An OpenMP Extension that Supports Thread-Level Speculation

Abstract

OpenMP directives are the de-facto standard for shared-memory parallel programming. However, OpenMP does not guarantee the correctness of the parallel execution of a given loop if runtime data dependences arise. Consequently, many highlyparallel regions cannot be safely parallelized with OpenMP due to the possibility of a dependence violation. In this paper, we propose to augment OpenMP capabilities, by adding Thread-Level Speculation (TLS) support. Our contribution is threefold. First, we have defined a new speculative clause for variables inside parallel loops. This clause ensures that all accesses to these variables will be carried out according to sequential semantics. Second, we have created a new, software-based TLS runtime library to ensure correctness in the parallel execution of OpenMP loops that include speculative variables. Third, we have developed a new GCC plugin, which seamlessly translates our OpenMP speculative clause into calls to our TLS runtime engine. The result is the ATLaS C Compiler framework, which takes advantage of TLS techniques to expand OpenMP functionalities, and guarantees the sequential semantics of any parallelized loop.

Using the Xeon Phi Platform to Run Speculatively-Parallelized Codes

“…The real-world applications include the 2-dimensional convex hull problem (2D-hull) [10], the Delaunay triangulation problem [13,29], and a C implementation of the TREE benchmark [5]. The synthetic benchmark is the Fast [4] benchmark.…”

“…We have also used a synthetic benchmark called Fast [4], which presents almost no dependences between iterations, and which was designed to test the efficiency of the speculative scheduling mechanism, with few iterations…”

“…The ATLaS framework [4] enhances OpenMP with a new clause to allow the speculative use of variables inside a program. The use of this tool is simple: A programmer only needs to include the list of speculative variables in the predefined speculative clause of al parallel for directive.…”

“…Figure 2 outlines the data structures needed by the speculative runtime library. In this section we will only briefly describe the main characteristics of this solution: A detailed explanation can be found in [4,15]. At the top of the figure, we can see two pointers to the non-spec and the most-spec threads, that are the threads in charge to the execution of the non-speculative and most-speculative chunks of iterations.…”

An Efficient Stream Data Processing Model for Multiuser Cryptographic Service

“…erefore, the effective use of synchronization and coordination is critical to achieving high performance. One way to achieve this goal is speculative execution, which enables concurrent synchronization through thread speculation or branch prediction [5][6][7][8]. Successful speculation will reduce the portion of continuous execution, but false speculation will increase revocation and recovery overhead.…”