cuSOLVER 自动并行计算许多矩阵?

cuSOLVER automatically parallelizes computation on many matrices?

我必须进行相同的计算(例如,获取 A1, A2, ... 的特征值)on many(>10^15) 矩阵,因此我希望尽可能多地使用线程。

但是我找不到说明线程数的 cuBLAS 或 cuSOLVER 代码。如果我使用 for 循环和 cuSOLVER 函数编写代码,cuSOLVER 会自动分配资源并并行计算吗?
或者是否有任何 cuSOLVER 或 cuBLAS API 可以控制线程数和并行化函数? ......

Does cuSOLVER automatically distribute resources and parallelize computations if I write code with for loop and cuSOLVER functions?

没有

Or is there any cuSOLVER or cuBLAS API that I can control the number of threads and parallelize functions?

没有

但是,如果您想阅读 CUSOLVER documentation, you will see that there is a batched sparse QR factorization routine. This can be used to solve eigenvalue problems