Reducing number of operations: The joy of algebraic ...

Reducing number of operations: The joy of algebraic ...

Reducing number of operations: The joy of algebraic transformations CS498DHP Program Optimization Number of operations and execution time Fewer number of operations does not necessarily mean shorter execution times. Because of scheduling in a parallel environment. Because of locality.

Because of communication in a parallel program. Nevertheless, although it has to be applied carefully, reducing the number of operations is one of the important optimizations. In this presentation, we discuss transformation to reduce the number of operations or reduce the length of scheduling in an idealized parallel environment where communication costs are zero.

Scheduling + Consider the expression tree: It can be shortened by applying Associativity and commutativity: [a+h+b*(c+g+d*e*f) ] or Associativity, commutativity and distributivity: [a+h+b*c+b*g+b*d*e*f].

The second expression is the sortest of the three. This means that with enough resources the third expression is the fastest although is has the most operations. + h

a * + b g +

* c * d f e Locality

Consider: do i=1.n c(i) = a(i)+b(i)+a(i)/b(i) end do do i=1,n x(i) = (a(i)+b(i))*t(i)+a(i)/b(i) end do

do i=1,n d(i) = a(i)/b(i) c(i) = a(i)+b(i)+d(i) end do do i=1,n x(i) = (a(i)+b(i))*t(i)+d(i) end do The sequence on the right executes fewer operations, but, if n is large enough, it

also incurs in more cache misses. (We assume that t is computed between the two loops so that they cannot be fused.) Communication in parallel programs Consider: cobegin

do i=1,n a(i) = .. end do send a(1:n) // receive a(1:n)

coend cobegin do i=1,n a(i) = .. end do

// do i=1,n a(i) = .. end do coend The sequence on the right executes more operation s, but it would execute faster if the send operation is expensive.

Approaches to reducing cost of computation Eliminate (syntactically) redundant computations. Apply algebraic transformations to reduce the number of operations. Decompose sequential computations for parallel execution. Apply algebraic transformations to reduce the

height of expressions trees and thus reduce execution time in a parallel environment. Elimination of redundant computations Many of the transformations were discussed in the context of compiler transformations. Common subexpression elimination Loop invariant removal

Elimination of redundant counters Loop unrolling (not discussed, but should have). It eliminates bookkeeping operations. However, compilers will not eliminate all redundant computations. Here is an example where user intervention is needed: The following sequence do i=1,n

s = a(i)+s end do do i=1,n-1 t = a(i)+t end do t May be replaced by do i=1,n-1

t = a(i)+t end do s=t+a(n) t This transformation is not usually done by compilers. 2. Another example, from C, is the loop

for (i = 0; i < n; i++) { for (j = 0; j < n; j++) { a[i,j]=0; } } Which, if a is n n, can be transformed into the loop below that has fewer bookkeeping operations.

b=a; for (i = 0; i < n*n; i++) { *b=0; b++; } Applying algebraic transformations to reduce the number of operations For example, the expressions a*(b*c)+ (b*a)*d+a*e can be transformed into (a*b)*(c+d)

+a*e by distributivity and then by associativity and distributivity into a*(b*(c+d)+e). Notice that associativity has to be applied with care. For example, suppose we are operating on floating point values and that x is very much larger than y and z=-x. Then (y+x)+z may give 0 as a result, while y+(x+z) gives y as an answer. The application of algebraic rules can be very sophisticated. Consider the

computation of xn. A nave implementation would require n-1 multiplications. However, if we represent n in binary as n=b0+2(b1+2(b2 + )) and notice that xn=xb0 (xb1+2(b2 + ))2, the number of multiplications can be reduced to O(log n). function power(x,n) (assume n>0) if n==1 then return x

if n%2==1 then return x*power(x,n-1) else x=power(x,n/2); return x*x Horners rule A polynomial A(x) = a0 + a1x + a2x + a3x + ... may be written as A(x) = a0 + x(a1 + x(a2 + x(a3 + ...))). As a result, a polynomial may be evaluated at a

point x', that is A(x') computed, in (n) time using Horner's rule. That is, repeated multiplications and additions, rather than the naive methods of raising x to powers, multiplying by the coefficient, and accumulating. Conventional matrix multiplication Asymptotic complexity: 2n3 operations Each recursion step (blocked version): 8 multiplications, 4 additions

