Search Results

Search found 6082 results on 244 pages for 'john son'.

Page 8/244 | < Previous Page | 4 5 6 7 8 9 10 11 12 13 14 15  | Next Page >

  • How John Got 15x Improvement Without Really Trying

    - by rchrd
    The following article was published on a Sun Microsystems website a number of years ago by John Feo. It is still useful and worth preserving. So I'm republishing it here.  How I Got 15x Improvement Without Really Trying John Feo, Sun Microsystems Taking ten "personal" program codes used in scientific and engineering research, the author was able to get from 2 to 15 times performance improvement easily by applying some simple general optimization techniques. Introduction Scientific research based on computer simulation depends on the simulation for advancement. The research can advance only as fast as the computational codes can execute. The codes' efficiency determines both the rate and quality of results. In the same amount of time, a faster program can generate more results and can carry out a more detailed simulation of physical phenomena than a slower program. Highly optimized programs help science advance quickly and insure that monies supporting scientific research are used as effectively as possible. Scientific computer codes divide into three broad categories: ISV, community, and personal. ISV codes are large, mature production codes developed and sold commercially. The codes improve slowly over time both in methods and capabilities, and they are well tuned for most vendor platforms. Since the codes are mature and complex, there are few opportunities to improve their performance solely through code optimization. Improvements of 10% to 15% are typical. Examples of ISV codes are DYNA3D, Gaussian, and Nastran. Community codes are non-commercial production codes used by a particular research field. Generally, they are developed and distributed by a single academic or research institution with assistance from the community. Most users just run the codes, but some develop new methods and extensions that feed back into the general release. The codes are available on most vendor platforms. Since these codes are younger than ISV codes, there are more opportunities to optimize the source code. Improvements of 50% are not unusual. Examples of community codes are AMBER, CHARM, BLAST, and FASTA. Personal codes are those written by single users or small research groups for their own use. These codes are not distributed, but may be passed from professor-to-student or student-to-student over several years. They form the primordial ocean of applications from which community and ISV codes emerge. Government research grants pay for the development of most personal codes. This paper reports on the nature and performance of this class of codes. Over the last year, I have looked at over two dozen personal codes from more than a dozen research institutions. The codes cover a variety of scientific fields, including astronomy, atmospheric sciences, bioinformatics, biology, chemistry, geology, and physics. The sources range from a few hundred lines to more than ten thousand lines, and are written in Fortran, Fortran 90, C, and C++. For the most part, the codes are modular, documented, and written in a clear, straightforward manner. They do not use complex language features, advanced data structures, programming tricks, or libraries. I had little trouble understanding what the codes did or how data structures were used. Most came with a makefile. Surprisingly, only one of the applications is parallel. All developers have access to parallel machines, so availability is not an issue. Several tried to parallelize their applications, but stopped after encountering difficulties. Lack of education and a perception that parallelism is difficult prevented most from trying. I parallelized several of the codes using OpenMP, and did not judge any of the codes as difficult to parallelize. Even more surprising than the lack of parallelism is the inefficiency of the codes. I was able to get large improvements in performance in a matter of a few days applying simple optimization techniques. Table 1 lists ten representative codes [names and affiliation are omitted to preserve anonymity]. Improvements on one processor range from 2x to 15.5x with a simple average of 4.75x. I did not use sophisticated performance tools or drill deep into the program's execution character as one would do when tuning ISV or community codes. Using only a profiler and source line timers, I identified inefficient sections of code and improved their performance by inspection. The changes were at a high level. I am sure there is another factor of 2 or 3 in each code, and more if the codes are parallelized. The study’s results show that personal scientific codes are running many times slower than they should and that the problem is pervasive. Computational scientists are not sloppy programmers; however, few are trained in the art of computer programming or code optimization. I found that most have a working knowledge of some programming language and standard software engineering practices; but they do not know, or think about, how to make their programs run faster. They simply do not know the standard techniques used to make codes run faster. In fact, they do not even perceive that such techniques exist. The case studies described in this paper show that applying simple, well known techniques can significantly increase the performance of personal codes. It is important that the scientific community and the Government agencies that support scientific research find ways to better educate academic scientific programmers. The inefficiency of their codes is so bad that it is retarding both the quality and progress of scientific research. # cacheperformance redundantoperations loopstructures performanceimprovement 1 x x 15.5 2 x 2.8 3 x x 2.5 4 x 2.1 5 x x 2.0 6 x 5.0 7 x 5.8 8 x 6.3 9 2.2 10 x x 3.3 Table 1 — Area of improvement and performance gains of 10 codes The remainder of the paper is organized as follows: sections 2, 3, and 4 discuss the three most common sources of inefficiencies in the codes studied. These are cache performance, redundant operations, and loop structures. Each section includes several examples. The last section summaries the work and suggests a possible solution to the issues raised. Optimizing cache performance Commodity microprocessor systems use caches to increase memory bandwidth and reduce memory latencies. Typical latencies from processor to L1, L2, local, and remote memory are 3, 10, 50, and 200 cycles, respectively. Moreover, bandwidth falls off dramatically as memory distances increase. Programs that do not use cache effectively run many times slower than programs that do. When optimizing for cache, the biggest performance gains are achieved by accessing data in cache order and reusing data to amortize the overhead of cache misses. Secondary considerations are prefetching, associativity, and replacement; however, the understanding and analysis required to optimize for the latter are probably beyond the capabilities of the non-expert. Much can be gained simply by accessing data in the correct order and maximizing data reuse. 