RACBVHs: Random-Accessible Compressed Bounding Volume Hierarchies

1. RACBVHs: Random-Accessible Compressed Bounding Volume Hierarchies Tae-Joon Kim Bochang Moon Duksu Kim Sung-EuiYoon KAIST (Korea Advanced Institute of Science and Technology) To appear at IEEE TVCG 09

2. Goal • Design a compact representation for interactive geometric and graphics applications of massive models • Support random access • Support various applications • Improve performance of applications

3. Motivation – Massive Models • Take high disk/memory spaces • Long data access time and low I/O performance Power plant with Hugo model(82M triangles) St. Matthew model(372M triangles)

4. Low Growth Rate of Data Access Speed Accumulated growth rate during 1999~2009 (log scale) Reducing data access time is critical! access speed disk access speed

5. Large Scale Applications • Ray tracing • Collision detection • Visibility queries • Dynamic simulation • Motion planning

6. Large Scale Applications • Common geometric data structures • Meshes • Acceleration hierarchies • k-d trees • Bounding volume hierarchies (BVHs) • Usually require random access on meshes and hierarchies Supporting random access is important!

7. Our Approach • Compress bounding volume hierarchies (BVHs) to reduce expensive data access time • Support random access • Support fast runtime decompression for performance improvements • Provide general API for various applications Random-Accessible Compressed BVHs

8. Ray Tracing of St. Matthew Model • 128M triangles (4GB) • 256M BVH (8GB) • 9.6:1 compression ratios for the BVH over original uncompressed BVH • 4.4:1 runtime performance improvement

9. Related Work • Mesh compression • Compression and random access • Tree and BVH compression

10. Mesh Compression • Designed to achieve a maximum compression ratio or efficient transmission • [Touma and Gotsman 98, Devillers and Gandoin 00] • Encode vertices • [Alliez and Desbrun 01, Kälberer et al. 05] • Encode edges • [Isenburg and Snoeyink 00] • Encode faces • [Gumhold and Strasser 98, Rossignac 99, Lee et al. 02] Do not directly support random access

11. Compression and Random Access • Quite common in video & audio encoding • E.g., MPEG video • Single or multi-resolution mesh compression • [Choe et al. 04, Yoon and Lindstrom 07, Choe et al. 09] • [Gobbetti et al. 06, Kim et al. 06] • Regular volumetric grids and images • [Ihm and Park 99, Rodler 99, Lefebvre and Hoppe 06]

12. Tree and BVH Compression • Tree compression • [Zaks 80, Zerling 85, Katajainen and Makinen 90] • Do not support random access • Encode bounding volumes • Quantization: [Cline et al. 06, Mahovsky 05, Terdiman 03] • Hierarchical encoding: [Rusinkiewicz and Levoy 00, Hubo et al. 06] • Encode tree structures • Assume a particular tree structure: [Lauterbach et al. 07 and 08] • Assume a particular layout for the tree: [Lefebvre and Hoppe 07]

13. Outline • Overview • Compression at preprocessing • Runtime data access framework • Results

14. Outline • Overview • Compression at preprocessing • Runtime data access framework • Results

15. Overview - Compression at Preprocessing BVH • Decompose the BVHs into set of clusters • Compress each cluster separately • Each cluster serves as an access point C1 C2 C3

16. Overview - Runtime BVH Access Framework Main memory Offset table Cluster c0 0 Cluster c1 1 Cached data Applications 2 Cluster c2 3 Cluster c3 Request data Requested data External drive 4 Cluster c4 5 Cluster c5 Compressed data pool … … … n Cluster cn Decompress using CPU power

17. Overview - Runtime BVH Access Framework Main memory Offset table Cluster c0 0 Cluster c1 1 Cached data Applications 2 Cluster c2 3 Cluster c3 Request data Requested data 4 Cluster c4 5 Cluster c5 Compressed data pool … … … n Cluster cn Preloaded data

18. Outline • Overview • Compression at preprocessing • Runtime data access framework • Results

19. Bounding Volume Hierarchies (BVHs) BV • A node of BVHs has: • Indices of children nodes • A bounding volume (BV) • We use the axis-aligned bounding box (AABB) and a binary BVH • Due to its simplicity and the wide use • Can be easilyextended to other types of BVs and k-ary BVHs Index of child node

20. Layouts of BVHs Layout of a BVH • Order of stored BV nodes in memory • Affect the performance of applications • Our method preserves the original layouts • Support any layouts • Use cache-efficient layouts for the best performance[Yoon and Manocha 06]

21. Cluster-based Compression • Cluster has: • Fixed number (e.g. 4K) of consecutive nodes in the layouts • Compute clusters with cache-coherent nodes • Implicitly performed by using cache-coherent layouts of BVHs C1 C2 C3

22. Encoding Bounding Volumes Parent BV Left BV • Quantize min and max extents of BVs • Predict child BVs using their parent BV • Use the simple median prediction • Encode prediction errors • Use a dictionary-based encoder for fast decompression Right BV Prediction error

23. Front-based Tree Structure Encoding n0 n0 • Maintain a front of nodes, one of whose child nodes is uncompressed yet • Average size of front: 13 • 4K nodes with cache-efficient layouts • Encode tree structures using only a few bits by connecting uncompressed nodes to a node in the front n1 n1 n2 n2 n3 n3 n4 n4 n9 n12 n5 n5 n6 n7 n8 n10 n11 n13 n14 0 1 2 3 n2 n3 n4 n5 Front : Not yet compressed nodes : Compressed nodes

24. Outline • Overview • Compression at preprocessing • Runtime data access framework • Results

25. BVH Access API • Various applications can transparently access our representation using the following API: • GetRootIndex (.) • GetBV (.) • IsLeaf (.) • GetLeftChildIndex (.) • GetRightChildIndex (.) • GetTriangleIndex (.) Offset table Cluster c0 0 Main memory Cluster c1 1 Cached data 2 Clusterc2 3 Applications Cluster c3 External drive 4 Cluster c4 Decompression 5 Cluster c5 Compressed data pool … … … n Cluster cn

26. Outline • Overview • Compression at preprocessing • Runtime data access framework • Results

27. Compression Results • 16 bits quantization for BVs • 4K nodes per cluster

28. Benchmark Applications • Ray tracing • Typically traverses large portions of BVHs • Collision detection • Accesses smaller and more localized portions of BVHs than ray tracing

29. Video for Benchmark Applications

30. Performance Improvement • Mainly due to higher I/O performance • Reduce the data size by using compression→ Reduce or remove the expensive disk I/O time • Use CPU power to efficiently decompress the data

31. Parallel Random Access • Ray tracing of St. Matthew model (128M tri.) using only primary rays Speedup

32. Limitation • Lower (by up to 3%) the performance with small models which can fit in main memory • Overhead of identifying clusters • Less tight BVs due to a conservative quantization

33. Summary & Conclusion • Low storage requirement • Achieve up to a 12:1 compression ratio • Improved performance with random accessibility • Achieve a 4:1 runtime performance improvement • Wide applicability • Allows various applications to access the compressed BVHs • High scalability • Achieve a 3.6:1 performance improvement when using 4 threads over single thread

34. Future Work • Compact in-core representations • Apply to highly parallel architectures • GPU • Larrabee • Apply to levels-of-detail (LOD) hierarchies

35. Acknowledgments • Members of SGLab. in KAIST • Model contributors • Funding agencies • KAIST seed grant • Ministry of Knowledge Economy • Samsung • Microsoft Research Asia • Korea Research Foundation

36. Any Questions? Q & A Thank you!