department of electronic engineering, tsinghua university nano-scale integrated circuit and system...
TRANSCRIPT
Department of Electronic Engineering, Tsinghua University
1Nano-scale Integrated Circuit and System Lab.
GPU Sparse LU Factorization and Its Application in Circuit Simulation
Nano-scale Integrated Circuit and System Lab.,EE Department, Tsinghua University
Ling Ren
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
2
Abstract
First work on GPU sparse LU factorization Algorithm description: elimination graph (EGraph) Algorithm analysis: parallelism in left-looking Algorithm implementation: timing order on GPU
Supplement to OpenCL BLAS Current cl_AMDBLAS has Triangular Solve but no LU Objective of LU:
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
3
Outline
Background Sparse LU factorization Dense LU factorization Summary
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
4
Background SPICE: the most popular circuit simulator
Simulating VSLI (~1 billion transistors) takes several days Bottleneck: Sparse LU factorization
Dynamic fluids, structural, economics …
Bottleneck
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
5
Outline
Background Sparse LU factorization Dense LU factorization Summary
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
6
Sparse LU factorization - related works [SuperLU 1999]
• Sequential, multi-thread, distributed versions• Incorporate Supernode, efficent for dense blocks
[Pardiso 2002]• Sequential, multi-thread, distributed, GPU [Christen2007]
versions• Adopt Supernode
But supernodes rarely form in circuit matrices [KLU 2010]
• Optimized for circuit matrices• Only sequential, use G/P left looking algorithm [G/P 1988]• Adopt BTF, without Supernode
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
Sparse LU factorization – left-looking Sequentially process each column When processing column k, use all the columns on the
left (1, 2, ..., k-1) to update column k. Update = vector multiply-and-add (MAD)
7
a a c b
c
a
b
c
a
b
c
a
b
read
write
•read+write>arithmeticUpdate
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
Algorithm description – EGraph Every column is updated with several columns on its left Nonzero structure of U determines the dependency
8
Vector MAD
(b)EGraph(a) Upper triangular
matrix U
nonzero
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
9
Algorithm analysis – two kinds of parallelism
Pipeline parallelism, alone with timing order
Column 1
Column 2
Column 3
Column 4
......
......
Overlapped factorization in pipeline mode
Thread 1
Thread 2
Divide columns into levels: columns in the same level are independent of each other Cluster mode: many columns factorized in parallel Pipeline mode: Overlap columns from different levels
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
10
Sparse LU factorization - workflow
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
11
Sparse LU factorization - preprocessing
Preprocessing: only once on CPU MC64 to ensure numerical stability [MC64]; Approximate Minimum Degree to reduce fill-ins
[AMD] ; pre-factorization (numeric factorization with partial
pivoting) to calculate the symbolic structure of L and U.
Sorting the nonzeros of L and U (introduced later)
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
12
Sparse LU factorization – on GPU GPU inputs
Location and values of nonzeros in A Location of nonzeros in L and U The Escheduler
GPU outputs Values of nonzeros in L and U
CSC (Compressed Sparse Column) format for sparse matrices A, L and U
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
13
Sparse LU factorization - avoid deadlock In traditional GPU programs, some wavefronts are inactive at
the beginning (limited resource etc.). They wait for other active wavefronts to finish and then become active.
But in sparse LU, we must ensure all wavefronts are active from the beginning
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
14
Sparse LU factorization - data formats
data formats for intermediate results:dense arrays vs. CSC
CSC (Compressed Sparse Column)• Can be put in local memory• Indexed accesses inconvenient (binary search)• Using too much local memory reduces active work-
groups, which leads to severe performance loss Dense arrays > CSC format: 2.5x
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
15
Sparse LU factorization - data locality Higher global memory bandwidth if consecutive
work-items access consecutive address
Improve data locality Nonzeros of L and U are out-of-order after preprocessing,
sort them according to row indices
1.7x speedup, overheads negligible Performed only once, incorporated into preprocessing
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
16
Experimental setups CPU
2 Xeon E5405 CPUs (8 cores in total) 2x6 MB L2 cache, 16GB ram
GPU AMD Radeon 5870 GPU
Testing matrices University of Florida Sparse Matrix Collection [Davis]
http://www.cise.ufl.edu/research/sparse/matrices/
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
17
Sparse LU factorization - Experimental results GPU speedups positively related to floating point
operations (flops)
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
18
Sparse LU factorization - Experimental results
Matrices divided into 4 groups First three groups according to Mflops
• GPU speedup positively related to Mflops 4th group: denormal floating point numbers
• Used to represent extremely small numbers• Very slowly on CPU, full speed support on GPU
An advantage of GPU in sparse LU and scientific computing
• Very high speedups for this group
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
19
Sparse LU factorization - Experimental results
Average speedup of each group
Group GPU bandwidth
GB / s
Over 1 CPU
Over 4 CPUs
Over 8 CPUs
Over KLU
1 0.81 0.41 0.24 0.22 0.58
2 10.97 2.43 0.85 0.55 3.64
3 52.59 10.53 3.65 2.58 15.58
4 36.82 26.86 8.01 4.48 25.61
All 15.91 4.51 1.64 1.13 6.25
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
20
Scalability – BBD Problem
How to use multiple GPUs?