6 out of the 10 codes studied here benefited from such high level optimizations. Array Accesses The most important cache optimization is the most basic: accessing Fortran array elements in column order and C array elements in row order. Four of the ten codes—1, 2, 4, and 10—got it wrong. Compilers will restructure nested loops to optimize cache performance, but may not do so if the loop structure is too complex, or the loop body includes conditionals, complex addressing, or function calls. In code 1, the compiler failed to invert a key loop because of complex addressing do I = 0, 1010, delta_x IM = I - delta_x IP = I + delta_x do J = 5, 995, delta_x JM = J - delta_x JP = J + delta_x T1 = CA1(IP, J) + CA1(I, JP) T2 = CA1(IM, J) + CA1(I, JM) S1 = T1 + T2 - 4 * CA1(I, J) CA(I, J) = CA1(I, J) + D * S1 end do end do In code 2, the culprit is conditionals do I = 1, N do J = 1, N If (IFLAG(I,J) .EQ. 0) then T1 = Value(I, J-1) T2 = Value(I-1, J) T3 = Value(I, J) T4 = Value(I+1, J) T5 = Value(I, J+1) Value(I,J) = 0.25 * (T1 + T2 + T5 + T4) Delta = ABS(T3 - Value(I,J)) If (Delta .GT. MaxDelta) MaxDelta = Delta endif enddo enddo I fixed both programs by inverting the loops by hand. Code 10 has three-dimensional arrays and triply nested loops. The structure of the most computationally intensive loops is too complex to invert automatically or by hand. The only practical solution is to transpose the arrays so that the dimension accessed by the innermost loop is in cache order. The arrays can be transposed at construction or prior to entering a computationally intensive section of code. The former requires all array references to be modified, while the latter is cost effective only if the cost of the transpose is amortized over many accesses. I used the second approach to optimize code 10. Code 5 has four-dimensional arrays and loops are nested four deep. For all of the reasons cited above the compiler is not able to restructure three key loops. Assume C arrays and let the four dimensions of the arrays be i, j, k, and l. In the original code, the index structure of the three loops is L1: for i L2: for i L3: for i for l for l for j for k for j for k for j for k for l So only L3 accesses array elements in cache order. L1 is a very complex loop—much too complex to invert. I brought the loop into cache alignment by transposing the second and fourth dimensions of the arrays. Since the code uses a macro to compute all array indexes, I effected the transpose at construction and changed the macro appropriately. The dimensions of the new arrays are now: i, l, k, and j. L3 is a simple loop and easily inverted. L2 has a loop-carried scalar dependence in k. By promoting the scalar name that carries the dependence to an array, I was able to invert the third and fourth subloops aligning the loop with cache. Code 5 is by far the most difficult of the four codes to optimize for array accesses; but the knowledge required to fix the problems is no more than that required for the other codes. I would judge this code at the limits of, but not beyond, the capabilities of appropriately trained computational scientists. Array Strides When a cache miss occurs, a line (64 bytes) rather than just one word is loaded into the cache. If data is accessed stride 1, than the cost of the miss is amortized over 8 words. Any stride other than one reduces the cost savings. Two of the ten codes studied suffered from non-unit strides. The codes represent two important classes of "strided" codes. Code 1 employs a multi-grid algorithm to reduce time to convergence. The grids are every tenth, fifth, second, and unit element. Since time to convergence is inversely proportional to the distance between elements, coarse grids converge quickly providing good starting values for finer grids. The better starting values further reduce the time to convergence. The downside is that grids of every nth element, n > 1, introduce non-unit strides into the computation. In the original code, much of the savings of the multi-grid algorithm were lost due to this problem. I eliminated the problem by compressing (copying) coarse grids into continuous memory, and rewriting the computation as a function of the compressed grid. On convergence, I copied the final values of the compressed grid back to the original grid. The savings gained from unit stride access of the compressed grid more than paid for the cost of copying. Using compressed grids, the loop from code 1 included in the previous section becomes do j = 1, GZ do i = 1, GZ T1 = CA(i+0, j-1) + CA(i-1, j+0) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) S1 = T1 + T4 - 4 * CA1(i+0, j+0) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 enddo enddo where CA and CA1 are compressed arrays of size GZ. Code 7 traverses a list of objects selecting objects for later processing. The labels of the selected objects are stored in an array. The selection step has unit stride, but the processing steps have irregular stride. A fix is to save the parameters of the selected objects in temporary arrays as they are selected, and pass the temporary arrays to the processing functions. The fix is practical if the same parameters are used in selection as in processing, or if processing comprises a series of distinct steps which use overlapping subsets of the parameters. Both conditions are true for code 7, so I achieved significant improvement by copying parameters to temporary arrays during selection. Data reuse In the previous sections, we optimized for spatial locality. It is also important to optimize for temporal locality. Once read, a datum should be used as much as possible before it is forced from cache. Loop fusion and loop unrolling are two techniques that increase temporal locality. Unfortunately, both techniques increase register pressure—as loop bodies become larger, the number of registers required to hold temporary values grows. Once register spilling occurs, any gains evaporate quickly. For multiprocessors with small register sets or small caches, the sweet spot can be very small. In the ten codes presented here, I found no opportunities for loop fusion and only two opportunities for loop unrolling (codes 1 and 3). In code 1, unrolling the outer and inner loop one iteration increases the number of result values computed by the loop body from 1 to 4, do J = 1, GZ-2, 2 do I = 1, GZ-2, 2 T1 = CA1(i+0, j-1) + CA1(i-1, j+0) T2 = CA1(i+1, j-1) + CA1(i+0, j+0) T3 = CA1(i+0, j+0) + CA1(i-1, j+1) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) T5 = CA1(i+2, j+0) + CA1(i+1, j+1) T6 = CA1(i+1, j+1) + CA1(i+0, j+2) T7 = CA1(i+2, j+1) + CA1(i+1, j+2) S1 = T1 + T4 - 4 * CA1(i+0, j+0) S2 = T2 + T5 - 4 * CA1(i+1, j+0) S3 = T3 + T6 - 4 * CA1(i+0, j+1) S4 = T4 + T7 - 4 * CA1(i+1, j+1) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 CA(i+1, j+0) = CA1(i+1, j+0) + DD * S2 CA(i+0, j+1) = CA1(i+0, j+1) + DD * S3 CA(i+1, j+1) = CA1(i+1, j+1) + DD * S4 enddo enddo The loop body executes 12 reads, whereas as the rolled loop shown in the previous section executes 20 reads to compute the same four values. In code 3, two loops are unrolled 8 times and one loop is unrolled 4 times. Here is the before for (k = 0; k < NK[u]; k++) { sum = 0.0; for (y = 0; y < NY; y++) { sum += W[y][u][k] * delta[y]; } backprop[i++]=sum; } and after code for (k = 0; k < KK - 8; k+=8) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (y = 0; y < NY; y++) { sum0 += W[y][0][k+0] * delta[y]; sum1 += W[y][0][k+1] * delta[y]; sum2 += W[y][0][k+2] * delta[y]; sum3 += W[y][0][k+3] * delta[y]; sum4 += W[y][0][k+4] * delta[y]; sum5 += W[y][0][k+5] * delta[y]; sum6 += W[y][0][k+6] * delta[y]; sum7 += W[y][0][k+7] * delta[y]; } backprop[k+0] = sum0; backprop[k+1] = sum1; backprop[k+2] = sum2; backprop[k+3] = sum3; backprop[k+4] = sum4; backprop[k+5] = sum5; backprop[k+6] = sum6; backprop[k+7] = sum7; } for one of the loops unrolled 8 times. Optimizing for temporal locality is the most difficult optimization considered in this paper. The concepts are not difficult, but the sweet spot is small. Identifying where the program can benefit from loop unrolling or loop fusion is not trivial. Moreover, it takes some effort to get it right. Still, educating scientific programmers about temporal locality and teaching them how to optimize for it will pay dividends. Reducing instruction count Execution time is a function of instruction count. Reduce the count and you usually reduce the time. The best solution is to use a more efficient algorithm; that is, an algorithm whose order of complexity is smaller, that converges quicker, or is more accurate. Optimizing source code without changing the algorithm yields smaller, but still significant, gains. This paper considers only the latter because the intent is to study how much better codes can run if written by programmers schooled in basic code optimization techniques. The ten codes studied benefited from three types of "instruction reducing" optimizations. The two most prevalent were hoisting invariant memory and data operations out of inner loops. The third was eliminating unnecessary data copying. The nature of these inefficiencies is language dependent. Memory operations The semantics of C make it difficult for the compiler to determine all the invariant memory operations in a loop. The problem is particularly acute for loops in functions since the compiler may not know the values of the function's parameters at every call site when compiling the function. Most compilers support pragmas to help resolve ambiguities; however, these pragmas are not comprehensive and there is no standard syntax. To guarantee that invariant memory operations are not executed repetitively, the user has little choice but to hoist the operations by hand. The problem is not as severe in Fortran programs because in the absence of equivalence statements, it is a violation of the language's semantics for two names to share memory. Codes 3 and 5 are C programs. In both cases, the compiler did not hoist all invariant memory operations from inner loops. Consider the following loop from code 3 for (y = 0; y < NY; y++) { i = 0; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += delta[y] * I1[i++]; } } } Since dW[y][u] can point to the same memory space as delta for one or more values of y and u, assignment to dW[y][u][k] may change the value of delta[y]. In reality, dW and delta do not overlap in memory, so I rewrote the loop as for (y = 0; y < NY; y++) { i = 0; Dy = delta[y]; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += Dy * I1[i++]; } } } Failure to hoist invariant memory operations may be due to complex address calculations. If the compiler can not determine that the address calculation is invariant, then it can hoist neither the calculation nor the associated memory operations. As noted above, code 5 uses a macro to address four-dimensional arrays #define MAT4D(a,q,i,j,k) (double *)((a)->data + (q)*(a)->strides[0] + (i)*(a)->strides[3] + (j)*(a)->strides[2] + (k)*(a)->strides[1]) The macro is too complex for the compiler to understand and so, it does not identify any subexpressions as loop invariant. The simplest way to eliminate the address calculation from the innermost loop (over i) is to define a0 = MAT4D(a,q,0,j,k) before the loop and then replace all instances of *MAT4D(a,q,i,j,k) in the loop with a0[i] A similar problem appears in code 6, a Fortran program. The key loop in this program is do n1 = 1, nh nx1 = (n1 - 1) / nz + 1 nz1 = n1 - nz * (nx1 - 1) do n2 = 1, nh nx2 = (n2 - 1) / nz + 1 nz2 = n2 - nz * (nx2 - 1) ndx = nx2 - nx1 ndy = nz2 - nz1 gxx = grn(1,ndx,ndy) gyy = grn(2,ndx,ndy) gxy = grn(3,ndx,ndy) balance(n1,1) = balance(n1,1) + (force(n2,1) * gxx + force(n2,2) * gxy) * h1 balance(n1,2) = balance(n1,2) + (force(n2,1) * gxy + force(n2,2) * gyy)*h1 end do end do The programmer has written this loop well—there are no loop invariant operations with respect to n1 and n2. However, the loop resides within an iterative loop over time and the index calculations are independent with respect to time. Trading space for time, I precomputed the index values prior to the entering the time loop and stored the values in two arrays. I then replaced the index calculations with reads of the arrays. Data operations Ways to reduce data operations can appear in many forms. Implementing a more efficient algorithm produces the biggest gains. The closest I came to an algorithm change was in code 4. This code computes the inner product of K-vectors A(i) and B(j), 0 = i < N, 0 = j < M, for most values of i and j. Since the program computes most of the NM possible inner products, it is more efficient to compute all the inner products in one triply-nested loop rather than one at a time when needed. The savings accrue from reading A(i) once for all B(j) vectors and from loop unrolling. for (i = 0; i < N; i+=8) { for (j = 0; j < M; j++) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (k = 0; k < K; k++) { sum0 += A[i+0][k] * B[j][k]; sum1 += A[i+1][k] * B[j][k]; sum2 += A[i+2][k] * B[j][k]; sum3 += A[i+3][k] * B[j][k]; sum4 += A[i+4][k] * B[j][k]; sum5 += A[i+5][k] * B[j][k]; sum6 += A[i+6][k] * B[j][k]; sum7 += A[i+7][k] * B[j][k]; } C[i+0][j] = sum0; C[i+1][j] = sum1; C[i+2][j] = sum2; C[i+3][j] = sum3; C[i+4][j] = sum4; C[i+5][j] = sum5; C[i+6][j] = sum6; C[i+7][j] = sum7; }} This change requires knowledge of a typical run; i.e., that most inner products are computed. The reasons for the change, however, derive from basic optimization concepts. It is the type of change easily made at development time by a knowledgeable programmer. In code 5, we have the data version of the index optimization in code 6. Here a very expensive computation is a function of the loop indices and so cannot be hoisted out of the loop; however, the computation is invariant with respect to an outer iterative loop over time. We can compute its value for each iteration of the computation loop prior to entering the time loop and save the values in an array. The increase in memory required to store the values is small in comparison to the large savings in time. The main loop in Code 8 is doubly nested. The inner loop includes a series of guarded computations; some are a function of the inner loop index but not the outer loop index while others are a function of the outer loop index but not the inner loop index for (j = 0; j < N; j++) { for (i = 0; i < M; i++) { r = i * hrmax; R = A[j]; temp = (PRM[3] == 0.0) ? 1.0 : pow(r, PRM[3]); high = temp * kcoeff * B[j] * PRM[2] * PRM[4]; low = high * PRM[6] * PRM[6] / (1.0 + pow(PRM[4] * PRM[6], 2.0)); kap = (R > PRM[6]) ? high * R * R / (1.0 + pow(PRM[4]*r, 2.0) : low * pow(R/PRM[6], PRM[5]); < rest of loop omitted > }} Note that the value of temp is invariant to j. Thus, we can hoist the computation for temp out of the loop and save its values in an array. for (i = 0; i < M; i++) { r = i * hrmax; TEMP[i] = pow(r, PRM[3]); } [N.B. – the case for PRM[3] = 0 is omitted and will be reintroduced later.] We now hoist out of the inner loop the computations invariant to i. Since the conditional guarding the value of kap is invariant to i, it behooves us to hoist the computation out of the inner loop, thereby executing the guard once rather than M times. The final version of the code is for (j = 0; j < N; j++) { R = rig[j] / 1000.; tmp1 = kcoeff * par[2] * beta[j] * par[4]; tmp2 = 1.0 + (par[4] * par[4] * par[6] * par[6]); tmp3 = 1.0 + (par[4] * par[4] * R * R); tmp4 = par[6] * par[6] / tmp2; tmp5 = R * R / tmp3; tmp6 = pow(R / par[6], par[5]); if ((par[3] == 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp5; } else if ((par[3] == 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp4 * tmp6; } else if ((par[3] != 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp5; } else if ((par[3] != 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp4 * tmp6; } for (i = 0; i < M; i++) { kap = KAP[i]; r = i * hrmax; < rest of loop omitted > } } Maybe not the prettiest piece of code, but certainly much more efficient than the original loop, Copy operations Several programs unnecessarily copy data from one data structure to another. This