RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.
Add another workspace memory resource that does not have the explicit memory limit. That is, after the change we have the following:
rmm::mr::get_current_device_resource() is default for all allocations, as before. It is used for the allocations with unlimited lifetime, e.g. returned to the user.
raft::get_workspace_resource() is for temporary allocations and forced to have fixed size, as before. However, it becomes smaller and should be used only for allocations, which do not scale with problem size. It defaults to a thin layer on top of the current_device_resource.
raft::get_large_workspace_resource()(new) is for temporary allocations, which can scale with the problem size. Unlike workspace_resource, its size is not fixed. By default, it points to the current_device_resource, but the user can set it to something backed by the host memory (e.g. managed memory) to avoid OOM exceptions when there's not enough device memory left.
Problem
We have a list of issues/preference/requirements, some of which contradict others
We rely on RMM to handle all allocations and we often use rmm::mr::pool_memory_resource for performance reasons (to avoid lots of cudaMalloc calls in the loops)
However, the design goal is to avoid setting allocators on our own and to give the full control to the user (hence the workaround in 2 was removed).
We introduced the workspace resource earlier to allow querying the available memory reliably and maximize the batch sizes accordingly (see also issue #1310). Without this, some of our batched algorithms either fail with OOM or severely underperform due to small batch sizes.
However, we cannot just put all of RAFT temporary allocations into the limited workspace_resource, because some of them scale with the problem size and would inevitably fail with OOM at some point.
Setting the workspace resource to the managed memory is not advisable as well for performance reasons: we have lots of small allocations in performance critical sections, so we need a pool, but a pool in the managed memory inevitably outgrows the device memory and makes the whole program slow.
Solution
I propose to split the workspace memory into two:
small, fixed-size workspace for small, frequent allocations
large workspace for the allocations that scale with the problem size
Notes:
We still leave the full control over the allocator types to the user.
Neither of the workspace resource should have unlimited lifetime / returned to the user. As a result, if the user sets the managed memory as the large workspace resource, the memory is guaranteed to be released after the function call.
We have the option to use the slow managed memory without a pool for large allocations, while still using a fast pool for small allocations.
We have more flexible control over which allocations are "large" and which are "small", so hopefully using the managed memory is not so bad for performance.
Brief
Add another workspace memory resource that does not have the explicit memory limit. That is, after the change we have the following:
rmm::mr::get_current_device_resource()
is default for all allocations, as before. It is used for the allocations with unlimited lifetime, e.g. returned to the user.raft::get_workspace_resource()
is for temporary allocations and forced to have fixed size, as before. However, it becomes smaller and should be used only for allocations, which do not scale with problem size. It defaults to a thin layer on top of thecurrent_device_resource
.raft::get_large_workspace_resource()
(new) is for temporary allocations, which can scale with the problem size. Unlikeworkspace_resource
, its size is not fixed. By default, it points to thecurrent_device_resource
, but the user can set it to something backed by the host memory (e.g. managed memory) to avoid OOM exceptions when there's not enough device memory left.Problem
We have a list of issues/preference/requirements, some of which contradict others
rmm::mr::pool_memory_resource
for performance reasons (to avoid lots of cudaMalloc calls in the loops)workspace_resource
, because some of them scale with the problem size and would inevitably fail with OOM at some point.Solution
I propose to split the workspace memory into two:
Notes: