Synthesis of fixed-point programs: the case of matrix multiplication
|
|
- Logan Snow
- 6 years ago
- Views:
Transcription
1 Synthesis of fixed-point programs: the case of matrix multiplication Mohamed Amine Najahi To cite this version: Mohamed Amine Najahi. Synthesis of fixed-point programs: the case of matrix multiplication. EJCIM: École Jeunes Chercheurs en Informatique Mathématique, Apr 213, Perpignan, France. 13th École Jeunes Chercheurs en Informatique Mathématique (EJCIM 213) Perpignan, April 12th, 213, 213. <lirmm > HAL Id: lirmm Submitted on 22 Feb 216 HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
2 13 th École Jeunes Chercheurs en Informatique Mathématique (EJCIM 213) Perpignan, April 12 th, 213 Synthesis of fixed-point programs: the case of matrix multiplication Amine Najahi Advisers: M. Martel and G. Revy Équipe-projet DALI, Univ. Perpignan Via Domitia LIRMM, CNRS: UMR Univ. Montpellier 2 DALI A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 1/26
3 How easy it is to program a product of matrices? A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 2/26
4 How easy it is to program a product of matrices? Well, in floating-point, it is very easy!! #define N 8 int main () { int i,j,k; float A[N][N]={...}; float B[N][N]={...}; float C[N][N]={,...,}; for (i = ; i < N ; i++) for (j = ; j < N ; j++) for (k = ; k < N ; k++) /* dot product of row i and column j */ C[i][j]+=A[i][k]*B[k][j]; return ; } A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 2/26
5 How easy it is to program a product of matrices? Well, in floating-point, it is very easy!! #define N 8 int main () { int i,j,k; float A[N][N]={...}; float B[N][N]={...}; float C[N][N]={,...,}; for (i = ; i < N ; i++) for (j = ; j < N ; j++) for (k = ; k < N ; k++) /* dot product of row i and column j */ C[i][j]+=A[i][k]*B[k][j]; return ; } But, what if the target does not have a floating-point unit? A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 2/26
6 Motivation Embedded systems are ubiquitous microprocessors and/or DSPs dedicated to one or a few specific tasks satisfy constraints: area, energy consumption, conception cost Some embedded systems do not have any FPU (floating-point unit) Embedded systems No FPU Highly used in audio and video applications demanding on floating-point computations A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 3/26
7 Motivation Embedded systems are ubiquitous microprocessors and/or DSPs dedicated to one or a few specific tasks satisfy constraints: area, energy consumption, conception cost Some embedded systems do not have any FPU (floating-point unit) Applications Embedded systems FP computations No FPU Highly used in audio and video applications demanding on floating-point computations A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 3/26
8 Motivation Embedded systems are ubiquitous microprocessors and/or DSPs dedicated to one or a few specific tasks satisfy constraints: area, energy consumption, conception cost Some embedded systems do not have any FPU (floating-point unit) Applications FP computations Embedded systems No FPU Software implementing floating point arithmetic Highly used in audio and video applications demanding on floating-point computations A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 3/26
9 Motivation Embedded systems are ubiquitous microprocessors and/or DSPs dedicated to one or a few specific tasks satisfy constraints: area, energy consumption, conception cost Some embedded systems do not have any FPU (floating-point unit) Applications FP computations Embedded systems Conversion No FPU Fixed point Float to Fix conversion is tackled by the ANR project DEFIS LIP6, IRISA, CEA, LIRMM, THALES and INPIXAL A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 3/26
10 Outline of the talk 1. Background of fixed-point arithmetic 1.1 Basics of fixed-point arithmetic 1.2 Numerical and combinatorial issues in fixed-point programs 1.3 CGPE 2. Matrix multiplication in fixed-point 2.1 An accurate algorithm 2.2 A compact algorithm 2.3 Closest pair algorithm 3. Conclusion A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 4/26
11 Background of fixed-point arithmetic Outline of the talk 1. Background of fixed-point arithmetic 1.1 Basics of fixed-point arithmetic 1.2 Numerical and combinatorial issues in fixed-point programs 1.3 CGPE 2. Matrix multiplication in fixed-point 2.1 An accurate algorithm 2.2 A compact algorithm 2.3 Closest pair algorithm 3. Conclusion A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 5/26
12 Background of fixed-point arithmetic Basics of fixed-point arithmetic Principles of fixed-point arithmetic Main idea of fixed-point arithmetic: interpret bit words as integers coupled with a scale factor: z 2 n 1 1 z = 13 Value in fixed-point = = = Integer part Fractional part A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 6/26
13 Background of fixed-point arithmetic Basics of fixed-point arithmetic Principles of fixed-point arithmetic Main idea of fixed-point arithmetic: interpret bit words as integers coupled with a scale factor: z 2 n 1 1 z = 13 Value in fixed-point = = = Integer part Fractional part The scale factor (or fixed-point format) is implicit, only the programmer is aware of it A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 6/26
14 Background of fixed-point arithmetic Basics of fixed-point arithmetic Principles of fixed-point arithmetic Main idea of fixed-point arithmetic: interpret bit words as integers coupled with a scale factor: z 2 n 1 1 z = 13 Value in fixed-point = = = Integer part Fractional part The scale factor (or fixed-point format) is implicit, only the programmer is aware of it Let us denote by Q a,b a fixed-point format with a integer bits and b fractional bits 1 1 ( ) 1 Q 1,7 1 1 Q 2,6 1 1 (2.3125) 1 Q,8 1 1 ( ) 1 A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 6/26
15 Background of fixed-point arithmetic Basics of fixed-point arithmetic Basic fixed-point operators Addition The two variables have to be in the same fixed-point format The sum of two Q a,b variables yields a Q a+1,b variable truncated A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 7/26
16 Background of fixed-point arithmetic Basics of fixed-point arithmetic Basic fixed-point operators Addition The two variables have to be in the same fixed-point format The sum of two Q a,b variables yields a Q a+1,b variable truncated Multiplication No need for the two variables to have the same fixed-point format The product of a Q a,b variable by a Q c,d variable yields a Q a+c,b+d variable truncated A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 7/26
17 Background of fixed-point arithmetic Numerical and combinatorial issues in fixed-point programs First example: a size 3 dot product Let us consider the arithmetic expression: (a b ) +(a 1 b 1 ) +(a 2 b 2 ) and the following input fixed-point formats: a b a 1 b 1 a 2 b 2 Value [.1,1.57] [,1.98] [.1,.87] [1.1,1.86] [,15.4] [2,3.3] Fixed-point format Q 1,7 Q 1,7 Q,8 Q 1,7 Q 4,4 Q 2,6 A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 8/26
