{"id":6518,"date":"2016-01-29T09:00:35","date_gmt":"2016-01-29T14:00:35","guid":{"rendered":"https:\/\/blogs.mathworks.com\/pick\/?p=6518"},"modified":"2016-01-07T17:19:39","modified_gmt":"2016-01-07T22:19:39","slug":"2d-wave-equation","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/pick\/2016\/01\/29\/2d-wave-equation\/","title":{"rendered":"2D Wave Equation"},"content":{"rendered":"
\r\n \r\n

Sean<\/a>'s pick this week is 2D Wave Equation<\/a> by Daniel Armyr<\/a>.\r\n <\/p>\r\n <\/introduction> \r\n

GPUs<\/a><\/h3>\r\n

MATLAB's Parallel Computing Toolbox<\/a> has direct support for Graphics Processing Units (GPUs or GPGPUs) for many different computations. Since most laptops, with\r\n the exception of high-end gaming ones, don't have powerful enough GPUs for scientific computing, I usually don't get to take\r\n advantage of this.\r\n <\/p>\r\n

However, I've been lucky enough over the last month or so to have access to a fancy portable desktop with a Tesla K40 GPU<\/a> in it. Daniel's app allows me to make waves and then compute the propagation on the cpu or gpu while measuring the difference\r\n in computational speed. (I switch to the GPU at 28s)\r\n <\/p>\r\n

\r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n \r\n <\/object>\r\n \r\n <\/p>\r\n

What did it take to switch over to the GPU from the CPU? Here's the code comparison<\/a>:\r\n <\/p>\r\n

<\/p>\r\n

Just the commands gpuArray<\/tt> and gather<\/tt> to move the arrays back and forth between the CPU and GPU. There are quite a few commands that run directly on the GPU,\r\n you can find them here<\/a>. This list grows bigger with every release of MATLAB. My personal favorite is iradon<\/tt> which was a big bottleneck in my previous life working with CT images.\r\n <\/p>\r\n

One thing I'd like to see Daniel do differently would be to check if a GPU is available. If you don't have a supported GPU,\r\n the app errors rather than failing gracefully or just disabling the GPU option. There's a convenient function, gpuDeviceCount<\/tt>, that you can use to see how many supported GPUs are on the machine.\r\n <\/p>

if<\/span> gpuDeviceCount == 0\r\n    warndlg(sprintf('You don''t have a supported GPU!\\nGo talk to your IT department :)'<\/span>),'No GPU Found'<\/span>)\r\nend<\/span><\/pre> 
Comments<\/a><\/h3>\r\n   
Do you use a GPU with MATLAB or have interest in trying one?<\/p>\r\n   
Give it a try and let us know what you think here<\/a> or leave a comment<\/a> for Daniel.\r\n   <\/p>