;border:1px solid;line-height:1.15;" |98 SMs
| style="margin:0px;border:1px solid;line-height:1.15;" |<code style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">#SBATCH --gres=gpu:nvidia_a100_80gb_pcie_7g.80gb:1</code>
|- style="line-height: 1.15; margin-top: 0px;"
| style="margin:0px;border:1px solid;line-height:1.15;" |'''MIG Slice (Single)'''
| style="margin:0px;border:1px solid;line-height:1.15;" |10GB
| style="margin:0px;border:1px solid;line-height:1.15;" |14 SMs
| style="margin:0px;border:1px solid;line-height:1.15;" |<code style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">#SBATCH --gres=gpu:nvidia_a100_80gb_pcie_1g.10gb:1</code>
|}

=== Combining & Using Multiple Slices in One Job ===
If your workload requires more than 10GB of VRAM or needs to scale across multiple independent GPU targets within a single job script, you can request multiple slices simultaneously.

==== 1. Requesting the Slices ====
To request multiple slices (for example, 2 slices to get 20GB of total VRAM), adjust your Slurm directive count at the end of the string:

'''''#SBATCH --gres=gpu:nvidia_a100_80gb_pcie_1g.10gb:2'''''

==== 2. Utilizing the Slices in Your Job ====
When you request multiple slices, Slurm will automatically provision them and expose them to your job environment.

* '''For Multi-Process/Multi-GPU Frameworks:''' Frameworks like PyTorch (DDP), TensorFlow, or custom multi-process Python scripts will automatically detect these as '''two distinct GPU devices''' (Device 0 and Device 1).
* '''For Parallel Task Execution:''' You can target specific slices within the same job submission using standard environment variables or Slurm arrays.

=== Troubleshooting & Device Visibility ===
Slurm automatically manages and sets up the environment for you. To verify your allocation or troubleshoot device visibility within your job script, you can print the assigned environment variables:
<span style="color: rgb(34, 34, 34); font-size: small; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(255, 255, 255); text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;"><div style="line-height: 1.15; margin-top: 0px;"><div style="line-height: 1.15; margin-top: 0px;"><div style="line-height: 1.15; margin-top: 0px;">'''''<div style="line-height: 1.15; margin-top: 0px;">echo "Allocated CUDA Devices: $CUDA_VISIBLE_DEVICES"</div>
<code role="text" style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">nvidia-smi
</code>
'''''</div></div></div><blockquote style="color: rgb(34, 34, 34); font-size: small; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(255, 255, 255); text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">''Important Tip:'<nowiki/>'' If your code or framework is not detecting the assigned MIG slices, double-check your script to ensure you are not accidentally overwriting or resetting the <code style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">$CUDA_VISIBLE_DEVICES</code> variable during your environment or module setup.</blockquote>

Magus02

2026-10-12T05:37:38Z

Bhuvnesh:

====What is HPC?====
<div style="font-size: 0.875em;">High performance computing (HPC) is the ability to process data and perform complex calculations at high speeds. To put it into perspective, a laptop or desktop with a 3 GHz processor can perform around 3 billion calculations per second. While that is much faster than any human can achieve, it pales in comparison to HPC solutions that can perform quadrillions of calculations per second.</div>
HPC is specifically needed for these reasons:

* It paves the way for new innovations in science, technology, business and academia.
* It improves processing speeds, which can be critical for many kinds of computing operations, applications and workloads.
* It helps lay the foundation for a reliable, fast IT infrastructure that can store, process and analyze massive amounts of data for various applications.

'''Specification:'''

* Total Cores = 8064 (AMD)
* 238 Tflops
* 57 Compute Nodes
* 2 Login Nodes
* 2 Master Nodes
* 12 High memory nodes with local SSD
* Better performance of applications, which require high performance storage but do not scale well across nodes. E.g. Gaussian.

