Differences

This shows you the differences between two versions of the page.

--- gpu [2017/07/17 09:52]
kocmanek [Performance tests]
+++ gpu [2017/09/11 11:26]
fucik
@@ Line 4: / Line 4: @@
 ===== Servers with GPU units =====
+GPU cluster ''gpu.q'' at Malá Strana:
-| machine                    | GPU; [[https://en.wikipedia.org/wiki/CUDA#Supported_GPUs|Capability]] [cc]  | cores | GPU RAM | Comment |
+| machine                    | GPU type | [[https://en.wikipedia.org/wiki/CUDA#GPUs_supported|cc]] | GPUs | GPU RAM | Comment |
-| titan                      | GeForce GTX 1080 Ti; cc6.1 | 1  | 11 GB           |  |
+| iridium                    | Quadro K2000        |  cc3.0 |   1|   2 GB |  |
-| titan-gpu                  | GeForce GTX Titan Z; cc3.5 | 2  | 6 GB each core  |  |
+| titan-gpu                  | GeForce GTX Titan Z |  cc3.5 |   2|   6 GB |  |
-| twister1; twister2; kronos | Tesla K40c; cc3.5          | 1  | 12 GB           |  |
+| twister1; twister2; kronos | Tesla K40c          |  cc3.5 |   1|  12 GB |  |
-| iridium                    | Quadro K2000; cc3.0        | 1  | 2 GB            |  |
+| dll1; dll2                 | GeForce GTX 1080    |  cc6.1 |   8|   8 GB |  |
-| victoria; arc              | GeForce GT 630; cc3.0      | 1  | 2 GB            | desktop machine |
+| titan                      | GeForce GTX 1080    |  cc6.1 |   1|   8 GB |  |
-| athena                     | GeForce GTX 1080; cc6.1    | 1  | 8 GB            | Tom's desktop machine |
+| dll3; dll4; dll5           | GeForce GTX 1080 Ti |  cc6.1 |  10|  11 GB | dll3 has only 9 GPUs since 2017/07 |
-| dll1; dll2                 | GeForce GTX 1080; cc6.1    | 8  | 8 GB each core  |  |
+| dll6                       | GeForce GTX 1080 Ti |  cc6.1 |   3|  11 GB |  |
-| dll3; dll4; dll5           | GeForce GTX 1080 Ti; cc6.1 | 10 | 11 GB each core |  |
-not used at the moment: GeForce GTX 570 (from twister2)
+Desktop machines:
+| machine                    | GPU type | [[https://en.wikipedia.org/wiki/CUDA#GPUs_supported|cc]] | GPUs | GPU RAM | Comment |
+| victoria; arc              | GeForce GT 630   | cc3.0 |  1 |  2 GB | desktop machine |
+| athena                     | GeForce GTX 1080 | cc6.1 |  1 |  8 GB | Tom's desktop machine |
+Not used at the moment: GeForce GTX 570 (from twister2)
 All machines have CUDA8.0 and should support both Theano and TensorFlow.
-Summary of future plans:
+=== Disk space ===
-  * Current Troja servers won't get any GPUs (the only option would be [[http://www.czc.cz/hp-quadro-k1200-4gb/171662/produkt?ppcbee-adtext-variant=Produkt%3B+kategorie+%2B+cena%3B+Pobo%C4%8Dky&gclid=CKbKkbrWrswCFQUq0wodHDELCw|Quadro K1200 4GB]], horribly cost-inefficient)
+All the GPU machines are at Malá Strana (not at Troja), so you should not use ''/lnet/tspec/work/'', but you should use:
-  * The old Quadro K2000 we have is a much more low end piece, so we can't test is in Troja.
+- ''/lnet/spec/work/'' (alias ''/net/work/'') - Lustre disk space at Malá Strana
-  * There is MetaCentrum which also has GPUs, so testing can be done there.
+- ''/net/cluster/TMP'' - NFS hard disk for temporary files, so slower than Lustre for most tasks
-  * It is impossible (wasteful in terms of space and forbidden by a dean regulation) to put non-rack machines to our servers rooms. So we won't be buying GeForce GTX 1080 (~20000CZK, out of stock now), for a non-rack machine since we most likely don't have any available.
+- ''/net/cluster/SSD'' - also NFS, but faster then TMP because of SSD
-  * Yes, there are grant applications under review which include rack machines with GPUs, e.g. 5x2 or something like that; more will be known in 2017.
+- ''/COMP.TMP'' - local (for each machine) space for temporary files (use it instead of ''/tmp''; over-filling ''/COMP.TMP'' should not halt the system).
