Thanks for your answer, it answers a part of my question. But what about this code that is pure C++ ? (can be seen on the previous link I sent) :
[cid:7fd825a6-4269-4ded-80ea-530ed872a545] How does one make sure that the server knows this task ? Best regards, Louis ________________________________ De : Pavel Tupitsyn <ptupit...@apache.org> Envoyé : lundi 8 avril 2024 11:34 À : user@ignite.apache.org <user@ignite.apache.org> Objet : Re: Use of distributed computing in C++ С++ client can execute Compute tasks that are written in Java and deployed to the server nodes: https://ignite.apache.org/releases/latest/cppdoc/classignite_1_1thin_1_1compute_1_1ComputeClient.html On Fri, Apr 5, 2024 at 3:41 PM Louis C <l...@outlook.fr<mailto:l...@outlook.fr>> wrote: Hello everyone, I was interested in the distributed computed in C++ in Ignite, as described here : https://ignite.apache.org/docs/latest/distributed-computing/distributed-computing I noted that the classes that will be executed must be known by the server, as said here : In order to run tasks on the remote nodes, make sure the class definitions of the tasks are available on the nodes. You can do this in two ways: * Add the classes to the classpath of the nodes; * Enable peer class loading<https://ignite.apache.org/docs/latest/code-deployment/peer-class-loading>. However these 2 points are mostly related to Java-based tasks, and I could not find anything regarding C++. So my question will be : how can a C++ client share an execution task on the server nodes ? Is it transparent ? Must we add a dll/.so on the server nodes ? In this case how to load this dll ? Thanks in advance, Louis C
