Blelloch, G. E., Chatterjee, S., Hardwick, J. C., Sipelstein, J., and Zagha, M. CVL: A C vector library. Tech. Rep. CMU-CS-93-114, School of Computer Science, Carnegie Mellon University, Feb. 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....techniques. The algorithm presented in [12] has been implemented in CILK [8, 27] which is a C based language for programming multithreaded computations. Several lightweight and ecient userlevel POSIX threads (or Pthreads [32] packages are available for shared memory programming 44 (see e.g. [43, 53]) VDS [23] is a load balancing tool which employs randomized work stealing techniques as its quantitative load balancing mechanism. An overview of existing tools that incorporate load balancing techniques and support di erent programming paradigms is provided in [26] 6 Discussion and Directions ....

G. Narlikar and G. Blelloch, \Pthreads for Dynamic and Irregular Parallelism," T.R. CMU-CS-98114, School of Computer Science, Carniege Mellon University, Pittsburgh, April 1998.

....microstructure by decomposing them into parallel loops with inter processor communication isolated between the loops. This decomposition is possible only for scans and reductions with an associative operator. The decomposition follows from the standard parallel algorithm for associative scans [45, 8]. Each processor goes through three phases as follows. 1. In phase 1, the processor sums the elements in its partition of the data. This is an elementwise loop requiring no interaction with other processors. 2. In phase 2, the sums from all the processors are scanned to obtain the correct ....

....This is an elementwise loop requiring no interaction with other processors. 2. In phase 2, the sums from all the processors are scanned to obtain the correct start offset. This involves inter processor communication (synchronization) This scanning operation can be accomplished using tree summing [8] or even with a simple serial loop [19] 3. In phase 3, the processor sums the elements of its section starting with the start offset derived from the second phase. This is another elementwise loop that requires no interaction with other processors. Note that we do not need to know the exact ....

BLELLOCH, G. E. Prefix sums and their applications. Tech. Rep. CMU-CS-90-190,School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, Nov. 1990.

....these redundancies, but, probably, there will always be some redundancy. Therefore, if the redundancy is unavoidable then it should be allowed in a PSP. But, if it is not needed, then it should be taken out, e.g. the structured types can be simulated at the higher level [3] 3 VCODE: A PSP VCODE[6, 5] is an intermediate level language which is claimed to be a candidate for PSP. In [6] authors write: VCODE is a small, simple language that could ultimately serve as a portable intermediate representation for high level data parallel languages. VCODE is a single stack, single assignment ....

Guy E. Blelloch, Siddhartha Chaterjee, Frritz Knabe, Jay Sipelstein, and Marco Zagha. VCODE Refrence Manual (Version 1.3). School of Computer Science, Carnegie Mellon University, Pittsburgh, USA, 1992.

No context found.

Blelloch, G. E., Chatterjee, S., Hardwick, J. C., Sipelstein, J., and Zagha, M. CVL: A C vector library. Tech. Rep. CMU-CS-93-114, School of Computer Science, Carnegie Mellon University, Feb. 1993.

....entire system may be ported to 1 a new parallel computer with little effort. This technique allowed us to quickly port the system to a number of parallel computers in this initial experiment. This paper describes the initial implementation scheme based on a portable vector library called CVL [3]. However, we believe that higher performance may be obtained in the future by more sophisticated compiler techniques. 1.2 The importance of nested parallelism Most data parallel languages, such as the array sublanguage of Fortran 90 [1] limit the type of the elements of a collection to base ....

....size, simply declared, or initialized from strings. Standard C stream I O is defined for vectors so that vectors may be read and written just like any other type. Int x = 3; given an initial value v Float vx(size(5) defined, all values are 0 v Int vy; declared v Int vz = [1 2 3 4 5] ; initialized vector of vector of Int v v Int vv = 1 2] 3 4 5] 6 7 8] vy cin; read from standard input The foreach construct is the basis for parallelism. In its body, the variables of the foreach argument list represent an individual element of the collection, and the ....

[Article contains additional citation context not shown here]

Guy Blelloch, Siddhartha Chatterjee, Jay Sippelstein, and Marco Zagha. CVL: a C Vector Library. School of Computer Science, Carnegie Mellon University, 1991.

....for vectors so that vectors may be read and written (sequentially) just like any other type. Note that string initialization is actual compiled into stream input, so that it incurs a runtime overhead proportional to the length of the string. Int x = 3; v Float vx(size(5) v Int vz = [1 2 3 4 5] ; v v Int vv = 1 2] 3 4 5] 6 7 8] v Int vy; vy cin; The foreach construct is the basis for parallelism. Its argument list specifies the element type of each vector and gives a new name that refers to an individual element in the body of the foreach. Inside the body, the ....

....may be read and written (sequentially) just like any other type. Note that string initialization is actual compiled into stream input, so that it incurs a runtime overhead proportional to the length of the string. Int x = 3; v Float vx(size(5) v Int vz = 1 2 3 4 5] v v Int vv = [1 2] [3 4 5] 6 7 8] v Int vy; vy cin; The foreach construct is the basis for parallelism. Its argument list specifies the element type of each vector and gives a new name that refers to an individual element in the body of the foreach. Inside the body, the statements are executed on the ....

[Article contains additional citation context not shown here]

G. Blelloch, S. Chatterjee, J. Sipelstein, and M. Zagha. CVL: a C Vector Library. School of Computer Science, Carnegie Mellon University, 1991.

....and the binary operator and identity element are parameters of the algorithm. Other parallel vector libraries have offered one of two approaches to providing scan functions. To ensure high performance, some provide specialized scan algorithms for a limited number of data types and operators [1]. This approach does not generalize to user defined types. More general libraries have accepted function pointers to allow the user to define any binary associative operator [9] This approach suffers a performance loss because the repeated invocation of the function may be unacceptably expensive. ....

G. Blelloch, S. Chatterjee, J. Sipelstein, and M. Zagha. CVL: a C Vector Library. School of Computer Science, Carnegie Mellon University, 1991.

....Chapel Hill, North Carolina 27599 3175 1 INTRODUCTION 1 1 Introduction This paper describes our efforts in implementing a version of Cvl (C Vector Library) for the MasPar MP 1 SIMD computer. Cvl is a library of rudimentary vector routines, callable from C, as described by G. Blelloch, et al. [2, 6]. UnCvl is an implementation of the Cvl library routines written in mpl, MasPar s parallel version of C. The main motivation for implementing Cvl on the MP 1 is to provide support for Nesl, a high level, portable parallel programming interface [3] Nesl compiles to an intermediate language called ....

....programs on the MasPar MP 1. 2 Problem Description The obvious goals of this project were to correctly implement all of the Cvl functions and to have the coded routines run as efficiently as possible on the MP 1 architecture. Working from the rather vague descriptions of the Cvl functions in [2, 6], and checking the results of our work against test programs known to run on the sequential version of Cvl, we have written code for all the documented functions and for many of the undocumented functions. 2 The question of efficiency is a bit more complicated since the efficiency of the library ....

G. Blelloch, S. Chatterjee, J. Sipelstein, and M. Zagha. Cvl: A C vector library. Carnegie Mellon University, School of Computer Science, 1991.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC