The Cilkview scalability analyzer

Abstract

The Cilkview scalability analyzer is a software tool for profiling, estimating scalability, and benchmarking multithreaded Cilk++ applications. Cilkview monitors logical parallelism during an instrumented execution of the Cilk++ application on a single processing core. As Cilkview executes, it analyzes logical dependencies within the computation to determine its work and span (criticalpath length). These metrics allow Cilkview to estimate parallelism and predict how the application will scale with the number of processing cores. In addition, Cilkview analyzes scheduling overhead using the concept of a "burdened dag," which allows it to diagnose performance problems in the application due to an insufficient grain size of parallel subcomputations.Cilkview employs the Pin dynamic-instrumentation framework to collect metrics during a serial execution of the application code. It operates directly on the optimized code rather than on a debug version. Metadata embedded by the Cilk++ compiler in the binary executable identifies the parallel control constructs in the executing application. This approach introduces little or no overhead to the program binary in normal runs.Cilkview can perform real-time scalability benchmarking automatically, producing gnuplot-compatible output that allows developers to compare an application's performance with the tool's predictions. If the program performs beneath the range of expectation, the programmer can be confident in seeking a cause such as insufficient memory bandwidth, false sharing, or contention, rather than inadequate parallelism or insufficient grain size.

The pochoir stencil compiler

“…It is straightforward to extend TRAP to perform r multiple cuts along each dimension to match the cache complexity of Frigo and Strumpen's algorithm while providing asymptotically more parallelism. Figure 9 shows the results of using the Cilkview scalability analyzer [20] to compare the parallelism of TRAP and STRAP on two typical benchmarks. We measured the two algorithms with uncoarsened base cases.…”

Executing Dynamic Data-Graph Computations Deterministically Using Chromatic Scheduling

“…The parallelism T 1 /T ∞ is plotted together with the actual speedup T 1 /T 12 achieved on a 12-core execution. Parallelism values were measured using the Cilkview scalability analyzer [53], which measures the work and span of a Cilk program by counting instructions. Speedup on 12 cores was computed as the ratio of the 1-core and 12-core running times.…”

A Review of Parallelization Tools and Introduction to Easypar

“…Code transformation technique removes the dependency among code segmentsand increases the possibility of concurrent execution. Prospector [91], Kremlin [6], Intel Parallel advisor [72], Cilkview [85]ParaAssist [38]and Alchemist [7]employ code transformation techniques for improving parallelization possibilities.. Parallel code generated using above mentioned techniquesneedverification.…”

“…Cilk++ [85], [92]concurrency platform helps programmer to use simple constructs in a program to parallelize a program. Cilk++ implements its own scheduler that takes care of parent child processes to be executed on different cores of multicore processor.…”