NPRG042 Programming in Parallel Environment [2022/23]

Results of CUDA Simulation

Name		v16k-e64k	v32k-e256k	v64k-e1024k	Total speedup
serial		865.47 ms (1.00x)	3468.85 ms (1.00x)	13946.7 ms (1.00x)	1.00x
solution		5.85241 ms (147.88x)	12.7826 ms (271.37x)	38.6158 ms (361.17x)	227.00x
adamekj9		5.90493 ms (146.57x)	12.5196 ms (277.07x)	43.7809 ms (318.56x)	221.06x
barosma		7.37584 ms (117.34x)	16.1263 ms (215.11x)	43.5332 ms (320.37x)	184.13x
beliansp		2.98744 ms (289.70x)	11.0578 ms (313.70x)	42.6876 ms (326.72x)	309.27x
dinhdo		5.45578 ms (158.63x)	11.6509 ms (297.73x)	36.3453 ms (383.73x)	244.53x
dobreva	Build log (dobreva) Compiling CUDA kernels kernels.obj ... Compiling and linking executable potential ... In file included from potential.cpp:8: ./implementation.hpp:114:9: error: ‘virtual void ProgramPotential<F, IDX_T, LEN_T>::iteration(std::vector<Point<F> >&)’ conflicts with a previous declaration 114 \| } \| ^ In file included from potential.cpp:8: ./implementation.hpp:41:13: note: previous declaration ‘ProgramPotential<F, IDX_T, LEN_T>::index_t ProgramPotential<F, IDX_T, LEN_T>::iteration’ 41 \| index_t iteration; \| ^~~~~~~~~ In file included from potential.cpp:8: ./implementation.hpp: In member function ‘virtual void ProgramPotential<F, IDX_T, LEN_T>::set_cuda_device()’: ./implementation.hpp:76:22: error: there are no arguments to ‘cudaInit’ that depend on a template parameter, so a declaration of ‘cudaInit’ must be available [-fpermissive] 76 \| cuda_error = cudaInit(); \| ^~~~~~~~ ./implementation.hpp:76:22: note: (if you use ‘-fpermissive’, G++ will accept your code, but allowing the use of an undeclared name is deprecated) ./implementation.hpp: In member function ‘virtual void ProgramPotential<F, IDX_T, LEN_T>::getVelocities(std::vector<Point<F> >&)’: ./implementation.hpp:127:39: error: expected primary-expression before ‘,’ token 127 \| cudaMemcpy(velocities.data(), , edges.size() * sizeof(edge_t), cudaMemcpyHostToDevice); \| ^ ./implementation.hpp:127:41: error: ‘edges’ was not declared in this scope 127 \| cudaMemcpy(velocities.data(), , edges.size() * sizeof(edge_t), cudaMemcpyHostToDevice); \| ^~~~~ potential.cpp: In function ‘int main(int, char*)’: potential.cpp:92:77: error: invalid new-expression of abstract class type ‘ProgramPotential<double, unsigned int, unsigned int>’ 92 \| IProgramPotential<real_t, index_t, length_t> program = new ProgramPotential<real_t, index_t, length_t>; \| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In file included from potential.cpp:8: ./implementation.hpp:16:7: note: because the following virtual functions are pure within ‘ProgramPotential<double, unsigned int, unsigned int>’: 16 \| class ProgramPotential : public IProgramPotential<F, IDX_T, LEN_T> \| ^~~~~~~~~~~~~~~~ In file included from ./implementation.hpp:6, from potential.cpp:8: ./internal/interface.hpp:98:22: note: ‘void IProgramPotential<F, IDX_T, LEN_T>::iteration(std::vector<Point<F> >&) [with F = double; IDX_T = unsigned int; LEN_T = unsigned int]’ 98 \| virtual void iteration(std::vector<point_t> &points) = 0; \| ^~~~~~~~~ make: *** [Makefile:22: potential] Error 1 Compilation of solution 'dobreva' failed.	MISSING	MISSING	MISSING	-
fedakri		11.5921 ms (74.66x)	34.8395 ms (99.57x)	130.048 ms (107.24x)	91.57x
goliasm		2.36476 ms (365.99x)	9.59834 ms (361.40x)	36.2542 ms (384.69x)	370.42x
hopfere		16.9191 ms (51.15x)	58.3078 ms (59.49x)	217.561 ms (64.10x)	57.74x
hotmarv_late		6.74322 ms (128.35x)	13.3564 ms (259.71x)	27.1634 ms (513.44x)	220.76x
hubatama		8.47679 ms (102.10x)	32.3848 ms (107.11x)	120.599 ms (115.65x)	108.00x
hubenami		WRONG	WRONG	WRONG	-