18 Background of fixed-point arithmetic Numerical and combinatorial issues in fixed-point programs First example: a size 3 dot product Let us consider the arithmetic expression: (a b ) +(a 1 b 1 ) +(a 2 b 2 ) and the following input fixed-point formats: a b a 1 b 1 a 2 b 2 Value [.1,1.57] [,1.98] [.1,.87] [1.1,1.86] [,15.4] [2,3.3] Fixed-point format Q 1,7 Q 1,7 Q,8 Q 1,7 Q 4,4 Q 2,6 Let us focus on 2 different schemes to compute the sum of products: Q 8,15 + Q 7,15 + Q 2,14 Q 7,15 a + b Q 1,15 Q 6,1 1 1 a b a 2 b 2 in full precision Q 3,15 + Q 6,1 Q 2,14 Q 1,15 a 2 b 2 a b a 1 b 1 (c +(c 1 + c 2 )) ((c + c 1 ) + c 2 ) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 8/26
19 Background of fixed-point arithmetic Numerical and combinatorial issues in fixed-point programs First example: a size 3 dot product Let us consider the arithmetic expression: (a b ) +(a 1 b 1 ) +(a 2 b 2 ) and the following input fixed-point formats: a b a 1 b 1 a 2 b 2 Value [.1,1.57] [,1.98] [.1,.87] [1.1,1.86] [,15.4] [2,3.3] Fixed-point format Q 1,7 Q 1,7 Q,8 Q 1,7 Q 4,4 Q 2,6 Let us focus on 2 different schemes to compute the sum of products: Q 8,8 + Q 7,9 + Q 2,14 Q 7,9 a + b Q 1,15 Q 6,1 1 1 a b a 2 b 2 with 16 bits precision Q 3,13 + Q 6,1 Q 2,14 Q 1,15 a 2 b 2 a b a 1 b 1 (c +(c 1 + c 2 )) ((c + c 1 ) + c 2 ) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 8/26
20 Background of fixed-point arithmetic CGPE The CGPE 1 software tool Written by Revy and Mouilleron to aid in emulating floating-point in software A tool that generates fast and certified code fast that reduce the evaluation latency on a given target, by using the target architecture features (as much as possible) certified for which we can bound the error entailed by the evaluation within the given target s arithmetic back-end middle-end front-end DAG computation Set of DAGs Filter 1 Filter n Decorated DAGs Code generator polynomial.xml <polynomial> <coefficient... >... <variable... >... <error... > </polynomial> architecture.xml <architecture> <instruction name="add" type="unsigned" inputs="32 32" output="32" latency="1" nodes="add dag 1..." macro="static inline..." gappa="..."... /> </architecture> C files Accuracy certificates 1 Code Generation for Polynomial Evaluation A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 9/26
21 Matrix multiplication in fixed-point Outline of the talk 1. Background of fixed-point arithmetic 1.1 Basics of fixed-point arithmetic 1.2 Numerical and combinatorial issues in fixed-point programs 1.3 CGPE 2. Matrix multiplication in fixed-point 2.1 An accurate algorithm 2.2 A compact algorithm 2.3 Closest pair algorithm 3. Conclusion A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 1/26
22 Matrix multiplication in fixed-point Defining the problem We are provided with a black box (CGPE) that synthesises code for dot-products in fixed-point arithmetic C code 2 real interval vectors Accuracy certificate 2 matrices A and B in I(R n n ) [ 4.54,7.78] [.789,.967]... A = [12.51,24.14] [.921,.791] [ 64,45.78] [.287,.7]... and, B = [125.1,245.14] [ 5.74,7.32] A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 11/26
23 Matrix multiplication in fixed-point Defining the problem We are provided with a black box (CGPE) that synthesises code for dot-products in fixed-point arithmetic C code 2 real interval vectors Accuracy certificate 2 matrices A and B in I(R n n ) [ 4.54,7.78] [.789,.967]... A = [12.51,24.14] [.921,.791] [ 64,45.78] [.287,.7]... and, B = [125.1,245.14] [ 5.74,7.32] We are asked to Generate code that evaluates all the products C = MN in fixed-point arithmetic where M A and N B A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 11/26
24 Matrix multiplication in fixed-point Tradeoffs to consider Remark: The suggested strategy should be efficient in terms of the tradeoffs below for all matrices of size smaller than 8 8 A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 12/26
25 Matrix multiplication in fixed-point Tradeoffs to consider Remark: The suggested strategy should be efficient in terms of the tradeoffs below for all matrices of size smaller than Size of the generated code We are targeting embedded systems code size should be as tight as possible A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 12/26
26 Matrix multiplication in fixed-point Tradeoffs to consider Remark: The suggested strategy should be efficient in terms of the tradeoffs below for all matrices of size smaller than Size of the generated code We are targeting embedded systems code size should be as tight as possible 2. Accuracy of the generated code Accuracy certificates should be produced that bound the absolute error The guaranteed absolute error should be as tight as possible A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 12/26
27 Matrix multiplication in fixed-point Tradeoffs to consider Remark: The suggested strategy should be efficient in terms of the tradeoffs below for all matrices of size smaller than Size of the generated code We are targeting embedded systems code size should be as tight as possible 2. Accuracy of the generated code Accuracy certificates should be produced that bound the absolute error The guaranteed absolute error should be as tight as possible 3. Speed of generation A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 12/26
28 Matrix multiplication in fixed-point An accurate algorithm An accurate algorithm Main idea: Generate a dot product code for each coefficient of the resulting matrix AccurateProduct Inputs: Two square matrices A I(R n n ) and B I(R n n ) Outputs: C code to compute the product MN for all M A and N B Steps: 1: for 1 < i n do 2: for 1 < j n do 3: cgpegendotproduct(a i,b j ); 4: end for 5: end for A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 13/26
29 Matrix multiplication in fixed-point An accurate algorithm An accurate algorithm Main idea: Generate a dot product code for each coefficient of the resulting matrix AccurateProduct Inputs: Two square matrices A I(R n n ) and B I(R n n ) Outputs: C code to compute the product MN for all M A and N B Steps: 1: for 1 < i n do 2: for 1 < j n do 3: cgpegendotproduct(a i,b j ); 4: end for 5: end for Illustration on the product of two 2 2 matrices C = ( ) C1,1 = cgpegendotproduct(a 1,B 1 ) C 1,2 = cgpegendotproduct(a 1,B 2 ) C 2,1 = cgpegendotproduct(a 2,B 1 ) C 2,2 = cgpegendotproduct(a 2,B 2 ) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 13/26
30 Matrix multiplication in fixed-point An accurate algorithm Analysis of AccurateProduct For square matrices of size n, n 2 calls to the cgpegendotproduct are issued Each dot product uses more than 2n instructions (n multiplications + n additions) The generated code for the product is proportional in size to 2n 3 More than 124 instructions for 8 8 matrices A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 14/26
31 Matrix multiplication in fixed-point An accurate algorithm Analysis of AccurateProduct For square matrices of size n, n 2 calls to the cgpegendotproduct are issued Each dot product uses more than 2n instructions (n multiplications + n additions) The generated code for the product is proportional in size to 2n 3 More than 124 instructions for 8 8 matrices Advantages Easy to generate code Two nested loops and n 2 calls to the routine cgpegendotproduct The reference in terms of numerical quality Drawbacks Code size is proportional to 2n 3 Similar code sizes are prohibitive in embedded systems A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 14/26
32 Matrix multiplication in fixed-point A compact algorithm A compact algorithm Main idea: Generate a unique dot product code for all the computations CompactProduct Inputs: Two square matrices A I(R n n ) and B I(R n n ) Outputs: C code to compute the product MN for all M A and N B Steps: 1: compute v such that v = A 1 A 2 A n 2: compute w such that w = B 1 B 2 B n 3: cgpegendotproduct(v,w); A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 15/26
33 Matrix multiplication in fixed-point A compact algorithm A compact algorithm Main idea: Generate a unique dot product code for all the computations CompactProduct Inputs: Two square matrices A I(R n n ) and B I(R n n ) Outputs: C code to compute the product MN for all M A and N B Steps: 1: compute v such that v = A 1 A 2 A n 2: compute w such that w = B 1 B 2 B n 3: cgpegendotproduct(v,w); Illustration on the product of two 2 2 matrices C = ( ) C1,1 = cgpegendotproduct(a 1 A 2,B 1 B 2 ) C 1,2 = cgpegendotproduct(a 1 A 2,B 1 B 2 ) C 2,1 = cgpegendotproduct(a 1 A 2,B 1 B 2 ) C 2,2 = cgpegendotproduct(a 1 A 2,B 1 B 2 ) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 15/26
34 Matrix multiplication in fixed-point A compact algorithm Analysis of CompactProduct For square matrices of size n, only one call to the cgpegendotproduct is issued The dot product uses around 2n instructions (n multiplications + n additions) The generated code for the product is proportional in size to 2n Around 16 instructions for 8 8 matrices A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 16/26
35 Matrix multiplication in fixed-point A compact algorithm Analysis of CompactProduct For square matrices of size n, only one call to the cgpegendotproduct is issued The dot product uses around 2n instructions (n multiplications + n additions) The generated code for the product is proportional in size to 2n Around 16 instructions for 8 8 matrices Advantages Easy to generate code Compute the union of all vectors of A and B and call the routine cgpegendotproduct The reference in terms of code size Drawbacks Numerical quality deteriorates dramatically A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 16/26
36 Matrix multiplication in fixed-point Closest pair algorithm A closest pair algorithm Main idea: Fuse together only rows or columns that are close to each other The Hausdorff distance d H dh : I(R n ) I(R { n ) R d H (A,B) = max max ai, } b ai i b i 1 i n A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 17/26
37 Matrix multiplication in fixed-point Closest pair algorithm A closest pair algorithm Main idea: Fuse together only rows or columns that are close to each other The Hausdorff distance d H dh : I(R n ) I(R { n ) R d H (A,B) = max max ai, } b ai i b i 1 i n Example Let A = ( [ 4, 7] ) [ 11, 12] ( ) and B = [ 2, 88] [ 23, 1] be two vectors in I ( R 2), we have: d H (A,B) = 11 (A,B) = ( ) [ 4, 88] [ 23, 12] a 1 a1 b 1 b1 R (A 1,B 1 ) d H (A 1,B 1 ) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 17/26
38 Matrix multiplication in fixed-point Closest pair algorithm ClosestPairFusion ClosestPairFusion Inputs: n vectors, v 1,...,v n in I(R m ) a routine findclosestpair based on d H a routine Union that applies the union operator the number k of output vectors Outputs: k vectors in I(R m ) Steps: 1: B = {v 1,...,v n } 2: while size(b) > k do 3: (u 1,u 2 ) = findclosestpair(b) 4: remove(u 1,B) 5: remove(u 2,B) 6: add(union(u 1,u 2 ),B) 7: end while A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 18/26
39 Matrix multiplication in fixed-point Closest pair algorithm Illustration of the ClosestPairFusion v 1 [ 4,4] [ 5,5] [ 5,5] [ 6,6] v 2 [ 2,2] [ 1,1] [ 3,3] [ 9,9] v 3 [ 7,7] [ 4,4] [ 12,12] [ 11,11] v 4 [ 8,8] [ 1,1] [ 1,1] [ 9,9] w 1 w 2 w 3 w 4 [ 3,3] [ 14,14] [ 5,5] [ 6,6] [ 1,1] [ 11,11] [ 3,3] [ 9,9] [ 4,4] [ 8,8] [ 11,11] [ 1,1] [ 9,9] [ 7,7] [ 1,1] [ 2,2] d H (v 1,v 2) d H (v 1,v 3) d H (v 1,v 4) d H (v 2,v 3) d H (v 2,v 4) d H (v 3,v 4) d H (w 1,w 2) d H (w 1,w 3) d H (w 1,w 4) d H (w 2,w 3) d H (w 2,w 4) d H (w 3,w 4) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 19/26
40 Matrix multiplication in fixed-point Closest pair algorithm Illustration of the ClosestPairFusion v 1 [ 4,4] [ 5,5] [ 5,5] [ 6,6] v 2 [ 2,2] [ 1,1] [ 3,3] [ 9,9] v 3 [ 7,7] [ 4,4] [ 12,12] [ 11,11] v 4 [ 8,8] [ 1,1] [ 1,1] [ 9,9] w 1 w 2 w 3 w 4 [ 3,3] [ 14,14] [ 5,5] [ 6,6] [ 1,1] [ 11,11] [ 3,3] [ 9,9] [ 4,4] [ 8,8] [ 11,11] [ 1,1] [ 9,9] [ 7,7] [ 1,1] [ 2,2] d H (v 1,v 2) d H (v 1,v 3) d H (v 1,v 4) d H (v 2,v 3) d H (v 2,v 4) d H (v 3,v 4) d H (w 1,w 2) d H (w 1,w 3) d H (w 1,w 4) d H (w 2,w 3) d H (w 2,w 4) d H (w 3,w 4) v 1 [ 4,4] [ 5,5] [ 5,5] [ 6,6] v 2 [ 2,2] [ 1,1] [ 3,3] [ 9,9] v 3 v 4 [ 8,8] [ 4,4] [ 12,12] [ 11,11] w 1 w 3 w 2 w 4 [ 5,5] [ 14,14] [ 6,6] [ 3,3] [ 11,11] [ 9,9] [ 11,11] [ 8,8] [ 1,1] [ 1,1] [ 7,7] [ 2,2] d H (v 1,v 2) d H (v 1,v 3 v 4) d H (v 2,v 3 v 4) d H (w 1 w 3,w 2) d H (w 1 w 3,w 4) d H (w 2,w 4) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 19/26
41 Matrix multiplication in fixed-point Closest pair algorithm Illustration of the ClosestPairFusion v 1 [ 4,4] [ 5,5] [ 5,5] [ 6,6] v 2 [ 2,2] [ 1,1] [ 3,3] [ 9,9] v 3 [ 7,7] [ 4,4] [ 12,12] [ 11,11] v 4 [ 8,8] [ 1,1] [ 1,1] [ 9,9] w 1 w 2 w 3 w 4 [ 3,3] [ 14,14] [ 5,5] [ 6,6] [ 1,1] [ 11,11] [ 3,3] [ 9,9] [ 4,4] [ 8,8] [ 11,11] [ 1,1] [ 9,9] [ 7,7] [ 1,1] [ 2,2] d H (v 1,v 2) d H (v 1,v 3) d H (v 1,v 4) d H (v 2,v 3) d H (v 2,v 4) d H (v 3,v 4) d H (w 1,w 2) d H (w 1,w 3) d H (w 1,w 4) d H (w 2,w 3) d H (w 2,w 4) d H (w 3,w 4) v 1 [ 4,4] [ 5,5] [ 5,5] [ 6,6] v 2 [ 2,2] [ 1,1] [ 3,3] [ 9,9] v 3 v 4 [ 8,8] [ 4,4] [ 12,12] [ 11,11] w 1 w 3 w 2 w 4 [ 5,5] [ 14,14] [ 6,6] [ 3,3] [ 11,11] [ 9,9] [ 11,11] [ 8,8] [ 1,1] [ 1,1] [ 7,7] [ 2,2] d H (v 1,v 2) d H (v 1,v 3 v 4) d H (v 2,v 3 v 4) d H (w 1 w 3,w 2) d H (w 1 w 3,w 4) d H (w 2,w 4) v 1 v 2 ( [ 4,4] [ 5,5] [ 5,5] ) [ 9,9] v 3 v 4 [ 8,8] [ 4,4] [ 12,12] [ 11,11] w 1 w 3 w 2 w 4 [ 5,5] [ 14,14] [ 3,3] [ 11,11] [ 11,11] [ 8,8] [ 1,1] [ 7,7] A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 19/26
42 Matrix multiplication in fixed-point Closest pair algorithm Analysis of the closest pair algorithm For square matrices of size n, k l calls to the cgpegendotproduct are issued Each dot product uses more than 2n instructions (n multiplications + n additions) The generated code for the product is proportional in size to 2nkl For 8 8 matrices, the table below gives the number of instructions l k Advantages Code size can be controlled through the parameters k and l Drawbacks Numerical quality deteriorates with small values of k and l A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 2/26
43 Matrix multiplication in fixed-point Closest pair algorithm Analysis of the closest pair algorithm For square matrices of size n, k l calls to the cgpegendotproduct are issued Each dot product uses more than 2n instructions (n multiplications + n additions) The generated code for the product is proportional in size to 2nkl For 8 8 matrices, the table below gives the number of instructions l k Advantages Code size can be controlled through the parameters k and l Drawbacks Numerical quality deteriorates with small values of k and l A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 2/26
44 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms These results were produced for interval matrices of size 8 8 The center of each interval is randomly selected in [ 1, 1] The diameter of the intervals is fixed to AccurateProduct Largest certified error:.1254 Mean certified error:.865 Number of instructions: 124 CompactProduct Largest certified error:.5585 Mean certified error:.5585 Number of instructions: 16 A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 21/26
45 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
46 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
47 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
48 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
49 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
50 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
51 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
52 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
53 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
54 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 22/26
55 Matrix multiplication in fixed-point Closest pair algorithm Let us compare these algorithms, cont d Certified bound for the absolute error Maximum certified error Mean certified error Code size Approximate number of instructions k (Number of fused rows and columns) A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 23/26
56 Conclusion Outline of the talk 1. Background of fixed-point arithmetic 1.1 Basics of fixed-point arithmetic 1.2 Numerical and combinatorial issues in fixed-point programs 1.3 CGPE 2. Matrix multiplication in fixed-point 2.1 An accurate algorithm 2.2 A compact algorithm 2.3 Closest pair algorithm 3. Conclusion A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 24/26
57 Conclusion In this talk: We suggested 3 strategies to generate code for matrix product in fixed-point arithmetic The accurate algorithm performs well in terms of numerical quality but is prohibitive The compact algorithm generates concise codes but deteriorates the numerical quality The Closest Pair algorithm enables the tradeoffs between code size and numerical quality A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 25/26
58 Conclusion In this talk: We suggested 3 strategies to generate code for matrix product in fixed-point arithmetic The accurate algorithm performs well in terms of numerical quality but is prohibitive The compact algorithm generates concise codes but deteriorates the numerical quality The Closest Pair algorithm enables the tradeoffs between code size and numerical quality For the future, we will be working on: Suggesting similar algorithms for the discrete convolution in fixed-point arithmetic Investigating the synthesis of VHDL code for building blocks like matrix multiplication A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 25/26
59 Conclusion 13 th École Jeunes Chercheurs en Informatique Mathématique (EJCIM 213) Perpignan, April 12 th, 213 Synthesis of fixed-point programs: the case of matrix multiplication Amine Najahi Advisers: M. Martel and G. Revy Équipe-projet DALI, Univ. Perpignan Via Domitia LIRMM, CNRS: UMR Univ. Montpellier 2 DALI A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 26/26
60 Conclusion Example of code generated by CGPE // File automatically generated by CGPE (Code Generation for Polynomial Evaluation ) // Scheme : (((( a+(s*(a1+(t*a2))))+((s*(t*t))*(a3+(t*a4))))+((s*(t*t))*((t*t)*(a5+(t*a6)))))+(((s*(t*t))*((t*t)*( T*T)))*(( a7+(t*a8))+((t*t)*(a9+(t*a1)))))) // Degree : [9,1] uint32_t func_(uint32_t T, uint32_t S) { uint32_t r = mul(t, x5a82685d ); // 1.31 uint32_t r1 = xb54f31f - r; // 1.31 uint32_t r2 = mul(s, r1); // 2.3 uint32_t r3 = x2 + r2; // 2.3 uint32_t r4 = mul(t, T); //.32 uint32_t r5 = mul(s, r4); // 1.31 uint32_t r6 = mul(t, x386fd5f4 ); // 1.31 uint32_t r7 = x43df72f7 - r6; // 1.31 uint32_t r8 = mul(r5, r7); // 2.3 uint32_t r9 = r3 + r8; // 2.3 uint32_t r1 = mul(t, x ); // 1.31 uint32_t r11 = x38b r1; // 1.31 uint32_t r12 = mul(r4, r11); // 1.31 uint32_t r13 = mul(r5, r12); // 2.3 uint32_t r14 = r9 + r13; // 2.3 uint32_t r15 = mul(r4, r4); //.32 uint32_t r16 = mul(r5, r15); // 1.31 uint32_t r17 = mul(t, x16c5cd9 ); // 1.31 uint32_t r18 = x1d7bf968 - r17; // 1.31 uint32_t r19 = mul(t, xfa9aa4 ); // 1.31 uint32_t r2 = x5dfffa4 - r19; // 1.31 uint32_t r21 = mul(r4, r2); // 1.31 uint32_t r22 = r18 + r21; // 1.31 uint32_t r23 = mul(r16, r22); // 2.3 uint32_t r24 = r14 + r23; // 2.3 return r24; } /* Error bound computed using MPFI: * [ b-283, * b -282] * ~ [ -2^{ },2^{ }] */ A. Najahi (DALI UPVD/LIRMM, CNRS, UM2) Synthesis of fixed-point programs: the case of matrix multiplication 27/26
BoxPlot++ Zeina Azmeh, Fady Hamoui, Marianne Huchard. To cite this version: HAL Id: lirmm
BoxPlot++ Zeina Azmeh, Fady Hamoui, Marianne Huchard To cite this version: Zeina Azmeh, Fady Hamoui, Marianne Huchard. BoxPlot++. RR-11001, 2011. HAL Id: lirmm-00557222 https://hal-lirmm.ccsd.cnrs.fr/lirmm-00557222
More informationRelabeling nodes according to the structure of the graph
Relabeling nodes according to the structure of the graph Ronan Hamon, Céline Robardet, Pierre Borgnat, Patrick Flandrin To cite this version: Ronan Hamon, Céline Robardet, Pierre Borgnat, Patrick Flandrin.
More informationDANCer: Dynamic Attributed Network with Community Structure Generator
DANCer: Dynamic Attributed Network with Community Structure Generator Oualid Benyahia, Christine Largeron, Baptiste Jeudy, Osmar Zaïane To cite this version: Oualid Benyahia, Christine Largeron, Baptiste
More informationThe optimal routing of augmented cubes.
The optimal routing of augmented cubes. Meirun Chen, Reza Naserasr To cite this version: Meirun Chen, Reza Naserasr. The optimal routing of augmented cubes.. Information Processing Letters, Elsevier, 28.
More informationEfficient implementation of interval matrix multiplication
Efficient implementation of interval matrix multiplication Hong Diep Nguyen To cite this version: Hong Diep Nguyen. Efficient implementation of interval matrix multiplication. Para 2010: State of the Art
More informationSelf-optimisation using runtime code generation for Wireless Sensor Networks Internet-of-Things
Self-optimisation using runtime code generation for Wireless Sensor Networks Internet-of-Things Caroline Quéva, Damien Couroussé, Henri-Pierre Charles To cite this version: Caroline Quéva, Damien Couroussé,
More informationAn Experimental Assessment of the 2D Visibility Complex
An Experimental Assessment of the D Visibility Complex Hazel Everett, Sylvain Lazard, Sylvain Petitjean, Linqiao Zhang To cite this version: Hazel Everett, Sylvain Lazard, Sylvain Petitjean, Linqiao Zhang.
