NVIDIA RTX Memory Utility (RTXMU)

Reduce memory consumption with an open-source solution.

RTXMU Product Benefits

RTXMU reduces the time it takes for a developer to integrate compaction and suballocation into an NVIDIA RTX™ title.

For applications using RTXMU, NVIDIA RTX GPUs get a ~50% reduction in memory footprint.

Scenes with thousands of small unique bottom-level acceleration structures (BLAS) benefit greatly from suballocation.

To support as many developers as possible, RTXMU is open source and available on GitHub.

Meet NVIDIA experts in DLSS, path tracing, and generative AI at this year's GDC.

The AS suballocator works around the 64 kilobyte (KB) / 4 megabyte (MB) buffer-alignment requirement by placing small AS allocations within a larger memory heap.

The AS suballocator still needs to fulfil the 256 byte (B) alignment required for AS allocations.

If the application requests 4MB or larger suballocation blocks, RTXMU uses placed resources with heaps that can provide 4MB alignment.

Sun Temple using RTXMU

Zero Day using RTXMU

If the build requests compaction, RTXMU will request the compaction size be written out to a chunk of video memory.

Once the compaction size has been copied from video memory to system memory, RTXMU allocates a suballocated compaction buffer to be used as the destination for the compaction copy.