Can traditional programming bridge the Ninja performance gap for parallel computing applications?

TLDR

Modern processors feature many cores, wide SIMD units, and complex memory hierarchies, making performance extraction difficult and leading some to argue that traditional programming approaches are inadequate. This study challenges that view by demonstrating that conventional programming techniques can achieve near‑optimal performance on multi‑core and many‑core systems for common parallel workloads. The authors quantify the “Ninja gap” between naïve C/C++ code and highly optimized implementations, then analyze algorithmic adjustments, compiler advances, and hardware support that reduce this gap. Across representative throughput benchmarks, the gap averages 24× (up to 53×) on a 6‑core Intel Core i7‑X980, but can be narrowed to about 1.3× with modest effort, a result that also holds for Intel MIC and indicates that radical language changes are unnecessary.

Abstract

Current processor trends of integrating more cores with wider SIMD units, along with a deeper and complex memory hierarchy, have made it increasingly more challenging to extract performance from applications. It is believed by some that traditional approaches to programming do not apply to these modern processors and hence radical new languages must be discovered. In this paper, we question this thinking and offer evidence in support of traditional programming methods and the performance-vs-programming effort effectiveness of common multi-core processors and upcoming many-core architectures in delivering significant speedup, and close-to-optimal performance for commonly used parallel computing workloads. We first quantify the extent of the " Ninja gap ", which is the performance gap between naively written C/C++ code that is parallelism unaware (often serial) and best-optimized code on modern multi-/many-core processors. Using a set of representative throughput computing benchmarks, we show that there is an average Ninja gap of 24X (up to 53X ) for a recent 6-core Intel® Core™ i7 X980 Westmere CPU, and that this gap if left unaddressed will inevitably increase. We show how a set of well-known algorithmic changes coupled with advancements in modern compiler technology can bring down the Ninja gap to an average of just 1.3X . These changes typically require low programming effort, as compared to the very high effort in producing Ninja code. We also discuss hardware support for programmability that can reduce the impact of these changes and even further increase programmer productivity. We show equally encouraging results for the upcoming Intel® Many Integrated Core architecture (Intel® MIC) which has more cores and wider SIMD. We thus demonstrate that we can contain the otherwise uncontrolled growth of the Ninja gap and offer a more stable and predictable performance growth over future architectures, offering strong evidence that radical language changes are not required.