<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Rajesh Kumar on Medium]]></title>
        <description><![CDATA[Stories by Rajesh Kumar on Medium]]></description>
        <link>https://medium.com/@rajeshkumar?source=rss-1c7f42db26f3------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*_gM0birXnFWW5IgG.jpg</url>
            <title>Stories by Rajesh Kumar on Medium</title>
            <link>https://medium.com/@rajeshkumar?source=rss-1c7f42db26f3------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 04:49:35 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@rajeshkumar/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Executing CUDA C++ Application on a remote machine using Nsight Eclipse]]></title>
            <link>https://medium.com/@rajeshkumar/executing-cuda-c-application-on-a-remote-machine-using-nsight-eclipse-fb8364029625?source=rss-1c7f42db26f3------2</link>
            <guid isPermaLink="false">https://medium.com/p/fb8364029625</guid>
            <dc:creator><![CDATA[Rajesh Kumar]]></dc:creator>
            <pubDate>Wed, 21 Jan 2015 19:57:55 GMT</pubDate>
            <atom:updated>2015-01-21T20:05:34.194Z</atom:updated>
            <content:encoded><![CDATA[<p>We will be creating a CUDA C++ application on local machine. We will build the project locally and transfer the executable to the remote machine. The executable will run on the remote machine and the result will be transferred to the local machine. The local machine need not have a GPU but <a href="https://developer.nvidia.com/cuda-downloads">CUDA toolkit</a> (preferably version 6.5) must be installed here. Both the machines should be Linux machines.</p><p>Start Nsight Eclipse and follow the following steps.</p><ol><li>File -&gt; New -&gt; CUDA C/C++ Project</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CGRjN42WFFFyr1AaRcg-3w.png" /></figure><p>2. Enter a project name, select executable (Empty Project in this example), and click Next.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*85UY7iKx9Cz7twFET9yKvA.png" /></figure><p>3. Select device linker mode as ‘whole program compilation’, and versions of PTX and GPU code (you can select multiple). Click Next.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E6XcnWzbzb8G7y2K4psQGQ.png" /></figure><p>4. Click on Manage and enter the details (user name and ip address) of your remote system. Select system type as SSH. Finally click Finish.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*c2sC4ozg3BPEtOUZni1wKg.png" /></figure><p>If you get the following error while connecting to the remote machine, Go to Window — &gt; preferences — &gt; General -&gt; Network. Make the Active Provider -&gt; Direct, where all the proxy entries are unchecked.</p><p><em>Failed to connect sshd on “x.x.x.x”: ProxySOCKS5: com.jcraft.jsch.JSchException: java.net.SocketException: Can’t connect to SOCKS proxy:Connection refused</em></p><p>5. Create a folder on the remote machine and set the path of this folder as the Project Path. Browse the CUDA toolkit path from remote system. Also select the CPU architecture (x86–64 bit in this example) of the remote system. Click Finish.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xMItyUrb8RdAgNb4ilPjwQ.png" /></figure><p>6. File -&gt; New -&gt; File</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PxuiHw8n9PAQKpQgsyxNig.png" /></figure><p>7. Enter a file name (fileName.cu) and click Finish.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D9tdggszle_NRZZnPJiDHQ.png" /></figure><p>8. Write your program in the above file.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rxxppLEmP-Pj-9DtmrJWFQ.png" /></figure><p>9. Right click on the project folder and click on properties.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OW_PJh7klwxS7Crrlg75xg.png" /></figure><p>10. Build -&gt; Target Syatem. Confirm the remote system details.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YCoxqANeZikdKMaiJpCPGg.png" /></figure><p>11. Build the project locally. (Can use the hammer icon).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*shy5ADBCWjgz4FP1AfLJKQ.png" /></figure><p>12. Right click on the project folder. Properties -&gt; Run/Debug Settings -&gt; Edit. Go to remote tab. Choose remote connection as local. Select ‘upload local executable’. Browse the path of the executable from your local system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RzKbEIaDP41BvYU3ummLsg.png" /></figure><p>13. Go to local tab and provide the path of the file containing main function.</p><p>Click Apply / OK.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zA3v2zeX6CheVTkySYkngw.png" /></figure><p>14. Right click on the project folder. Run As -&gt; Remote C/C++ application.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WGeNGfeOVPcNY0GWjL-yYA.png" /></figure><p>15. Check the result in the console.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WVSF0bAJYB7sqpnfWBXBkg.png" /></figure><p>Cheers..!!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fb8364029625" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>