problem occurs in both Fortran and C programs, although it manifests itself differently in the two languages. Code 1 declares two arrays—one for old values and one for new values. At the end of each iteration, the array of new values is copied to the array of old values to reset the data structures for the next iteration. This problem occurs in Fortran programs not included in this study and in both Fortran 77 and Fortran 90 code. Introducing pointers to the arrays and swapping pointer values is an obvious way to eliminate the copying; but pointers is not a feature that many Fortran programmers know well or are comfortable using. An easy solution not involving pointers is to extend the dimension of the value array by 1 and use the last dimension to differentiate between arrays at different times. For example, if the data space is N x N, declare the array (N, N, 2). Then store the problem’s initial values in (_, _, 2) and define the scalar names new = 2 and old = 1. At the start of each iteration, swap old and new to reset the arrays. The old–new copy problem did not appear in any C program. In programs that had new and old values, the code swapped pointers to reset data structures. Where unnecessary coping did occur is in structure assignment and parameter passing. Structures in C are handled much like scalars. Assignment causes the data space of the right-hand name to be copied to the data space of the left-hand name. Similarly, when a structure is passed to a function, the data space of the actual parameter is copied to the data space of the formal parameter. If the structure is large and the assignment or function call is in an inner loop, then copying costs can grow quite large. While none of the ten programs considered here manifested this problem, it did occur in programs not included in the study. A simple fix is always to refer to structures via pointers. Optimizing loop structures Since scientific programs spend almost all their time in loops, efficient loops are the key to good performance. Conditionals, function calls, little instruction level parallelism, and large numbers of temporary values make it difficult for the compiler to generate tightly packed, highly efficient code. Conditionals and function calls introduce jumps that disrupt code flow. Users should eliminate or isolate conditionls to their own loops as much as possible. Often logical expressions can be substituted for if-then-else statements. For example, code 2 includes the following snippet MaxDelta = 0.0 do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) if (Delta > MaxDelta) MaxDelta = Delta enddo enddo if (MaxDelta .gt. 0.001) goto 200 Since the only use of MaxDelta is to control the jump to 200 and all that matters is whether or not it is greater than 0.001, I made MaxDelta a boolean and rewrote the snippet as MaxDelta = .false. do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) MaxDelta = MaxDelta .or. (Delta .gt. 0.001) enddo enddo if (MaxDelta) goto 200 thereby, eliminating the conditional expression from the inner loop. A microprocessor can execute many instructions per instruction cycle. Typically, it can execute one or more memory, floating point, integer, and jump operations. To be executed simultaneously, the operations must be independent. Thick loops tend to have more instruction level parallelism than thin loops. Moreover, they reduce memory traffice by maximizing data reuse. Loop unrolling and loop fusion are two techniques to increase the size of loop bodies. Several of the codes studied benefitted from loop unrolling, but none benefitted from loop fusion. This observation is not too surpising since it is the general tendency of programmers to write thick loops. As loops become thicker, the number of temporary values grows, increasing register pressure. If registers spill, then memory traffic increases and code flow is disrupted. A thick loop with many temporary values may execute slower than an equivalent series of thin loops. The biggest gain will be achieved if the thick loop can be split into a series of independent loops eliminating the need to write and read temporary arrays. I found such an occasion in code 10 where I split the loop do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do into two disjoint loops do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) end do end do do i = 1, n do j = 1, m C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do Conclusions Over the course of the last year, I have had the opportunity to work with over two dozen academic scientific programmers at leading research universities. Their research interests span a broad range of scientific fields. Except for two programs that relied almost exclusively on library routines (matrix multiply and fast Fourier transform), I was able to improve significantly the single processor performance of all codes. Improvements range from 2x to 15.5x with a simple average of 4.75x. Changes to the source code were at a very high level. I did not use sophisticated techniques or programming tools to discover inefficiencies or effect the changes. Only one code was parallel despite the availability of parallel systems to all developers. Clearly, we have a problem—personal scientific research codes are highly inefficient and not running parallel. The developers are unaware of simple optimization techniques to make programs run faster. They lack education in the art of code optimization and parallel programming. I do not believe we can fix the problem by publishing additional books or training manuals. To date, the developers in questions have not studied the books or manual available, and are unlikely to do so in the future. Short courses are a possible solution, but I believe they are too concentrated to be much use. The general concepts can be taught in a three or four day course, but that is not enough time for students to practice what they learn and acquire the experience to apply and extend the concepts to their codes. Practice is the key to becoming proficient at optimization. I recommend that graduate students be required to take a semester length course in optimization and parallel programming. We would never give someone access to state-of-the-art scientific equipment costing hundreds of thousands of dollars without first requiring them to demonstrate that they know how to use the equipment. Yet the criterion for time on state-of-the-art supercomputers is at most an interesting project. Requestors are never asked to demonstrate that they know how to use the system, or can use the system effectively. A semester course would teach them the required skills. Government agencies that fund academic scientific research pay for most of the computer systems supporting scientific research as well as the development of most personal scientific codes. These agencies should require graduate schools to offer a course in optimization and parallel programming as a requirement for funding. About the Author John Feo received his Ph.D. in Computer Science from The University of Texas at Austin in 1986. After graduate school, Dr. Feo worked at Lawrence Livermore National Laboratory where he was the Group Leader of the Computer Research Group and principal investigator of the Sisal Language Project. In 1997, Dr. Feo joined Tera Computer Company where he was project manager for the MTA, and oversaw the programming and evaluation of the MTA at the San Diego Supercomputer Center. In 2000, Dr. Feo joined Sun Microsystems as an HPC application specialist. He works with university research groups to optimize and parallelize scientific codes. Dr. Feo has published over two dozen research articles in the areas of parallel parallel programming, parallel programming languages, and application performance.

