Ke Meng

Hi, @lilux618. libgrape-lite is not ready for the graph500 benchmark and BFS has some specific algorithmic optimizations that libgrape-lite does not apply. This is because these optimizations are hard to...

> how can I change it in source code level? The BFS source is passed in via gflags. Specifically, the source is set at [here](https://github.com/alibaba/libgrape-lite/blob/0d0471ce2dcf90f4544c489e3af086bee752e320/examples/analytical_apps/run_cuda_app.h#L184). > Further more, how do...

The definition of `setColumnDefsAndConstraints` is in header file, and it tires to call [dynamic_cast](https://github.com/hyrise/sql-parser/blob/master/src/sql/CreateStatement.h#L74). LLVM/MLIR project [implicitly add](https://github.com/llvm/llvm-project/blob/7f9e6f6fa6547b9c5648d96fd96920467c8b3e84/llvm/cmake/modules/AddLLVM.cmake#L47) `-fno-rtti` flag (this flag disable `dynamic_cast`) and this will cause a compilation...

Hi @Bouncner, any chance to get this landed? This fix does not introduce any changes but only removes the function definition from the header file.

> In the long run, we would like to remove the cluttered `statements.cpp`. For that, could you please move the function to a new `CreateStatement.cpp`? Hi @Bouncner, the above issue...

Hi @is-shidian, vineyard currently can only share data on host memory, while many training/inference/sampling engines leverage GPU to accelerate tensor-centric task, so data sharing also has a high probability of...

@Y-jiji Yes

@Y-jiji For example, A graph task(learning like GNN, analytics like Label propagation) loads the topology of a real-world graph into GPU but only traverses the graph without changing it (Reloading...

Hi @CSWater, I have left some comments for you, I suggest you consider the following at an early stage: - Make the GPU-feature optional. - Allocate the GPU memory on-demand,...

Hi @CSWater, I have following suggestion: - We should pushdown the `ENABLE_GPU` marcos in `/gpu` to make the regular code path more clear. GPU-related methods are always callable but return...