Strassens Algorithm Asymptotic complexity: O(nlog27) = O(n2.8) operations Each recursion step: 7 multiplications, 18 additions/subtractions Asymptotic complexity is solution of T(n)=7T(n/2)+18(n/2)2 Winograd Asymptotic complexity: O(n2.8..)operations Each recursion step: 7 multiplications, 15 additions/subtractions

Parallel matrix multiplication Parallel matrix multiplication can be accomplished without redundant operations. First observe that the time to compute a sum of n elements, given enough resources, is log 2 n . Time:

Time: With sufficient replication and computational resources matrix multiplication can take just one multiplication step and log 2 n additions Copying can also be done in logarithmic steps Parallelism and redundancy Algebra rules can be applied to reduce

tree height. In some cases, the height of the tree is reduced at the expense of an increase in the number of operations Parallel Prefix Redundancy in parallel sorting. Sorting networks.

Comparator (2-sorter) inputs outputs x min(x, y)

y max(x, y) Comparison Network n/2 comparisons per stage

1 0 0 0 0

1 0 0 1 0

1 1 0 1 1

1 d stages Sorting Networks inputs outputs 1

0 0 1 0 0 1 1 Sorting Network

0 0 0 0 1 1 1 1

sorted Insertion Sort Network inputs outputs depth 2n 3 comparator stages

comparators Odd-even transposition sort O(n) O(n2) Bubblesort

O(n) O(n2) Bitonic sort O(log(n)2) O(nlog(n)2)

Odd-even mergesort O(log(n)2) O(nlog(n)2) Shellsort O(log(n)2)


Recently Viewed Presentations



    Price optimism. 1. Trend: Although price is still the single most important factor to most shoppers when food and grocery shopping, its impact as a driver of store choice has waned slightly
  • Chapter Nine - SSCC

    Chapter Nine - SSCC

    Choose: $1 in each of the three period or $8 in the third Gives constant payment x forever Cashflow Important cashflow: Perpetuity You can rent an apartment for $1000 each month (r=0.5%=0.005) You can buy it P=300.000 Renting vs buying?...
  • tropism powerpoint

    tropism powerpoint

    There are many types of tropisms : Phototropism. Geotropism Thigmotropism. 3 main types. TROPISM. Plant growth in response to a stimulus ("tropo" - "turn") *Can be positive or negative. PHOTOTROPISM. The growth response of a plant in response to light...
  • Diapositiva 1

    Diapositiva 1

    INDIETRO Piccole donne di Louisa May Alcott, apparso per la prima volta negli Stati Uniti nel 1868. Come ben si sa la storia corre da un Natale all'altro raccontando pensieri, emozioni, desideri, progetti, piccole avventure e imprese di quattro sorelle...
  • Constructing low star discrepancy point sets with genetic

    Constructing low star discrepancy point sets with genetic

    Human-Competitiveness 2/5. Criterion (D): The results are publishable in its own right as new scientific results independent of the fact they were mechanically created We have published our papers in the most prestigious journals of the field: ACM Transactions on...
  • CKEC - Debbie Waggoner - home

    CKEC - Debbie Waggoner - home

    Glenn Miller was a famous big band leader in the 1930s and '40s. During World War II, he and his big band were important ambassadors for American values, playing swing music throughout Europe and broadcasting their music over the radio...
  • Title


    Chapter 5 :: Digital Building Blocks Digital Design and Computer Architecture David Money Harris and Sarah L. Harris
  • Design Research for a Context-aware Capture System to Support ...

    Design Research for a Context-aware Capture System to Support ...

    Design Research for a Context-Aware PIM System Abe Crystal - [email protected] Metadata Research Center ... Initial version of metadata schema. Spring 2006 Deployment in one biology class, with ethnographic observation and interviews. Spring 2007 Usability testing with field biology task.