    Read the article

  • Web Experience Management: Segmentation & Targeting - Chalk Talk with John

    - by Michael Snow
    Today's post comes from our WebCenter friend, John Brunswick.  Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} Having trouble getting your arms around the differences between Web Content Management (WCM) and Web Experience Management (WEM)?  Told through story, the video below outlines the differences in an easy to understand manner. By following the journey of Mr. and Mrs. Smith on their adventure to find the best amusement park in two neighboring towns, we can clearly see what an impact context and relevancy play in our decision making within online channels.  Just as when we search to connect with the best products and services for our needs, the Smiths have their grandchildren coming to visit next week and finding the best park is essential to guarantee a great family vacation.  One town effectively Segments and Targets visitors to enhance their experience, reducing the effort needed to learn about their park. Have a look below to join the Smiths in their search.    Learn MORE about how you might measure up: Deliver Engaging Digital Experiences Drive Digital Marketing SuccessAccess Free Assessment Tool

    Read the article

  • Référencement : Google ressuscite la balise « Meta Keywords » pour son service Google Actualités

    Référencement : Google ressuscite la balise « Meta Keywords » Pour son service Google Actualités Détrompez-vous, les balises META keyword ne sont pas complètement tombées dans les oubliettes. Google annonce sur le site officiel de Google News une nouvelle balise-meta appelée « news_keywords » qui permet à la fois aux rédacteurs de s'exprimer librement sur leurs articles et à Google Actualités de mieux cerner les thématiques de chaque article. [IMG]http://idelways.developpez.com/news/images/Google-news-logo.jpg[/IMG] La balise META news_keywords autorise aux éditeurs de spécifier une série de mots clés séparés par des virgules pour cha...