'''Memory'''

* 4 GB Per core for Compute Nodes, with a total memory = 29,184 GB
* 16 GB Per core for High Memory Nodes, with total memory = 12,288 GB

===='''Compute Specification:'''====
{| class=" wikitable" style="height: 49px;" width="1084"
|-
| width="115"|'''COMPUTE NODES'''
| width="162"|''' PROCESSOR'''
| width="86"|''' RAM'''
| width="79"|''' CORE'''
| width="86"|''' GPU '''
| width="91"|''' NO OF NODES'''
| width="81"|'''TOTAL RAM'''
| width="89"|'''TOTAL CORES'''
|}
{| class=" wikitable" style="height: 118px;" width="912"
|-
||Compute Node
||AMD EPYC 7763
||512 GB
||128
||-
||57
||29184GB
||7296
|-
||High Memory Node
||AMD EPYC 7543
||1024GB
||64
||-
||12
||12228GB
||768
|-
||GPU
||INTEL XEON PLATINUM 8358
||512GB
||64
||Tesla A100*4
||2
||1024GB
||128
|-
||Total
||
||
||
||
||
||42436
||8192
|-
||Storage
||750TB PFS
||
||
||
||
||
||
|}

===='''How to connect to Snu HPC ?'''====
 Enumerated here are the various types.

* '''SSH : '''Host magus02.snu.edu.in ( default port 22 )
* '''From Linux Machine : '''$ssh user@magus02.snu.edu.in
* '''From Windows Machine using ''''''putty.'''

===='''How to Reset Password on Magus02 ?'''====
 Step1. Login to magus02 account.

[[File:magus 02 pic 1.1716181911641.png|frameless|578x578px]]

Step2: Use Passwd Command.

[[File:magus 03 pic.1716182070345.png|frameless|558x558px]] 

==='''How to Connect to Magus from outside SNU Network ?'''===
To gain access to MEGUS from outside the SNU network, a VPN connection is mandatory. To acquire VPN access, kindly click on the [https://forms.office.com/Pages/ResponsePage.aspx?id=12YS_ftig0aFL0FyYUAqLN3tPd66H1xCnswYz1fby35UMkhBR0lMSlJLUlZGT0wwTzJTOUtLNEpZVy4u VPN Request Form] or utilize the option of accessing it by scanning the barcode provided below.

===='''How to Transfers data using WinSCP Windows (Client).'''====
Step1 : Open the WinSCP and fill all the details, After that click on the Login Button.

[[File:magus 04 pic.1716182129572.png|frameless|589x589px]]

Step 2 : Click Yes on Warning tab.

Step 3: Connected, Now you can transfer the data using Right click of key Panel.

[[File:magus o5 pic.1716182172254.png|left|frameless|530x530px]]

'''Applications on Megus'''

[[Vasp5.4|Vasp]], [[02HPC Quantum Espresso Magus02|Quantum espresso]], [[Gaussian16|Gaussian]], Siesta, [[Gromacs]], OpenFOAM.

==='''Queue Structure on Magus02.'''===
{| class="wikitable" style="background-color: #fafafa; height: 418px;" border="1" width="1523"
|-
| style="background-color:#bababa;" class="" width="60" |S.No.
| style="background-color:#bababa;" class="" width="115" |Processor Architecture
| style="background-color:#bababa;" class="" width="81" |Priority
| style="background-color:#bababa;" class="" width="147" |Queue Name
| style="background-color:#bababa;" class="" width="110" |Min no of Cores Required to submit job
| style="background-color:#bababa;" class="" width="99" |Max no of cores allowed per job
| style="background-color:#bababa;" class="" width="68" |No of nodes in the queue
| style="background-color:#bababa;" class="" width="92" |Wall Time
| style="background-color:#bababa;" class="" width="111" |MAX Cores User can Use in the Queue
| style="background-color:#bababa;" class="" width="111" |Max No. of Pending Jobs Per user
| style="background-color:#bababa;" class="" width="111" |Max No. of Running Jobs Per user
|-
| width="60" |'''1'''

| width="115" |
AMD EPYC 7763

| width="81" |
50

| width="147" |
Medium

| width="110" |
32

| width="99" |
128

| width="68" |
17

| width="92" |
1 week

| width="111" |
512

| width="111" | 2
| width="111" | 2
|-
| width="60" |'''2'''

| width="115" |
AMD EPYC 7543

| width="81" |
50

| width="147" |
Small

| width="110" |
8

| width="99" |
32

| width="68" |
16

| width="92" |
1 Month

| width="111" |
128

| width="111" |2
| width="111" |4
|-
| width="60" |'''3'''

| width="115" |
AMD EPYC 7543

| width="81" |
50

| width="147" |
High_mem

| width="110" |
32

| width="99" |
256

| width="68" |
12

| width="92" |
2 weeks

| width="111" |
1024

| width="111" |2
| width="111" |2
|-
| width="60" |'''4'''

| width="115" |
AMD EPYC 7763

| width="81" |
50

| width="147" |
Large

| width="110" |
64

| width="99" |
256

| width="68" |
24

| width="92" |
2 weeks

| width="111" |
1024

| width="111" |2
| width="111" |2
|-
| width="60" |'''5'''

| width="115" |
Intel Xeon Platinum 8358

| width="81" |
50

| width="147" |
GPU

| width="110" |
16

| width="99" |
16

| width="68" |
2

| width="92" |
1 week

| width="111" |
32

| width="111" |
| width="111" |2
|-
| width="60" |'''6'''

| width="115" |
AMD EPYC 7763

| width="81" |
Low

| width="147" |
Long

| width="110" |
32

| width="99" |
1024

| width="68" |
All lm nodes

| width="92" |
1 Month

| width="111" |
1024

| width="111" |
| width="111" |8
|}
====='''[[Slurm Command Reference:]]'''=====
'''[[index.php?title=Media:Linux Command Reference.pdf|Linux_Command_Reference.pdf]]'''

'''[Https://wiki.snu.edu.in/index.php?title=GPU MIG SLICE&preload= How to Use GPU MIG Slice]'''

GPU MIG SLICE

2026-10-12T05:33:38Z

Bhuvnesh: Created page with " === Why MIG is Essential for Magus === Our core HPC environment, '''Magus02''', houses a total of '''8 A100 GPU cards'''. Because our physical GPU footprint is highly concent..."

=== Why MIG is Essential for Magus ===
Our core HPC environment, '''Magus02''', houses a total of '''8 A100 GPU cards'''. Because our physical GPU footprint is highly concentrated, treating each A100 as an "all-or-nothing" resource creates severe scheduling bottlenecks and immense resource wastage.

Prior to this upgrade, a user running a small script that only required 5GB of memory would lock an entire 80GB card, effectively wasting 75GB of enterprise VRAM and blocking other researchers from working. With only 8 cards total on Magus02, this under-utilization is no longer sustainable. MIG directly solves this problem.

=== Why Use MIG? (And How It Benefits You) ===
MIG allows a single A100 physical GPU to be securely partitioned into 7 separate, isolated instances. This directly addresses the cluster bottlenecks discussed in our policy meeting by delivering:

* '''Drastically Reduced Resource Wastage:''' Instead of only 8 users locking down the entire cluster, Magus02 can now dynamically host up to 56 individual GPU jobs simultaneously, ensuring no VRAM sits idle.
* '''Faster Queue Times:''' You can now run small-to-medium jobs immediately without waiting days in the queue for a full 80GB GPU to become free.
* '''Hardware-Level Isolation:''' Each MIG slice has its own dedicated memory, cache, and compute cores. Your job's performance will remain entirely unaffected by other users sharing the same physical card.
* '''Support for Multi-Process Frameworks:''' Because each MIG instance operates as a fully distinct GPU device, your multi-process frameworks can seamlessly bind to individual slices. This allows for clean, parallel data-processing tasks without complex internal scheduling.