More informationThe Proportional Colouring Problem: Optimizing Buffers in Radio Mesh Networks
The Proportional Colouring Problem: Optimizing Buffers in Radio Mesh Networks Florian Huc, Claudia Linhares Sales, Hervé Rivano To cite this version: Florian Huc, Claudia Linhares Sales, Hervé Rivano.
More informationFault-Tolerant Storage Servers for the Databases of Redundant Web Servers in a Computing Grid
Fault-Tolerant s for the Databases of Redundant Web Servers in a Computing Grid Minhwan Ok To cite this version: Minhwan Ok. Fault-Tolerant s for the Databases of Redundant Web Servers in a Computing Grid.
More informationScan chain encryption in Test Standards
Scan chain encryption in Test Standards Mathieu Da Silva, Giorgio Di Natale, Marie-Lise Flottes, Bruno Rouzeyre To cite this version: Mathieu Da Silva, Giorgio Di Natale, Marie-Lise Flottes, Bruno Rouzeyre.
More informationStream Ciphers: A Practical Solution for Efficient Homomorphic-Ciphertext Compression
Stream Ciphers: A Practical Solution for Efficient Homomorphic-Ciphertext Compression Anne Canteaut, Sergiu Carpov, Caroline Fontaine, Tancrède Lepoint, María Naya-Plasencia, Pascal Paillier, Renaud Sirdey
More informationLinux: Understanding Process-Level Power Consumption
Linux: Understanding Process-Level Power Consumption Aurélien Bourdon, Adel Noureddine, Romain Rouvoy, Lionel Seinturier To cite this version: Aurélien Bourdon, Adel Noureddine, Romain Rouvoy, Lionel Seinturier.
More informationLaHC at CLEF 2015 SBS Lab
LaHC at CLEF 2015 SBS Lab Nawal Ould-Amer, Mathias Géry To cite this version: Nawal Ould-Amer, Mathias Géry. LaHC at CLEF 2015 SBS Lab. Conference and Labs of the Evaluation Forum, Sep 2015, Toulouse,
More informationHardware support for UNUM floating point arithmetic
Hardware support for UNUM floating point arithmetic Andrea Bocco, Yves Durand, Florent De Dinechin To cite this version: Andrea Bocco, Yves Durand, Florent De Dinechin. Hardware support for UNUM floating
More informationMoveability and Collision Analysis for Fully-Parallel Manipulators
Moveability and Collision Analysis for Fully-Parallel Manipulators Damien Chablat, Philippe Wenger To cite this version: Damien Chablat, Philippe Wenger. Moveability and Collision Analysis for Fully-Parallel
More informationRegularization parameter estimation for non-negative hyperspectral image deconvolution:supplementary material
Regularization parameter estimation for non-negative hyperspectral image deconvolution:supplementary material Yingying Song, David Brie, El-Hadi Djermoune, Simon Henrot To cite this version: Yingying Song,
More informationCatalogue of architectural patterns characterized by constraint components, Version 1.0
Catalogue of architectural patterns characterized by constraint components, Version 1.0 Tu Minh Ton That, Chouki Tibermacine, Salah Sadou To cite this version: Tu Minh Ton That, Chouki Tibermacine, Salah
More informationStudy on Feebly Open Set with Respect to an Ideal Topological Spaces
Study on Feebly Open Set with Respect to an Ideal Topological Spaces Yiezi K. Al Talkany, Suadud H. Al Ismael To cite this version: Yiezi K. Al Talkany, Suadud H. Al Ismael. Study on Feebly Open Set with
More informationBranch-and-price algorithms for the Bi-Objective Vehicle Routing Problem with Time Windows
Branch-and-price algorithms for the Bi-Objective Vehicle Routing Problem with Time Windows Estèle Glize, Nicolas Jozefowiez, Sandra Ulrich Ngueveu To cite this version: Estèle Glize, Nicolas Jozefowiez,
More informationTeaching Encapsulation and Modularity in Object-Oriented Languages with Access Graphs
Teaching Encapsulation and Modularity in Object-Oriented Languages with Access Graphs Gilles Ardourel, Marianne Huchard To cite this version: Gilles Ardourel, Marianne Huchard. Teaching Encapsulation and
More informationEfficient Gradient Method for Locally Optimizing the Periodic/Aperiodic Ambiguity Function
Efficient Gradient Method for Locally Optimizing the Periodic/Aperiodic Ambiguity Function F Arlery, R assab, U Tan, F Lehmann To cite this version: F Arlery, R assab, U Tan, F Lehmann. Efficient Gradient
More informationTaking Benefit from the User Density in Large Cities for Delivering SMS
Taking Benefit from the User Density in Large Cities for Delivering SMS Yannick Léo, Anthony Busson, Carlos Sarraute, Eric Fleury To cite this version: Yannick Léo, Anthony Busson, Carlos Sarraute, Eric
More informationReal-Time Collision Detection for Dynamic Virtual Environments
Real-Time Collision Detection for Dynamic Virtual Environments Gabriel Zachmann, Matthias Teschner, Stefan Kimmerle, Bruno Heidelberger, Laks Raghupathi, Arnulph Fuhrmann To cite this version: Gabriel
More informationMokka, main guidelines and future
Mokka, main guidelines and future P. Mora De Freitas To cite this version: P. Mora De Freitas. Mokka, main guidelines and future. H. Videau; J-C. Brient. International Conference on Linear Collider, Apr
More informationMultimedia CTI Services for Telecommunication Systems
Multimedia CTI Services for Telecommunication Systems Xavier Scharff, Pascal Lorenz, Zoubir Mammeri To cite this version: Xavier Scharff, Pascal Lorenz, Zoubir Mammeri. Multimedia CTI Services for Telecommunication
More informationComputing and maximizing the exact reliability of wireless backhaul networks
Computing and maximizing the exact reliability of wireless backhaul networks David Coudert, James Luedtke, Eduardo Moreno, Konstantinos Priftis To cite this version: David Coudert, James Luedtke, Eduardo
More informationSDLS: a Matlab package for solving conic least-squares problems
SDLS: a Matlab package for solving conic least-squares problems Didier Henrion, Jérôme Malick To cite this version: Didier Henrion, Jérôme Malick. SDLS: a Matlab package for solving conic least-squares
More informationSetup of epiphytic assistance systems with SEPIA
Setup of epiphytic assistance systems with SEPIA Blandine Ginon, Stéphanie Jean-Daubias, Pierre-Antoine Champin, Marie Lefevre To cite this version: Blandine Ginon, Stéphanie Jean-Daubias, Pierre-Antoine
More informationTacked Link List - An Improved Linked List for Advance Resource Reservation
Tacked Link List - An Improved Linked List for Advance Resource Reservation Li-Bing Wu, Jing Fan, Lei Nie, Bing-Yi Liu To cite this version: Li-Bing Wu, Jing Fan, Lei Nie, Bing-Yi Liu. Tacked Link List
More informationInduced minors and well-quasi-ordering
Induced minors and well-quasi-ordering Jaroslaw Blasiok, Marcin Kamiński, Jean-Florent Raymond, Théophile Trunck To cite this version: Jaroslaw Blasiok, Marcin Kamiński, Jean-Florent Raymond, Théophile
More informationLinked data from your pocket: The Android RDFContentProvider
Linked data from your pocket: The Android RDFContentProvider Jérôme David, Jérôme Euzenat To cite this version: Jérôme David, Jérôme Euzenat. Linked data from your pocket: The Android RDFContentProvider.