    Read the article

  • Intel Parallel XE : Intel publie un eBook gratuit pour découvrir son outil d'optimisation des développements parallèles

    Intel Parallel XE : Intel publie un eBook gratuit Pour découvrir son outil d'optimisation des développements parallèles Les solutions Intel Parallel développées, comme leur nm l'indique, par Intel sont un ensemble d'outils qui permettent une meilleure optimisation des développements parallèles pour tirer partie des architectures multi-coeurs. « La nouvelle gamme d'outils Intel Parallel XE permet aux équipes de développement de délivrer le code en temps et en heure avec le niveau de performance le plus élevé eet le minimum de défauts du cluster au desktop jusqu'au périphériques », explique Intel. Pour aider les développeurs dans la découverte et la prise en main d'Int...

    Read the article

  • Eclipse sort la version 1.0 d'Orion, son environnement de développement Web dans le Cloud

    Eclipse sort la version 1.0 d'Orion son environnement de développement Web dans le Cloud L'événement EclipseCon Europe a été l'occasion pour les développeurs d'Orion de dévoiler la première version da la plateforme de développement dans le Cloud de la fondation. Orion avait été présenté au stade de prototype en mars 2011 et mettait à la disposition des développeurs des outils d'intégration et de développement Web pouvant être utilisés dans un navigateur, sans nécessiter l'installation d'outils supplémentaires. Après plusieurs mois de tests, de correction des bugs et d'ajout des nouveautés en fonction des retours des utilisateurs, Orion est prêt pour une utilisation en envi...

    Read the article

  • Microsoft dévoile le fonctionnement de son futur Marketplace pour Windows Phone 7, les applications

    Mise à jour du 15/06/10 Microsoft dévoile le fonctionnement de son futur Marketplace Pour Windows Phone 7 : comme sur l'AppStore les applications seront filtrées Lors de la conférence annuel du TechEd de la semaine dernière, Microsoft a - en toute discrétion - livré des informations sur sa future galerie d'applications pour Windows Phone 7. Une des confirmations les plus intéressantes du ReMIX 2010 de mai dernier (retrouvez l'intégralité du ReMIX 2010, la conférence de Microsoft France de mai dernier entièrement dédiée aux développeurs, en webcast) concerne l'appari...

    Read the article

  • WebSphere Application Server V8 : IBM améliore les capacités techniques de son serveur d'applications, testez-le gratuitement

    WebSphere Application Server V8 : meilleures capacités techniques Pour le nouveau serveur d'applications d'IBM pour développeurs, testez-le gratuitement IBM vient de présenter la nouvelle version de son serveur d'applications (le plus vendu au monde) : WebSphere Application Server V8. WAS est « en tête des benchmarks et considéré par les analystes comme le serveur d'applications le plus solide du marché », se félicite IBM. Sa nouvelle version optimise le déploiement d'applications accessibles à partir de terminaux de tous formats : PC, smartphones, tablettes, etc. Autre nouveauté, WAS v8 supporte les langages Ruby et Python, il accélère le chargement...

    Read the article

  • ColdFusion Builder 2 : Adobe annonce la disponibilité de la beta de son EDI fondé sur Eclipse

    Adobe annonce la disponibilité de la beta de ColdFusion Builder 2 son EDI fondé sur Eclipse ColdFusion Builder 2, l'environnement de développement d'Adobe fondé sur Eclipse permettant le développement d'applications ColdFusion, vient de passer en beta. Cette version intègre des améliorations qui permettent aux développeurs de tester et de déployer leurs applications plus rapidement, de personnaliser leur environnement de travail pour améliorer leur workflow , et de développer des fonctionnalités avec des extensions créées avec CFML ColdFusion Builder 2 Beta intègre les fonctionnalités suivantes*: Un onglet d'aide qui permet une navigation rapide vers la prochaine bal...

    Read the article

  • ECMA International adopte JSON comme standard, le format d'échange de données continue son ascension

    ECMA International adopte JSON comme standard, le format d'échange de données continue son ascension JSON (JavaScript Object Notation) a été adopté comme standard ECMA suite à un vote de l'Assemblée Générale. Cette nouvelle norme s'est vue attribuer le numéro 404, ce qui ne manque pas de rappeler celui du code d'erreur du protocole de communication HTTP sur le réseau Internet, renvoyé par un serveur HTTP pour indiquer que la ressource demandée (généralement une page web) n'existe pas.Rappelons...

