Programming/Cuda - Revision history

Pysdlb at 14:03, 5 January 2023

2023-01-05T14:03:12Z

2023-01-05T14:01:42Z

2022-11-16T11:04:08Z

2022-11-10T16:26:05Z

2022-11-08T09:54:27Z

2022-11-07T16:56:37Z

2022-11-07T16:52:03Z

2022-11-02T12:27:05Z

2021-10-26T13:47:56Z

2021-05-24T09:10:25Z

@@ Line 45: / Line 45: @@
 For applications that require this information:
-* A40 is >6.0
+* A40 is 8.5
 * P100 is 6.0
@@ Line 170: / Line 170: @@
 * module add cuda/8.0.61
 * module add cuda/9.0.176
-* module add cuda/11.0.3 (to be removed - superseded)
+* module add cuda/10.1.168
 * module add cuda/11.5.0

@@ Line 168: / Line 168: @@
 The following modules are available:
 * module add cuda/8.0.61
 * module add cuda/9.0.176

← Older revision		Revision as of 11:04, 16 November 2022
Line 226:		Line 226:
	* [https://www.youtube.com/watch?v=_41LCMFpsFs&t=6s 'Youtube - CUDA training video']		* [https://www.youtube.com/watch?v=_41LCMFpsFs&t=6s 'Youtube - CUDA training video']

−	~~==Navigation==~~	+	{{Librariespagenav}}
−
−	* [[Main_Page\|Home]]
−	* [[Applications\|Application support]]
−	* [[General\|General]]
−	~~* [[Programming\|Programming support]] *~~

@@ Line 5: / Line 5: @@
 {|
 |style="width:5%; border-width: 0;cellpadding=0" | [[File:icon_tick.png]]
-|style="width:95%; border-width: 0;cellpadding=0" | The CUDA platform is designed to work with programming languages such as [[programming/C|C]], [[programming/C-Plusplus|C++]]. Fortran CUDA is possible through the use of PGI-FORTRAN, which is now available.
+|style="width:95%; border-width: 0;cellpadding=0" | The CUDA platform is designed to work with programming languages such as [[programming/C|C]], [[programming/C-Plusplus|C++]].
 |-
 |}

@@ Line 26: / Line 26: @@
 ====NVidia A40====
-Key features of the Tesla A40 GPU accelerator include:
+Key features of the Ampere A40 GPU accelerator include:
 * 48 GB GDDDR6 with ECC (696 GB/s bandwidth)

@@ Line 22: / Line 22: @@
 ===Specification===
-Viper has 4 K40m (GPU01-GPU04) and 1 P100 (GPU05)
+Viper has 4 A40 (GPU01-GPU04) and 1 P100 (GPU05)
 ====NVidia K40m====
-Key features of the Tesla K40 GPU accelerator include:
+Key features of the Tesla A40 GPU accelerator include:
-* 12GB of ultra-fast GDDR5 memory allows users to process 2X larger datasets, enabling them to rapidly analyze massive volumes of data.
+* 48 GB GDDDR6 with ECC (696 GB/s bandwidth)
-* 2,880 CUDA® parallel processing cores deliver application acceleration by up to 10X compared to using a CPU alone.
+* 10,752 CUDA-based cores
-* Dynamic Parallelism enables GPU threads to dynamically spawn new threads, enabling users to quickly and easily crunch through adaptive and dynamic data structures.
+* 84 Second-generation RT core
-* PCIe Gen-3 interconnect support accelerates data movement by 2X compared to PCIe Gen-2 technology.
+* 336 Third-generation Tensor cores
 ====NVidia P100====
@@ Line 199: / Line 199: @@
 #SBATCH -e %N.%j.%a.err
 #SBATCH -p gpu
-#SBATCH --gres=gpu:tesla
+#SBATCH --gres=gpu
 #SBATCH --exclusive
 #SBATCH --mail-user= your email address here

@@ Line 5: / Line 5: @@
 {|
 |style="width:5%; border-width: 0;cellpadding=0" | [[File:icon_tick.png]]
-|style="width:95%; border-width: 0;cellpadding=0" | The CUDA platform is designed to work with programming languages such as [[programming/C|C]], [[programming/C-Plusplus|C++]]. Fortran CUDA is possible through the use of PGI-fortran, which is now available.
+|style="width:95%; border-width: 0;cellpadding=0" | The CUDA platform is designed to work with programming languages such as [[programming/C|C]], [[programming/C-Plusplus|C++]]. Fortran CUDA is possible through the use of PGI-FORTRAN, which is now available.
 |-
 |}
@@ Line 221: / Line 221: @@
 *MPI (CPU nodes only)
-== Further Information ==
+== Next Steps ==
 * [http://www.nvidia.com/object/cuda_home_new.html http://www.nvidia.com/object/cuda_home_new.html]

@@ Line 1: / Line 1: @@
 == Programming Details ==
-CUDA is a parallel computing platform and application programming interface (API) model created by NVidia. It allows you to program a CUDA-enabled graphics processing unit (GPU) for general-purpose processing.
+CUDA is a parallel computing platform and application programming interface (API) model created by Nvidia. It allows you to program a CUDA-enabled graphics processing unit (GPU) for general-purpose processing.
 {|
@@ Line 96: / Line 96: @@
 ====Profiling====
-NVidia provide a good profiling tool call nvprof; its invocation is shown below:
+Nvidia provide a good profiling tool call nvprof; its invocation is shown below:
 <pre>
-$ module load cuda/9.0.176
+$ module load cuda/11.5.0
 $ nvprof ./mycudaprogram
 </pre>
@@ Line 172: / Line 172: @@
 * module add cuda/8.0.61
 * module add cuda/9.0.176
+* module add cuda/11.0.3 (to be removed - superseded)
 ==== Compilation ====
@@ Line 179: / Line 180: @@
 <pre style="background-color: black; color: white; border: 2px solid black; font-family: monospace, sans-serif;">
-[username@login01 ~]$ module add cuda/9.0.176
+[username@login01 ~]$ module add cuda/11.5.0
 [username@login01 ~]$ nvcc -o testGPU testGPU.cu
 </pre>
@@ Line 202: / Line 203: @@
 #SBATCH --mail-user= your email address here
-module add cuda/10.1.168
+module add cuda/11.5.0
 /home/user/CUDA/testGPU

@@ Line 1: / Line 1: @@
 == Programming Details ==
-CUDA is a parallel computing platform and application programming interface (API) model created by NVidia. It allows you to program a CUDA-enabled graphics processing unit (GPU) for general purpose processing.
+CUDA is a parallel computing platform and application programming interface (API) model created by NVidia. It allows you to program a CUDA-enabled graphics processing unit (GPU) for general-purpose processing.
 {|
@@ Line 37: / Line 37: @@
 Key features of the P100 GPU accelerator include:
-* 16GB HBM2 Memory with a  Type PCI Express 3.0 x16 interface (bandwidth 720 GBps)
+* 16GB HBM2 Memory with a  Type PCI Express 3.0 x16 interface (bandwidth 720 Gbps)
-* 3584 CUDA cores graphics Engine NVIDIA Tesla P100
+* 3584 CUDA cores Graphics Engine NVIDIA Tesla P100
 * Bus Type PCI Express 3.0 x16, API Supported OpenCL, OpenACC
@@ Line 200: / Line 200: @@
 #SBATCH --gres=gpu:tesla
 #SBATCH --exclusive
 module add cuda/10.1.168