More informationHow to simulate a volume-controlled flooding with mathematical morphology operators?
How to simulate a volume-controlled flooding with mathematical morphology operators? Serge Beucher To cite this version: Serge Beucher. How to simulate a volume-controlled flooding with mathematical morphology
More informationTHE COVERING OF ANCHORED RECTANGLES UP TO FIVE POINTS
THE COVERING OF ANCHORED RECTANGLES UP TO FIVE POINTS Antoine Mhanna To cite this version: Antoine Mhanna. THE COVERING OF ANCHORED RECTANGLES UP TO FIVE POINTS. 016. HAL Id: hal-0158188
More informationXBenchMatch: a Benchmark for XML Schema Matching Tools
XBenchMatch: a Benchmark for XML Schema Matching Tools Fabien Duchateau, Zohra Bellahsene, Ela Hunt To cite this version: Fabien Duchateau, Zohra Bellahsene, Ela Hunt. XBenchMatch: a Benchmark for XML
More informationAssisted Policy Management for SPARQL Endpoints Access Control
Assisted Policy Management for SPARQL Endpoints Access Control Luca Costabello, Serena Villata, Iacopo Vagliano, Fabien Gandon To cite this version: Luca Costabello, Serena Villata, Iacopo Vagliano, Fabien
More informationEvery 3-connected, essentially 11-connected line graph is hamiltonian
Every 3-connected, essentially 11-connected line graph is hamiltonian Hong-Jian Lai, Yehong Shao, Ju Zhou, Hehui Wu To cite this version: Hong-Jian Lai, Yehong Shao, Ju Zhou, Hehui Wu. Every 3-connected,
More informationRepresentation of Finite Games as Network Congestion Games
Representation of Finite Games as Network Congestion Games Igal Milchtaich To cite this version: Igal Milchtaich. Representation of Finite Games as Network Congestion Games. Roberto Cominetti and Sylvain
More informationComparison of spatial indexes
Comparison of spatial indexes Nathalie Andrea Barbosa Roa To cite this version: Nathalie Andrea Barbosa Roa. Comparison of spatial indexes. [Research Report] Rapport LAAS n 16631,., 13p. HAL
More informationRobust IP and UDP-lite header recovery for packetized multimedia transmission
Robust IP and UDP-lite header recovery for packetized multimedia transmission Michel Kieffer, François Mériaux To cite this version: Michel Kieffer, François Mériaux. Robust IP and UDP-lite header recovery
More informationX-Kaapi C programming interface
X-Kaapi C programming interface Fabien Le Mentec, Vincent Danjean, Thierry Gautier To cite this version: Fabien Le Mentec, Vincent Danjean, Thierry Gautier. X-Kaapi C programming interface. [Technical
More informationLight field video dataset captured by a R8 Raytrix camera (with disparity maps)
Light field video dataset captured by a R8 Raytrix camera (with disparity maps) Laurent Guillo, Xiaoran Jiang, Gauthier Lafruit, Christine Guillemot To cite this version: Laurent Guillo, Xiaoran Jiang,
More informationBlind Browsing on Hand-Held Devices: Touching the Web... to Understand it Better
Blind Browsing on Hand-Held Devices: Touching the Web... to Understand it Better Waseem Safi Fabrice Maurel Jean-Marc Routoure Pierre Beust Gaël Dias To cite this version: Waseem Safi Fabrice Maurel Jean-Marc
More informationComparison of radiosity and ray-tracing methods for coupled rooms
Comparison of radiosity and ray-tracing methods for coupled rooms Jimmy Dondaine, Alain Le Bot, Joel Rech, Sébastien Mussa Peretto To cite this version: Jimmy Dondaine, Alain Le Bot, Joel Rech, Sébastien
More informationAn Efficient Numerical Inverse Scattering Algorithm for Generalized Zakharov-Shabat Equations with Two Potential Functions
An Efficient Numerical Inverse Scattering Algorithm for Generalized Zakharov-Shabat Equations with Two Potential Functions Huaibin Tang, Qinghua Zhang To cite this version: Huaibin Tang, Qinghua Zhang.
More informationThe Connectivity Order of Links
The Connectivity Order of Links Stéphane Dugowson To cite this version: Stéphane Dugowson. The Connectivity Order of Links. 4 pages, 2 figures. 2008. HAL Id: hal-00275717 https://hal.archives-ouvertes.fr/hal-00275717
More informationVery Tight Coupling between LTE and WiFi: a Practical Analysis
Very Tight Coupling between LTE and WiFi: a Practical Analysis Younes Khadraoui, Xavier Lagrange, Annie Gravey To cite this version: Younes Khadraoui, Xavier Lagrange, Annie Gravey. Very Tight Coupling
More informationComparator: A Tool for Quantifying Behavioural Compatibility
Comparator: A Tool for Quantifying Behavioural Compatibility Meriem Ouederni, Gwen Salaün, Javier Cámara, Ernesto Pimentel To cite this version: Meriem Ouederni, Gwen Salaün, Javier Cámara, Ernesto Pimentel.
More informationYAM++ : A multi-strategy based approach for Ontology matching task
YAM++ : A multi-strategy based approach for Ontology matching task Duy Hoa Ngo, Zohra Bellahsene To cite this version: Duy Hoa Ngo, Zohra Bellahsene. YAM++ : A multi-strategy based approach for Ontology
More informationA Generic Architecture of CCSDS Low Density Parity Check Decoder for Near-Earth Applications
A Generic Architecture of CCSDS Low Density Parity Check Decoder for Near-Earth Applications Fabien Demangel, Nicolas Fau, Nicolas Drabik, François Charot, Christophe Wolinski To cite this version: Fabien
More informationService Reconfiguration in the DANAH Assistive System
Service Reconfiguration in the DANAH Assistive System Said Lankri, Pascal Berruet, Jean-Luc Philippe To cite this version: Said Lankri, Pascal Berruet, Jean-Luc Philippe. Service Reconfiguration in the
More informationA Voronoi-Based Hybrid Meshing Method
A Voronoi-Based Hybrid Meshing Method Jeanne Pellerin, Lévy Bruno, Guillaume Caumon To cite this version: Jeanne Pellerin, Lévy Bruno, Guillaume Caumon. A Voronoi-Based Hybrid Meshing Method. 2012. hal-00770939
More informationMalware models for network and service management
Malware models for network and service management Jérôme François, Radu State, Olivier Festor To cite this version: Jérôme François, Radu State, Olivier Festor. Malware models for network and service management.
More informationDeformetrica: a software for statistical analysis of anatomical shapes
Deformetrica: a software for statistical analysis of anatomical shapes Alexandre Routier, Marcel Prastawa, Benjamin Charlier, Cédric Doucet, Joan Alexis Glaunès, Stanley Durrleman To cite this version:
More information[Demo] A webtool for analyzing land-use planning documents
[Demo] A webtool for analyzing land-use planning documents M.A. Farvardin, Eric Kergosien, Mathieu Roche, Maguelonne Teisseire To cite this version: M.A. Farvardin, Eric Kergosien, Mathieu Roche, Maguelonne
More informationQuickRanking: Fast Algorithm For Sorting And Ranking Data
QuickRanking: Fast Algorithm For Sorting And Ranking Data Laurent Ott To cite this version: Laurent Ott. QuickRanking: Fast Algorithm For Sorting And Ranking Data. Fichiers produits par l auteur. 2015.