hurtisov	Build log (hurtisov) Compiling CUDA kernels kernels.obj ... kernels.h(65): error: incomplete type is not allowed kernels.h(65): error: identifier "point_t" is undefined kernels.h(65): error: identifier "points" is undefined kernels.h(65): error: identifier "velocities" is undefined kernels.h(65): error: identifier "real_t" is undefined kernels.h(65): error: expected a ")" kernels.h(66): error: incomplete type is not allowed kernels.h(66): error: identifier "edge_t" is undefined kernels.h(66): error: identifier "edges" is undefined kernels.h(66): error: identifier "length_t" is undefined kernels.h(66): error: identifier "lengths" is undefined kernels.h(66): error: expected a ")" kernels.h(67): error: identifier "index_t" is undefined kernels.h(67): error: "point_t" is not a type name kernels.h(67): error: "point_t" is not a type name kernels.h(67): error: "real_t" is not a type name kernels.h(67): error: "real_t" is not a type name kernels.cu(41): error: expected a ")" kernels.cu(41): error: declaration is incompatible with "<error-type> run_repulsive_force_kernel" kernels.h(65): here kernels.cu(43): error: identifier "pointsCount" is undefined kernels.cu(44): error: identifier "vertexRepulsion" is undefined kernels.cu(44): error: identifier "timeQuantum" is undefined kernels.cu(44): error: identifier "vertexMass" is undefined kernels.cu(80): error: declaration is incompatible with "<error-type> run_compulsive_forces_kernel" kernels.h(66): here kernels.cu(83): error: identifier "pointsCount" is undefined 25 errors detected in the compilation of "kernels.cu". make: *** [Makefile:27: kernels.obj] Error 2 Compilation of solution 'hurtisov' failed.	MISSING	MISSING	MISSING	-
jungmak_late		8.75838 ms (98.82x)	20.5573 ms (168.74x)	65.0866 ms (214.28x)	144.84x
kapustb		5.25004 ms (164.85x)	12.1182 ms (286.25x)	38.5778 ms (361.52x)	243.40x
kastovso		7.98451 ms (108.39x)	31.5442 ms (109.97x)	117.384 ms (118.81x)	112.21x
klemend		8.50086 ms (101.81x)	30.9353 ms (112.13x)	114.035 ms (122.30x)	111.45x
kormosa		9.58775 ms (90.27x)	32.4252 ms (106.98x)	130.84 ms (106.59x)	100.65x
kripnema		3.3925 ms (255.11x)	8.76659 ms (395.69x)	28.4229 ms (490.69x)	353.56x
krnakva					-
krnakva_late		7.30408 ms (118.49x)	16.8216 ms (206.21x)	52.2564 ms (266.89x)	176.10x
lamplotm		8.2167 ms (105.33x)	31.544 ms (109.97x)	117.677 ms (118.52x)	111.01x
lukasej1		7.46731 ms (115.90x)	31.6708 ms (109.53x)	117.311 ms (118.89x)	114.64x
malyjir1		9.4727 ms (91.36x)	34.1913 ms (101.45x)	135.408 ms (103.00x)	98.33x
markomi1		15.949 ms (54.26x)	62.5306 ms (55.47x)	255.999 ms (54.48x)	54.73x
markvaja_late		6.09953 ms (141.89x)	12.6961 ms (273.22x)	39.543 ms (352.70x)	221.52x
nagymar_late		11.0177 ms (78.55x)	22.0166 ms (157.56x)	83.9838 ms (166.06x)	119.53x
nekardaj		5.47502 ms (158.08x)	11.7375 ms (295.54x)	36.7679 ms (379.32x)	242.99x
obornyt		9.30944 ms (92.97x)	31.5553 ms (109.93x)	120.05 ms (116.17x)	105.41x
pacalm		5.01115 ms (172.71x)	11.4888 ms (301.93x)	36.7248 ms (379.76x)	255.64x
petrask1		20.535 ms (42.15x)	62.7053 ms (55.32x)	209.228 ms (66.66x)	52.81x
piroutej		15.3213 ms (56.49x)	48.2585 ms (71.88x)	171.184 ms (81.47x)	68.35x
polakluk		5.40509 ms (160.12x)	11.9182 ms (291.05x)	38.8798 ms (358.71x)	240.60x
provaja		7.94259 ms (108.97x)	31.1739 ms (111.27x)	119.85 ms (116.37x)	112.12x
stochelm					-
stochelm_late		5.96438 ms (145.11x)	12.4915 ms (277.70x)	38.8816 ms (358.70x)	225.90x
strycekr		7.60488 ms (113.80x)	31.3471 ms (110.66x)	123.533 ms (112.90x)	112.44x
svarcjar_late		5.51609 ms (156.90x)	11.8283 ms (293.27x)	37.4693 ms (372.22x)	240.58x
svestka1		32.8951 ms (26.31x)	79.9452 ms (43.39x)	237.104 ms (58.82x)	38.43x
tichavso					-
tichavso_late		5.74675 ms (150.60x)	21.2082 ms (163.56x)	82.0752 ms (169.93x)	160.95x
tomasekjac_late		2.42724 ms (356.57x)	9.15505 ms (378.90x)	34.4143 ms (405.26x)	379.20x
trefilpa		8.41299 ms (102.87x)	32.1884 ms (107.77x)	120.109 ms (116.12x)	108.65x
truchanm		9.24159 ms (93.65x)	35.0264 ms (99.04x)	120.859 ms (115.40x)	101.90x
vaico		15.694 ms (55.15x)	35.3586 ms (98.10x)	133.863 ms (104.19x)	79.10x
valenmic		9.85352 ms (87.83x)	32.7952 ms (105.77x)	123.298 ms (113.11x)	101.08x
vermesa		19.9257 ms (43.43x)	62.7564 ms (55.27x)	207.48 ms (67.22x)	53.58x
veselmi7		6.06584 ms (142.68x)	13.0544 ms (265.72x)	40.2156 ms (346.80x)	219.69x
vilimev_late		5.93078 ms (145.93x)	12.3824 ms (280.14x)	39.1548 ms (356.19x)	226.76x
vopalkoz		5.62238 ms (153.93x)	12.3103 ms (281.78x)	42.342 ms (329.38x)	229.34x
yanushop_late		127.26 ms (6.80x)	540.433 ms (6.42x)	2223.2 ms (6.27x)	6.49x