=== Updated GPU & CPU Configurations ===
To align with this shift toward efficiency, we have also '''lowered the minimum CPU requirement for GPU jobs from 16 cores to 4 cores''', allowing you to run much more compact and efficient workloads without wasting CPU cycles.
{| class="wikitable" style="color: rgb(34, 34, 34); font-size: small; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(255, 255, 255); text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; margin-bottom: 32px; font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;"
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Resource Type'''
| style="margin:0px;border:1px solid;line-height:1.15;" |'''VRAM'''
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Compute (SM)'''
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Slurm Directive'''
|- style="line-height: 1.15; margin-top: 0px;"
| style="margin:0px;border:1px solid;line-height:1.15;" |'''Full GPU'''
| style="margin:0px;border:1px solid;line-height:1.15;" |80GB
| style="margin:0px;border:1px solid;line-height:1.15;" |98 SMs
| style="margin:0px;border:1px solid;line-height:1.15;" |<code style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">#SBATCH --gres=gpu:nvidia_a100_80gb_pcie_7g.80gb:1</code>
|- style="line-height: 1.15; margin-top: 0px;"
| style="margin:0px;border:1px solid;line-height:1.15;" |'''MIG Slice (Single)'''
| style="margin:0px;border:1px solid;line-height:1.15;" |10GB
| style="margin:0px;border:1px solid;line-height:1.15;" |14 SMs
| style="margin:0px;border:1px solid;line-height:1.15;" |<code style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">#SBATCH --gres=gpu:nvidia_a100_80gb_pcie_1g.10gb:1</code>
|}

=== Combining & Using Multiple Slices in One Job ===
If your workload requires more than 10GB of VRAM or needs to scale across multiple independent GPU targets within a single job script, you can request multiple slices simultaneously.

==== 1. Requesting the Slices ====
To request multiple slices (for example, 2 slices to get 20GB of total VRAM), adjust your Slurm directive count at the end of the string:

'''''#SBATCH --gres=gpu:nvidia_a100_80gb_pcie_1g.10gb:2'''''

==== 2. Utilizing the Slices in Your Job ====
When you request multiple slices, Slurm will automatically provision them and expose them to your job environment.

* '''For Multi-Process/Multi-GPU Frameworks:''' Frameworks like PyTorch (DDP), TensorFlow, or custom multi-process Python scripts will automatically detect these as '''two distinct GPU devices''' (Device 0 and Device 1).
* '''For Parallel Task Execution:''' You can target specific slices within the same job submission using standard environment variables or Slurm arrays.

=== Troubleshooting & Device Visibility ===
Slurm automatically manages and sets up the environment for you. To verify your allocation or troubleshoot device visibility within your job script, you can print the assigned environment variables:
<span style="color: rgb(34, 34, 34); font-size: small; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(255, 255, 255); text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;"><div style="line-height: 1.15; margin-top: 0px;"><div style="line-height: 1.15; margin-top: 0px;"><div style="line-height: 1.15; margin-top: 0px;">'''''<div style="line-height: 1.15; margin-top: 0px;">echo "Allocated CUDA Devices: $CUDA_VISIBLE_DEVICES"</div>
<code role="text" style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">nvidia-smi
</code>
'''''</div></div></div><blockquote style="color: rgb(34, 34, 34); font-size: small; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(255, 255, 255); text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">https://fonts.gstatic.com/s/e/notoemoji/17.0/1f4a1/32.png '''Important Tip:''' If your code or framework is not detecting the assigned MIG slices, double-check your script to ensure you are not accidentally overwriting or resetting the <code style="font-family: "Google Sans Text", sans-serif; line-height: 1.15; margin-top: 0px;">$CUDA_VISIBLE_DEVICES</code> variable during your environment or module setup.</blockquote>

Main Page

2026-08-04T11:23:06Z

Bhuvnesh:

{| style="width: 100%;" width="1334" data-mce-style="height: 26px ;background-color:#005197;"
| style="vertical-align:middle;text-align:left;border-left-style:none;border-right-style:none;border-top-style:none;border-bottom-style:none;background-color:;font-family:Verdana, Arial, Helvetica, sans-serif;font-size:11px;border:1px dashed rgb(187, 187, 187);" class="col-white-bg" |
=== Information Technology Resource Centre : ===
'''<big>Configuration of Wi-Fi</big>''' [[File:down13.jpg|frameless|25x25px]]
* <big>[[Configuration of Wi-Fi for SNU-STAFF or SNU-STUDENT on Android Device]]</big>
* <big>[[Configuration of Wi-Fi for SNU-STAFF or SNU-STUDENT on Apple Device]]</big>
* <big>[[Configuration of Wi-Fi for SNU-STAFF or SNU-STUDENT on Window Device]]</big>
* [[Configuration of Wifi for Students|<big>Configuration of Wifi for Students</big>]]
*[[Configuration of Wifi System for Guest|<big>Configuration of Wifi System for Guest</big>]]
*[[How to Approve Guest Wifi Request|<big>How to Approve Guest Wifi Request</big>]]
*[[How to connect Eduroam WIfi|<big>How to connect Eduroam WIfi</big>]] 