More informationTraffic Grooming in Bidirectional WDM Ring Networks
Traffic Grooming in Bidirectional WDM Ring Networks Jean-Claude Bermond, David Coudert, Xavier Munoz, Ignasi Sau To cite this version: Jean-Claude Bermond, David Coudert, Xavier Munoz, Ignasi Sau. Traffic
More informationAcyclic Coloring of Graphs of Maximum Degree
Acyclic Coloring of Graphs of Maximum Degree Guillaume Fertin, André Raspaud To cite this version: Guillaume Fertin, André Raspaud. Acyclic Coloring of Graphs of Maximum Degree. Stefan Felsner. 005 European
More informationImplementing decimal floating-point arithmetic through binary: some suggestions
Implementing decimal floating-point arithmetic through binary: some suggestions Nicolas Brisebarre, Milos Ercegovac, Nicolas Louvet, Erik Martin-Dorel, Jean-Michel Muller, Adrien Panhaleux To cite this
More informationIntroClassJava: A Benchmark of 297 Small and Buggy Java Programs
IntroClassJava: A Benchmark of 297 Small and Buggy Java Programs Thomas Durieux, Martin Monperrus To cite this version: Thomas Durieux, Martin Monperrus. IntroClassJava: A Benchmark of 297 Small and Buggy
More informationAn FCA Framework for Knowledge Discovery in SPARQL Query Answers
An FCA Framework for Knowledge Discovery in SPARQL Query Answers Melisachew Wudage Chekol, Amedeo Napoli To cite this version: Melisachew Wudage Chekol, Amedeo Napoli. An FCA Framework for Knowledge Discovery
More informationASAP.V2 and ASAP.V3: Sequential optimization of an Algorithm Selector and a Scheduler
ASAP.V2 and ASAP.V3: Sequential optimization of an Algorithm Selector and a Scheduler François Gonard, Marc Schoenauer, Michele Sebag To cite this version: François Gonard, Marc Schoenauer, Michele Sebag.
More informationSimulations of VANET Scenarios with OPNET and SUMO
Simulations of VANET Scenarios with OPNET and SUMO Florent Kaisser, Christophe Gransart, Marion Berbineau To cite this version: Florent Kaisser, Christophe Gransart, Marion Berbineau. Simulations of VANET
More informationImplementing an Automatic Functional Test Pattern Generation for Mixed-Signal Boards in a Maintenance Context
Implementing an Automatic Functional Test Pattern Generation for Mixed-Signal Boards in a Maintenance Context Bertrand Gilles, Laurent Tchamnda Nana, Valérie-Anne Nicolas To cite this version: Bertrand
More informationThe New Territory of Lightweight Security in a Cloud Computing Environment
The New Territory of Lightweight Security in a Cloud Computing Environment Shu-Ching Wang, Shih-Chi Tseng, Hsin-Met Chuan, Kuo-Qin Yan, Szu-Hao Tsai To cite this version: Shu-Ching Wang, Shih-Chi Tseng,
More informationFrom medical imaging to numerical simulations
From medical imaging to numerical simulations Christophe Prud Homme, Vincent Chabannes, Marcela Szopos, Alexandre Ancel, Julien Jomier To cite this version: Christophe Prud Homme, Vincent Chabannes, Marcela
More informationlambda-min Decoding Algorithm of Regular and Irregular LDPC Codes
lambda-min Decoding Algorithm of Regular and Irregular LDPC Codes Emmanuel Boutillon, Frédéric Guillou, Jean-Luc Danger To cite this version: Emmanuel Boutillon, Frédéric Guillou, Jean-Luc Danger lambda-min
More informationReverse-engineering of UML 2.0 Sequence Diagrams from Execution Traces
Reverse-engineering of UML 2.0 Sequence Diagrams from Execution Traces Romain Delamare, Benoit Baudry, Yves Le Traon To cite this version: Romain Delamare, Benoit Baudry, Yves Le Traon. Reverse-engineering
More informationMulti-mode Operator for SHA-2 Hash Functions
Multi-mode Operator for SHA-2 Hash Functions Ryan Glabb, Laurent Imbert, Graham Jullien, Arnaud Tisserand, Nicolas Veyrat-Charvillon To cite this version: Ryan Glabb, Laurent Imbert, Graham Jullien, Arnaud
More informationWeed Leaf Recognition in Complex Natural Scenes by Model-Guided Edge Pairing
Weed Leaf Recognition in Complex Natural Scenes by Model-Guided Edge Pairing Benoit De Mezzo, Gilles Rabatel, Christophe Fiorio To cite this version: Benoit De Mezzo, Gilles Rabatel, Christophe Fiorio.
More informationAccurate Conversion of Earth-Fixed Earth-Centered Coordinates to Geodetic Coordinates
Accurate Conversion of Earth-Fixed Earth-Centered Coordinates to Geodetic Coordinates Karl Osen To cite this version: Karl Osen. Accurate Conversion of Earth-Fixed Earth-Centered Coordinates to Geodetic
More informationWorkspace and joint space analysis of the 3-RPS parallel robot
Workspace and joint space analysis of the 3-RPS parallel robot Damien Chablat, Ranjan Jha, Fabrice Rouillier, Guillaume Moroz To cite this version: Damien Chablat, Ranjan Jha, Fabrice Rouillier, Guillaume
More informationReal-Time and Resilient Intrusion Detection: A Flow-Based Approach
Real-Time and Resilient Intrusion Detection: A Flow-Based Approach Rick Hofstede, Aiko Pras To cite this version: Rick Hofstede, Aiko Pras. Real-Time and Resilient Intrusion Detection: A Flow-Based Approach.
More informationModularity for Java and How OSGi Can Help
Modularity for Java and How OSGi Can Help Richard Hall To cite this version: Richard Hall. Modularity for Java and How OSGi Can Help. présentation invitée à DECOR04. 2004. HAL Id: hal-00003299
More informationFramework for Hierarchical and Distributed Smart Grid Management
Framework for Hierarchical and Distributed Smart Grid Management Rémi Bonnefoi, Christophe Moy, Jacques Palicot To cite this version: Rémi Bonnefoi, Christophe Moy, Jacques Palicot. Framework for Hierarchical
More informationKernel perfect and critical kernel imperfect digraphs structure
Kernel perfect and critical kernel imperfect digraphs structure Hortensia Galeana-Sánchez, Mucuy-Kak Guevara To cite this version: Hortensia Galeana-Sánchez, Mucuy-Kak Guevara. Kernel perfect and critical
More informationHySCaS: Hybrid Stereoscopic Calibration Software
HySCaS: Hybrid Stereoscopic Calibration Software Guillaume Caron, Damien Eynard To cite this version: Guillaume Caron, Damien Eynard. HySCaS: Hybrid Stereoscopic Calibration Software. SPIE newsroom in
More informationYANG-Based Configuration Modeling - The SecSIP IPS Case Study
YANG-Based Configuration Modeling - The SecSIP IPS Case Study Abdelkader Lahmadi, Emmanuel Nataf, Olivier Festor To cite this version: Abdelkader Lahmadi, Emmanuel Nataf, Olivier Festor. YANG-Based Configuration
More informationBugMaps-Granger: A Tool for Causality Analysis between Source Code Metrics and Bugs
BugMaps-Granger: A Tool for Causality Analysis between Source Code Metrics and Bugs Cesar Couto, Pedro Pires, Marco Tulio Valente, Roberto Bigonha, Andre Hora, Nicolas Anquetil To cite this version: Cesar
More informationFIT IoT-LAB: The Largest IoT Open Experimental Testbed
FIT IoT-LAB: The Largest IoT Open Experimental Testbed Eric Fleury, Nathalie Mitton, Thomas Noel, Cédric Adjih To cite this version: Eric Fleury, Nathalie Mitton, Thomas Noel, Cédric Adjih. FIT IoT-LAB:
More informationQuasi-tilings. Dominique Rossin, Daniel Krob, Sebastien Desreux
Quasi-tilings Dominique Rossin, Daniel Krob, Sebastien Desreux To cite this version: Dominique Rossin, Daniel Krob, Sebastien Desreux. Quasi-tilings. FPSAC/SFCA 03, 2003, Linkoping, Sweden. 2003.