    Read the article

  • Oracle muscle son Cloud avec de nouvelles applications hébergées et une plateforme pour Java

    Oracle muscle son Cloud Avec de nouvelles applications hébergées et une plateforme pour Java Qu'il est loin le temps où le PDG d'Oracle voyait le Cloud Computing comme une aberration technologique. Depuis, Salesforce.com a montré qu'il était une entreprise viable et s'est diversifié dans les bases de données, SAP a multiplié les rachats de spécialistes du SaaS (applications à la demande) et Microsoft a placé ses outils professionnels (ERP et

    Read the article

  • SAP multiplie les annonces sur son Cloud, nouvel environnement de développement et nouvelles applications prévus

    SAP multiplie les annonces sur son Cloud Nouvel environnement de développement et nouvelles applications prévus Le Cloud Computing était au coeur du SAPPHIRE NOW 2011 et sera au centre des préoccupation de SAP France pour l'année à venir. L'éditeur allemand affirme clairement aujourd'hui qu'il s'agit, pour lui, d'un « changement de paradigme sur le marché, d'une façon orchestrée qui offre une véritable valeur métier aux entreprises ». Pour renforcer sa stratégie, SAP multiplie d'ailleurs les annonces. Comme par exemple celle du rach...

    Read the article

  • Fixya publie son rapport sur les navigateurs mobiles les plus utilisés, Safari en tête de liste

    Fixya publie son rapport sur les navigateurs mobiles les plus utilisés, Safari en tête de liste Malgré les 73,5% de part d'Android sur le marché du smartphone, une enquête de Fixya qui a comparé l'utilisation de cinq principales plateformes de navigation ( Internet -Google- , Chrome -Google-, Opera -Opera- , Explorer -Windows- et Safari -Apple- ) révèle que Safari est le navigateur mobile le plus utilisé. Avec plus de la moitié des parts d'utilisation (58,12%), le navigateur d'Apple se distingue...

    Read the article

  • Google et Microsoft envisagent de créer une liste des utilisateurs d'IPv6 pour accélérer son adoptio

    Mise à jour du 29/03/10 NB : Les commentaires sur cette mise à jour commencent ici dans le topic Google et Microsoft discutent pour créer une liste des utilisateurs d'IPv6 Pour accélérer son adoption : à la fois bonne et mauvaise solution Google, Microsoft (ainsi que Netflix) ont entamé des discussions dans l'optique de créer une liste commune des internautes qui utiliseront l'IPv6, le futur protocole d'internet qui pourra palier à la pénurie d'adresse IPv4 (lire ci-avant). La nouvel...

    Read the article

  • Android connait une croissance record, un graphique présente son évolution sur les 18 derniers mois

    Mise à jour du 27.05.2010 par Katleen Android connait une croissance record, un graphique présente son évolution sur les 18 derniers mois Un visuel infographique très complet, retraçant l'évolution d'Android au cours des 18 derniers mois. Il s'appuie sur les dernières statistiques dévoilées par Google il y a quelques jours lors de sa conférence I/O. Des chiffres rassurants suite à l'abandon des logiciels de Google au profit de ceux de Microsoft par plusieurs constructeurs. Les points clés transmis par ces i...

    Read the article

  • Mozilla revendique 40% de part de marché européen pour Firefox, et Chrome continue son progrès selon

    Mise à jour du 02/04/10 (MAJ Djug) Mozilla revendique 40% de part de marché européen pour Firefox et Chrome continue sa monté en puissance selon les derniers chiffres de NetApplications Mozilla vient de publier un document «The State of the Internet» dans lequel elle revendique 40% de part de marché européen pour son navigateur Firefox durant le premier trimestre de 2010. [IMG]http://djug.developpez.com/rsc/firefox_share.jpg[/IMG] Selon ce document, 350 millions de personne utilisent Firefox à travers le monde ce qui représente 30% de part de marché mondiale du navigateur Web. D'un...

    Read the article

  • Les entreprises peuvent tester gratuitement Windows 7 jusqu'au 31/12/2010, Microsoft prolonge son pr

    Les entreprises peuvent tester gratuitement Windows 7 jusqu'au 31/12/2010, Microsoft prolonge son programme d'essai Depuis septembre 2009, Microsoft permet aux entreprises de tester gratuitement Windows 7 (32 ou 64 bits) pendant 90 jours. Ce programme, qui devait bientôt prendre fin, rencontre un très large succès. Aussi, Microsoft a décidé de le poursuivre jusqu'au 31 décembre 2010. Les professionnels pourront donc continuer de télécharger gratuitement Windows 7 Entreprise et de l'utiliser pendant près de 3 mois. Dans ce délai, les entreprises pourront effectuer différents tests : compatibilité applicative et matérielle, stratégies de déploiement, etc. Attention cependan...

    Read the article

  • Microsoft annule son projet de tablette Courier, l'objet ne sera pas mis en production

    Mise à jour du 30.04.2010 par Katleen Microsoft annule son projet de tablette Courier, l'objet ne sera pas mis en production L'information est courte, claire et concise. Microsoft vient à la fois de confirmer l'existence d'une tablette Courier, et d'en annoncer la mort. Voici donc un rival de moins pour l'iPad d'Apple. C'est Frank Shaw, chargé de communication pour Microsoft, qui a fait -il y a à peine quelques heures- la déclaration suivante aux médias américains : «A tout moment, de nouvelles idées sont expérimentées, testées et incubées. C'est dans l'ADN de Microsoft. Le projet «Courier» en est un exemple. Sa technologie sera évaluée pour un usage futur, mais nous ne prévoyon...