 === Individual acquisitions: NVIDIA Academic Hardware Grants ==
@@ Line 43: / Line 47: @@
 In this section will be explained how to use cluster properly.
+==== Set-up CUDA and CUDNN ====
+You can add following command into your ~/.bashrc
+  CUDNN_version=6.0
+  CUDA_version=8.0
+  CUDA_DIR_OPT=/opt/cuda-$CUDA_version
+  if [ -d "$CUDA_DIR_OPT" ] ; then
+    CUDA_DIR=$CUDA_DIR_OPT
+    export CUDA_HOME=$CUDA_DIR
+    export THEANO_FLAGS="cuda.root=$CUDA_HOME,device=gpu,floatX=float32"
+    export PATH=$PATH:$CUDA_DIR/bin
+    export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CUDA_DIR/cudnn/$CUDNN_version/lib64:$CUDA_DIR/lib64
+    export CPATH=$CUDA_DIR/cudnn/$CUDNN_version/include:$CPATH
+  fi
 ==== TensorFlow Environment ====
@@ Line 108: / Line 129: @@
 * [[http://www.trustedreviews.com/nvidia-geforce-gtx-1080-review-performance-benchmarks-and-conclusion-page-2| 980 vs 1080 vs Titan X (not the Titan Z we have)]]
-In the following table is the experiment conducted by Tom Kocmi. You can replicate experiment: /a/merkur3/kocmanek/Projects/GPUBenchmark (you will need to prepare environment of TensorFlow11 or use my ANACONDA). The benchmark uses 2GB model.
+In the following table is the experiment conducted by Tom Kocmi. You can replicate experiment: /a/merkur3/kocmanek/Projects/GPUBenchmark (you will need to prepare environment of TensorFlow11 or use my ANACONDA). The benchmark uses 2GB model of seq2seq machine translation in Neural Monkey (De > EN). If not specified, the benchmark had an access only to one GPU.
 | machine | Setup; CPU/GPU; [[https://en.wikipedia.org/wiki/CUDA#Supported_GPUs|Capability]] [cc] | Walltime | Note |
 | athena     | GeForce GTX 1080; cc6.1            |    9:55:58 | Tom's desktop  |
-| dll2       | (2 GPU) GeForce GTX 1080; cc6.1    |   10:19:40 | with CUDA_VISIBLE_DEVICES=0 |
+| dll2       | GeForce GTX 1080; cc6.1            |   10:19:40 |  |
 | titan      | GeForce GTX 1080 Ti                |   10:45:11 | (new result with correct CUDA version) |
 | dll1       | (2 GPU) GeForce GTX 1080; cc6.1    |   12:34:34 | Probably only one GPU was used |
-| dll2       | (2 GPU) GeForce GTX 1080; cc6.1    |   13:01:05 | Only one GPU was used |
+| dll2       | GeForce GTX 1080; cc6.1            |   13:01:05 | Only one GPU was used |
 | titan-gpu  | (2 GPU) GeForce GTX Titan Z; cc3.5 |   16:05:24 | Probably only one GPU was used |
 | kronos-dev | Tesla K40c; cc3.5                  |   22:41:01 |  |
@@ Line 130: / Line 151: @@
 === Second Benchmark ===
-The previous benchmark only compares the speed of processing units within the GPUs and do not take into account the size of memory. Therefore I have conducted another benchmark, this time for each graphic card I have increased the batch size as much as possible so the model still could fit into the GPU (the previous benchmark had batch size 20). This way the results should be more representative of the power for each GPU.
+The previous benchmark only compares the speed of processing units within the GPUs and do not take into account the size of memory. Therefore I have conducted another benchmark, this time for each graphic card I have increased the batch size as much as possible so the model still could fit into the GPU (the previous benchmark model had batch size 20). This way the results should be more representative of the power for each GPU.
 | GPU; Cuda capability       | GPU RAM |  Walltime | Batch size | Machine |
-| Tesla K40c; cc3.5          |   12 GB |           |       2400 | kronos |
+| Tesla K40c; cc3.5          |   12 GB |           |        |  |
 | GeForce GTX 1080 Ti; cc6.1 |   11 GB |  00:55:56 |       2300 | dll5 |
 | GeForce GTX 1080; cc6.1    |    8 GB |  01:10:57 |       1700 | dll1 |

[ Back to the navigation ] [ Back to the content ]

Institute of Formal and Applied Linguistics Wiki

Differences