More informationRecordMe: A Smartphone Application for Experimental Collections of Large Amount of Data Respecting Volunteer s Privacy
RecordMe: A Smartphone Application for Experimental Collections of Large Amount of Data Respecting Volunteer s Privacy David Blachon, François Portet, Laurent Besacier, Stéphan Tassart To cite this version:
More informationA N-dimensional Stochastic Control Algorithm for Electricity Asset Management on PC cluster and Blue Gene Supercomputer
A N-dimensional Stochastic Control Algorithm for Electricity Asset Management on PC cluster and Blue Gene Supercomputer Stéphane Vialle, Xavier Warin, Patrick Mercier To cite this version: Stéphane Vialle,
More informationIMPLEMENTATION OF MOTION ESTIMATION BASED ON HETEROGENEOUS PARALLEL COMPUTING SYSTEM WITH OPENC
IMPLEMENTATION OF MOTION ESTIMATION BASED ON HETEROGENEOUS PARALLEL COMPUTING SYSTEM WITH OPENC Jinglin Zhang, Jean François Nezan, Jean-Gabriel Cousin To cite this version: Jinglin Zhang, Jean François
More informationConstraint Programming and Combinatorial Optimisation in Numberjack
Constraint Programming and Combinatorial Optimisation in Numberjack Emmanuel Hébrard, O Mahony Eoin, O Sullivan Barry To cite this version: Emmanuel Hébrard, O Mahony Eoin, O Sullivan Barry. Constraint
More informationZigbee Wireless Sensor Network Nodes Deployment Strategy for Digital Agricultural Data Acquisition
Zigbee Wireless Sensor Network Nodes Deployment Strategy for Digital Agricultural Data Acquisition Xinjian Xiang, Xiaoqing Guo To cite this version: Xinjian Xiang, Xiaoqing Guo. Zigbee Wireless Sensor
More informationOptimizing correctly-rounded reciprocal square roots for embedded VLIW cores
Optimizing correctly-rounded reciprocal square roots for embedded VLIW cores Claude-Pierre Jeannerod, Guillaume Revy To cite this version: Claude-Pierre Jeannerod, Guillaume Revy. Optimizing correctly-rounded
More informationApproximate Computing with Runtime Code Generation on Resource-Constrained Embedded Devices
Approximate Computing with Runtime Code Generation on Resource-Constrained Embedded Devices Damien Couroussé, Caroline Quéva, Henri-Pierre Charles To cite this version: Damien Couroussé, Caroline Quéva,
More informationScalewelis: a Scalable Query-based Faceted Search System on Top of SPARQL Endpoints
Scalewelis: a Scalable Query-based Faceted Search System on Top of SPARQL Endpoints Joris Guyonvarc H, Sébastien Ferré To cite this version: Joris Guyonvarc H, Sébastien Ferré. Scalewelis: a Scalable Query-based
More informationPrimitive roots of bi-periodic infinite pictures
Primitive roots of bi-periodic infinite pictures Nicolas Bacquey To cite this version: Nicolas Bacquey. Primitive roots of bi-periodic infinite pictures. Words 5, Sep 5, Kiel, Germany. Words 5, Local Proceedings.
More informationA 64-Kbytes ITTAGE indirect branch predictor
A 64-Kbytes ITTAGE indirect branch André Seznec To cite this version: André Seznec. A 64-Kbytes ITTAGE indirect branch. JWAC-2: Championship Branch Prediction, Jun 2011, San Jose, United States. 2011,.
More informationFuzzy sensor for the perception of colour
Fuzzy sensor for the perception of colour Eric Benoit, Laurent Foulloy, Sylvie Galichet, Gilles Mauris To cite this version: Eric Benoit, Laurent Foulloy, Sylvie Galichet, Gilles Mauris. Fuzzy sensor for
More informationA Methodology for Improving Software Design Lifecycle in Embedded Control Systems
A Methodology for Improving Software Design Lifecycle in Embedded Control Systems Mohamed El Mongi Ben Gaïd, Rémy Kocik, Yves Sorel, Rédha Hamouche To cite this version: Mohamed El Mongi Ben Gaïd, Rémy
More informationQuality of Service Enhancement by Using an Integer Bloom Filter Based Data Deduplication Mechanism in the Cloud Storage Environment
Quality of Service Enhancement by Using an Integer Bloom Filter Based Data Deduplication Mechanism in the Cloud Storage Environment Kuo-Qin Yan, Yung-Hsiang Su, Hsin-Met Chuan, Shu-Ching Wang, Bo-Wei Chen
More informationTime-Division Multiplexing Architecture for Hybrid Filter Bank A/D converters
Time-Division Multiplexing Architecture for Hybrid Filter Bank A/D converters Davud Asemani, Jacques Oksman To cite this version: Davud Asemani, Jacques Oksman Time-Division Multiplexing Architecture for
More informationLossless and Lossy Minimal Redundancy Pyramidal Decomposition for Scalable Image Compression Technique
Lossless and Lossy Minimal Redundancy Pyramidal Decomposition for Scalable Image Compression Technique Marie Babel, Olivier Déforges To cite this version: Marie Babel, Olivier Déforges. Lossless and Lossy
More informationMultidimensional Data Streams Summarization Using Extended Tilted-Time Windows
Multidimensional Data Streams Summarization Using Extended Tilted-Time Windows Yoann Pitarch, Pascal Poncelet, Anne Laurent, Marc Plantevit To cite this version: Yoann Pitarch, Pascal Poncelet, Anne Laurent,
More informationCombined video and laser camera for inspection of old mine shafts
Combined video and laser camera for inspection of old mine shafts Laurent Cauvin To cite this version: Laurent Cauvin. Combined video and laser camera for inspection of old mine shafts. 3. International
More informationOpen Digital Forms. Hiep Le, Thomas Rebele, Fabian Suchanek. HAL Id: hal
Open Digital Forms Hiep Le, Thomas Rebele, Fabian Suchanek To cite this version: Hiep Le, Thomas Rebele, Fabian Suchanek. Open Digital Forms. Research and Advanced Technology for Digital Libraries - 20th
More informationModelling approaches for anion-exclusion in compacted Na-bentonite
Modelling approaches for anion-exclusion in compacted Na-bentonite Christophe Tournassat, C. A. J. Appelo To cite this version: Christophe Tournassat, C. A. J. Appelo. Modelling approaches for anion-exclusion
More informationAn Object Oriented Finite Element Toolkit
An Object Oriented Finite Element Toolkit Rachid Touzani To cite this version: Rachid Touzani. An Object Oriented Finite Element Toolkit. Mang, H.A.; Rammerstorfer, F.G.; Eberhardsteiner, J. Fifth World
More information