    Read the article

  • De nouvelles informations sur Windows 8 et son Windows Server associé révélées au compte goutte par

    Mise à jour du 10.06.2010 par Katleen De nouvelles informations sur Windows 8 et son Windows Server associé révélées au compte goutte par un cadre de Microsoft Si l'on se réfère au cycle de vie des produits de Microsoft, on constate qu'au niveau des clients et des serveurs d'OS les sorties alternent entre une majeure, puis une mineure, et ainsi de suite, tous les deux ans. La mise à jour la plus récente de la version serveur de Windows 7 s'appelle Windows Server 2008 R2, et elle était mineure (sortie en 2009). On peut donc logiquement s'attendre à des changements majeurs pour la prochaine mouture. Dans une interview récente, Bob Muglia, Président de l'unité Tools and Servers chez Micr...

    Read the article

  • iAds : déjà 50 % du marché américain pour la régie publicitaire d'Apple avant même son lancement, pr

    Mise à jour du 08/06/10 iAds possèderait déjà 50 % du marché US des annonces mobiles Avant même son lancement, prévu pour le 1er juillet iAds, la nouvelle régie publicitaire d'Apple pour applications mobiles, sera officiellement lancée le 1er juillet prochain. Elle concernera les applications tournant sur les iPhones et iPod Touch qui embarqueront iOS 4 (ex-iPhone OS), le nouvel OS mobile d'Apple. Lors du WWDC, Steve Jobs a d'ores et déjà dévoilé plusieurs grands noms d'annonceurs impliqués dans le projet. Parmi eux, on compte bien sûr Disney (dont Jobs est un actionnaire influent) mais aussi des marques aussi différent...

    Read the article

  • Zend dévoile son approche "Mobile First" de Zend Studio pour le développement d'applications mobiles multiplateformes

    Zend dévoile son approche "Mobile First" de Zend Studio Pour le développement d'applications mobiles multiplateformes Zend Technologies, la compagnie spécialisée en PHP, a dévoilé lors de sa dernière conférence annuelle une évolution majeure de Zend Studio. La nouvelle version du produit permet aux développeurs de créer des applications multiplateformes compatibles avec les principaux systèmes d'exploitation mobiles tels que Android et iOS. [IMG]http://www.developpez.net/forums/u51346-a17-i41.png[/IMG] Zend, fondée par Andi Gutmans et Zeev Suraski, lance une interface Cloud qui permet de créer par glisser-déposer des interfaces graphiques en ut...

    Read the article

  • Shumway : Mozilla lance son implémentation JavaScript et HTML5 de Flash, les démos du projet disponibles

    Shumway : Mozilla lance son implémentation JavaScript de Flash les utilisateurs peuvent tester le projet open source basé sur les technos du Web Même si HTML5 est vu comme un remplaçant de Flash, il existe encore une grande quantité de contenu Flash sur le Web. Pour permettre le rendu de ces contenus sans avoir besoin du lecteur Flash Player, Mozilla a lancé le projet Shumway. Shumway est une machine virtuelle écrite en JavaScript, couplé avec les technologies HTML5, qui permet le rendu des fichiers SWF sans avoir besoin de Flash Player. Shumway est construit avec deux objectifs principaux selon un billet de blog de Mozilla : faire avancer le Web ouvert en toute séc...

    Read the article

  • Le gouvernement fait un point sur l'avancement et la fiabilité d'IDéNum, son projet d'identité numér

    Mise à jour du 03.06.2010 par Katleen Le gouvernement fait un point sur l'avancement et la fiabilité d'IDéNum, son projet d'identité numérique centralisée Nathalie Kosciusko-Morizet (NKM), notre secrétaire d'État chargée de la Prospective et du Développement de l'économie numérique, a lancé en février 2010 le chantier du projet IDéNum (voir news précédente). Hier, elle avait réuni les 58 organismes partenaires de l'aventure pour une réunion de point d'étape. Parmi les organismes associés, certains souhaitent pour l'instant rester anonymes et garder leur participation secrète. Pour les autres, la liste comprend : ACSEL, AFNIC, Agorabox, Almetis, Agence nationale de la sécurité des systèmes d'informat...

    Read the article

  • Chrome dépasse la barre des 10 % de part de marché, Safari atteint son plus haut historique

    Chrome dépasse la barre des 10 % de parts de marché Safari atteint son plus haut historique Mise à jour du 02/02/11, par Hinault Romaric Le navigateur de Google continue sa progression. Au cours du mois de janvier, Chrome vient de franchir la barre symbolique des 10% (10,70% de part de marché) pour la première fois selon NetMarketShare. Le mois de janvier a été un mois record pour Chrome, mais aussi pour Safari, le navigateur d'Apple, qui a atteint pour la première fois 6,30% de part de marché. Internet Explorer en revanche a enregistré une baisse de près de 4% (56% de part de marché en janvier 2011). On note égal...

    Read the article

  • PhoneGap Build 1.0 : Adobe relance son outil de conversion d'applications mobiles multiplateforme et enrichit son offre Cloud

    PhoneGap Build 1.0 : Adobe relance l'outil de conversion d'applications mobiles multiplateforme Et enrichit son offre Cloud avec 5 autres services Dans le cadre de sa conférence Create theWeb, Adobe a dévoilé hier une palette d'outils et de services pour les designers et les développeurs. Des outils très orientés HTML5, CSS3 et JavaScript. Leur but : « créer plus facilement des sites web, des contenus numériques et des applications mobiles ». La liste est assez longue. On y trouve PhoneGap Build 1.0 et une gamme Edge qui se compose à présent de Edge Animate 1.0, EdgeInspect 1.0 (ex-Shadow), Edge Web Fonts, Edge Code et en avant-première Edge Reflow.

    Read the article

< Previous Page | 4 5 6 7 8 9 10 11 12 13 14 15  | Next Page >