'''<big>Net ID Setup</big>''' [[File:download (5).png|frameless|21x21px]]

*[[Setting up your NetID Password|<big>Setting up your NetID Password</big>]] : <big>(First Time Setup, Forget Password, Change Password)</big> 

'''<big>Email ID Setup</big>''' [[File:download (8).1716959100424.jpg|frameless|25x25px]]

* <big>[[Setting up your University Email ID Password.]]</big>

'''<big>Microsoft Teams Setup</big>''' [[File:RE1Mu3b.1747285649387.png|link=https://wiki.snu.edu.in/File:RE1Mu3b.1747285649387.png|frameless|101x101px]]

* <big>[[Microsoft Teams Installation and Guide]]</big>
* <big>[[Manage attendee audio permissions]]</big>
* <big>[[Schedule Microsoft Teams meeting from Microsoft Outlook]]</big>
* <big>[[How to Join Microsoft Teams meeting from web client]]</big>
* <big>[[Create a Class in MS Teams]]</big>
* <big>[[Change Meeting Permissions for Attendees through MS Teams]]</big>

'''<big>Configuration of VPN</big>''' [[File:download (6).1716183181171.png|frameless|27x27px]]
*[[How to connect VPN-IVANTI (Students)|<big>How to connect VPN-IVANTI (Students)</big>]]
*[[How to connect VPN-IVANTI (Staff)|<big>How to connect VPN-IVANTI (Staff)</big>]]

'''<big>High Performance Computing cluster (Magus)</big>''' [[File:images (2).png|frameless|42x42px]]
*<big>[[HPC|old HPC Magus01]]</big>

*<big>[[Magus02|HPC Cluster Magus02]]</big>
*<big>[[Open FOAM]]</big>
*<big>[[Nanoelectronics Computation Laboratory - NCL]]</big> 

'''<big>Kronos HPC LAB Cluster</big>'''
*<big>[[How to connect Kronos HPC Lab Cluster]]</big>
*<big>[[How to Transfers data using WinSCP Windows (Client)]]</big>

'''<big>Academia Installation</big>''' [[File:images.png|frameless|21x21px]]

*<big>[[Academia NativeApps Client Installation]]</big> 

<big>'''DIGII Campus Application'''</big> [[File:download5678.png|frameless|38x38px]]
*<big>[[How to create Request (For Night and Day Out Pass)]]</big>
*<big>[[How to download the Digii Campus Application]]</big>
*<big>[[How to login Collpoll Application- Parent's|How to login DigiiCampus Application- Parent's]]</big>
*<big>[[How to login in Digii (Formely Coll Poll)|How to login in DigiiCampus Application]] - '''For Students'''</big>
*[[Parental Approval Process|<big>Parental Approval Process</big>]] - <big>'''For Parent's'''</big> 

<big>'''Obtaining a Free Windows 11 Education License for Students'''</big> [[File:win11.1753764385890.jpg|frameless|38x38px]]
*<big>[[Active students at University are eligible to obtain a free Windows 11 Education license]]</big>

====='''<big>[[Get Informed: Frequently Asked Questions"]]</big>'''=====
|-
| style="border-top-style:none;" |
|}

Main Page

2026-08-04T11:20:45Z

Bhuvnesh:

Main Page

2026-08-04T11:16:17Z

Bhuvnesh:

Main Page

2026-08-04T11:14:00Z

Bhuvnesh:

{| class="greyscale" style="width: 100%;" width="1334" data-mce-style="height: 26px ;background-color:#005197;"
| style="vertical-align:middle;text-align:left;border-left-style:none;border-right-style:none;border-top-style:none;border-bottom-style:none;background-color:;font-family:Verdana, Arial, Helvetica, sans-serif;font-size:11px;border:1px dashed rgb(187, 187, 187);" class="col-white-bg" |
=== Information Technology Resource Centre : ===
'''<big>Configuration of Wi-Fi</big>''' [[File:down13.jpg|frameless|25x25px]]
* <big>[[Configuration of Wi-Fi for SNU-STAFF or SNU-STUDENT on Android Device]]</big>
* <big>[[Configuration of Wi-Fi for SNU-STAFF or SNU-STUDENT on Apple Device]]</big>
* <big>[[Configuration of Wi-Fi for SNU-STAFF or SNU-STUDENT on Window Device]]</big>
* [[Configuration of Wifi for Students|<big>Configuration of Wifi for Students</big>]]
*[[Configuration of Wifi System for Guest|<big>Configuration of Wifi System for Guest</big>]]
*[[How to Approve Guest Wifi Request|<big>How to Approve Guest Wifi Request</big>]]
*[[How to connect Eduroam WIfi|<big>How to connect Eduroam WIfi</big>]] 

'''<big>Net ID Setup</big>''' [[File:download (5).png|frameless|21x21px]]

*[[Setting up your NetID Password|<big>Setting up your NetID Password</big>]] : <big>(First Time Setup, Forget Password, Change Password)</big> 

'''<big>Email ID Setup</big>''' [[File:download (8).1716959100424.jpg|frameless|25x25px]]

* <big>[[Setting up your University Email ID Password.]]</big>

'''<big>Microsoft Teams Setup</big>''' [[File:RE1Mu3b.1747285649387.png|link=https://wiki.snu.edu.in/File:RE1Mu3b.1747285649387.png|frameless|101x101px]]

* <big>[[Microsoft Teams Installation and Guide]]</big>
* <big>[[Manage attendee audio permissions]]</big>
* <big>[[Schedule Microsoft Teams meeting from Microsoft Outlook]]</big>
* <big>[[How to Join Microsoft Teams meeting from web client]]</big>
* <big>[[Create a Class in MS Teams]]</big>
* <big>[[Change Meeting Permissions for Attendees through MS Teams]]</big>

'''<big>Configuration of VPN</big>''' [[File:download (6).1716183181171.png|frameless|27x27px]]
*[[How to connect VPN-IVANTI (Students)|<big>How to connect VPN-IVANTI (Students)</big>]]
*[[How to connect VPN-IVANTI (Staff)|<big>How to connect VPN-IVANTI (Staff)</big>]]

'''<big>High Performance Computing cluster (Magus)</big>''' [[File:images (2).png|frameless|42x42px]]
*<big>[[HPC|old HPC Magus01]]</big>

*<big>[[Magus02|HPC Cluster Magus02]]</big>
*<big>[[Open FOAM]]</big>
*<big>[[Nanoelectronics Computation Laboratory - NCL]]</big> 

'''<big>Kronos HPC LAB Cluster</big>'''
*<big>[[How to connect Kronos HPC Lab Cluster]]</big>
*<big>[[How to Transfers data using WinSCP Windows (Client)]]</big>

'''<big>Academia Installation</big>''' [[File:images.png|frameless|21x21px]]

*<big>[[Academia NativeApps Client Installation]]</big> 

<big>'''DIGII Campus Application'''</big> [[File:download5678.png|frameless|38x38px]]
*<big>[[How to create Request (For Night and Day Out Pass)]]</big>
*<big>[[How to download the Digii Campus Application]]</big>
*<big>[[How to login Collpoll Application- Parent's|How to login DigiiCampus Application- Parent's]]</big>
*<big>[[How to login in Digii (Formely Coll Poll)|How to login in DigiiCampus Application]] - '''For Students'''</big>
*[[Parental Approval Process|<big>Parental Approval Process</big>]] - <big>'''For Parent's'''</big> 

<big>'''Obtaining a Free Windows 11 Education License for Students'''</big> [[File:win11.1753764385890.jpg|frameless|38x38px]]
*<big>[[Active students at University are eligible to obtain a free Windows 11 Education license]]</big>

====='''<big>[[Get Informed: Frequently Asked Questions"]]</big>'''=====
|-
| style="border-top-style:none;" |
|}

File:flexiskin-images-favicon.png

2026-08-04T11:11:30Z