Circuit-partition-based simulation algorithm bordered-block-diagonal (BBD) Diagonal blocks are factorized
independently
But An becomes dense. So we need dense LU factorization
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
21
Outline
Background Sparse LU factorization Dense LU factorization Summary
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
22
Dense LU Factorization – blocked algorithm Three core operations
Dense LU factorization Triangular matrix inversion Matrix multiplication
Suitable for GPU GEMM most frequent GEMM very efficient on GPU
• 920 Gflop/s (single), 290 Gflop/s (double)finished LU + inverse GEMM
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
23
443 Gflop/s (single), 163 Gflop/s (double)
Dense LU Factorization – performance
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
24
Comparison to previous studies
Dense LU Factorization – related works
Performance of Dense LU FactorizationWork Hardware Single Double
[Galoppo2005] GTX 7800 10 --
[Volkov2008] GTX 8800 179 --
[Tomov2010] 8 Xeon Harpertown 100 50
[Tomov2010] GTX 280 300 --
[Tomov2010] 8 Xeon Harpertown + GTX 280 388 99
Ours Radeon 5870 443 163
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
25
Dense LU Factorization – further improvement
CPU BLAS for Gaussian elimination 100 Gflop/s GEMM can be further improved
Scalability to multiple GPUs Blocked dense LU: independent GEMMs on multiple GPUs Diagonal blocks in BBD on multiple GPUs Linear performance improvement expected
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
26
Summary First work on GPU sparse LU factorization
Exploit parallelism of left-looking algorithm Blocked dense LU factorization
443 Gflop/s (single), 163 Gflop/s (double)
Supplement to OpenCL BLAS Accelerate SPICE simulators
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
27
Reference [SPICE] L. W. Nagel, “SPICE 2: A computer program to stimulate semiconductor
circuits,” Ph.D. dissertation, University of California, Berkeley, 1975. [SuperLU1999] J. W. Demmel, S. C. Eisenstat, J. R. Gilbert, X. S. Li, and J. W. H. Liu,
“A supernodal approach to sparse partial pivoting,” SIAM J. Matrix Analysis and Applications, vol. 20, no. 3, pp. 720–755, 1999
[Pardiso2002] O. Schenk and K. Gartner, “Solving unsymmetric sparse systems of linear equations with pardiso,” Computational Science - ICCS 2002, vol. 2330, pp. 355–363, 2002.
[G/P 1988] J. R. Gilbert and T. Peierls, “Sparse partial pivoting in time proportional to arithmetic operations,” SIAM J. Sci. Statist. Comput., vol. 9, pp. 862– 874, 1988
[KLU2010] T. A. Davis and E. Palamadai Natarajan, “Algorithm 907: KLU, a direct sparse solver for circuit simulation problems,” ACM Trans. Math. Softw., vol. 37, pp. 36:1–36:17, September 2010.
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
28
Reference [Christen2007] M. Christen, O. Schenk, and H. Burkhart, “General-purpose sparse
matrix building blocks using the nvidia cuda technology platform,” 2007. [Davis] T. A. Davis and Y. Hu, “The university of florida sparse matrix collection,” to
appear in ACM Transactions on Mathematical Software. [Galoppo2005] N. Galoppo, N. K. Govindaraju, M. Henson, and D. Manocha, “LU-
GPU: Efficient algorithms for solving dense linear systems on graphics hardware,” SC Conference, vol. 0, p. 3, 2005.
[Volkov2008] V. Volkov and J. Demmel, “LU, QR and Cholesky factorizations using vector capabilities of gpus,” EECS Department, University of California, Berkeley, Tech. Rep. UCB/EECS-2008-49, May 2008.
[Tomov2010] S. Tomov, J. Dongarra, and M. Baboulin, “Towards dense linear algebra for hybrid gpu accelerated manycore systems,” Parallel Comput., vol. 36, pp. 232–240, June 2010.
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
29
Reference [MC64] I. S. Duff and J. Koster, “The design and use of algorithms for permuting
large entries to the diagonal of sparse matrices,” SIAM J. Matrix Anal. and Applics, no. 4, pp. 889–901, 1997.
[AMD] P. R. Amestoy, Enseeiht-Irit, T. A. Davis, and I. S. Duff, “Algorithm 837: AMD, an approximate minimum degree ordering algorithm,” ACM Trans. Math. Softw., vol. 30, pp. 381–388, September 2004.
Department of Electronic Engineering, Tsinghua University
30Nano-scale Integrated Circuit and System Lab.
Thank you !
Nano-scale Integrated Circuit and System Lab.,EE Department, Tsinghua University
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
31
Sparse LU factorization – Terminology Elimination Graph Definition
An edge from j to k iff U(j, k) != 0 In the following context, node = column
• Level Definition– The length of the longest
path from any source node to itself.
– Source nodes have no incoming edges.
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
32
Sparse LU factorization - Experimental results
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
33
Dense LU factorization – Basic algorithm
Factorize to get and
Blocked LU factorization
Nano-scale Integrated Circuit and System Lab.
Department of Electronic Engineering, Tsinghua University
34
Dense LU factorization – Basic algorithm
Repeat the process to obtain , , and so on