Cilk

Abstract

Cilk (pronounced "silk") is a C-based runtime system for multithreaded parallel programming.In this paper, we document the efficiency of the Cilk work-stealing scheduler, both empirically and analytically. We show that on real and synthetic applications, the "work" and "critical path" of a Cilk computation can be used to accurately model performance. Consequently, a Cilk programmer can focus on reducing the work and critical path of his computation, insulated from load balancing and other rtmtime scheduling issues. We also prove that for the class of "fully strict" (well-structured) programs, the Cilk scheduler achieves space, time, and communication bounds all within a constant factor of optimal. The Cilk rmrtime system currently runs on the Connection Machine CM5 MPP, the Intel Paragon MPP, the Silicon Graphics Power Challenge SMP, and the MIT Phish network of workstations.Applications written in Cilk include protein folding, graphic rendering, backtrack search, and the *Socrates chess program, which won third prize in the

MCFlow: A Real-Time Multi-core Aware Middleware for Dependent Task Graphs

“…Parallel programming languages, extensions, and libraries, such as Cilk [21], OpenMP [22] and Intel Thread Building Blocks [23] provide different forms of support for parallel programs. Charm++ [24] recently has been extended to support manycore architectures [25].…”

Computational science: Emerging opportunities and challenges

“…A number of alternative programming techniques have been proposed including partitioned global address spaces [4], active messages [5], and Cilk [6]. Each of these has some advantages for fine-grained computations with complex access patterns, and some of their ideas can be found in recent parallel language efforts like Chapel [7], X10 [8], and Fortress [9].…”

Lazy Threads: Implementing a Fast Parallel Call

“…Here we describe the synchronizer transformations for two forks followed by a join. 4 In the previous section, we showed how we were able to encode the inlet address and continuation address in the return continuation. The return continuation for a child was changed at most once by the parent if the child suspended.…”

“…These approaches have been used to implement many parallel languages, e.g., Mul-T [23], Id90 [9,30], CCϩϩ [6], Charm [20], Opus [25], Cilk [4], Olden [5], and Cid [29]. The common goal is to reduce the overhead associated with managing the logical parallelism.…”

“…In many cases, the cost of the fork is reduced by severely restricting what can be done in a thread. These approaches, among others, have been used in implementing parallel programming languages such as ABCL [40], CCϩϩ [6], Charm [20], Cid [29], Cik [4], Concert [21], Id90 [9,30], Mul-T [23], and Olden [5]. Still, a fork remains substantially more expensive than a simple sequential call.…”