Register-Pressure-Aware Instruction Scheduling Using Ant Colony Optimization

Abstract

This paper describes a new approach to register-pressure-aware instruction scheduling, using Ant Colony Optimization (ACO) . ACO is a nature-inspired optimization technique that researchers have successfully applied to NP-hard sequencing problems like the Traveling Salesman Problem (TSP) and its derivatives. In this work, we describe an ACO algorithm for solving the long-standing compiler optimization problem of balancing Instruction-Level Parallelism (ILP) and Register Pressure (RP) in pre-allocation instruction scheduling. Three different cost functions are studied for estimating RP during instruction scheduling. The proposed ACO algorithm is implemented in the LLVM open-source compiler, and its performance is evaluated experimentally on three different machines with three different instruction-set architectures: Intel x86, ARM, and AMD GPU. The proposed ACO algorithm is compared to an exact Branch-and-Bound (B&B) algorithm proposed in previous work. On x86 and ARM, both algorithms are evaluated relative to LLVM's generic scheduler, while on the AMD GPU, the algorithms are evaluated relative to AMD's production scheduler. The experimental results show that using SPECrate 2017 Floating Point, the proposed algorithm gives geometric-mean improvements of 1.13% and 1.25% in execution speed on x86 and ARM, respectively, relative to the LLVM scheduler. Using PlaidML on an AMD GPU, it gives a geometric-mean improvement of 7.14% in execution speed relative to the AMD scheduler. The proposed ACO algorithm gives approximately the same execution-time results as the B&B algorithm, with each algorithm outperforming the other on a substantial number of hard scheduling regions. ACO gives better results than B&B on many large instances that B&B times out on. Both ACO and B&B outperform the LLVM algorithm on the CPU and the AMD algorithm on the GPU.

Mathematical Modeling and Exact Optimizing of University Course Scheduling COnsidering Preferences of Professors

“…Lu et al [1] optimized the scheduling problem, allocating resources and facilities, taking into account almost the entire set of constraints and desirable goals that have been determined in advance. Shobaki et al [2] and Burke et al [3] stated that the scheduling problem is a problem with four parameters: a finite set of times, a finite set of resources, and a finite set of sessions so that a series of constraints as much as possible can be observed.…”

“…As it was noted before, educational scheduling can be defined as an allocation problem that aims to allocate a set of resources (courses, professors, etc.) to a limited number of time intervals [2].…”

“…Another issue that forces universities to use a method to find a timetable is the limited time for planning. So, there is a need for an efficient method to create a weekly curriculum with high speed and quality in such a way that it meets the needs of professors and students while satisfying the UCS constraints [2].…”

Path Optimization of Welding Robot Based on Ant Colony and Genetic Algorithm

“…Data shows that the ACO algorithm and B&B model are almost the same in computing time consumption. However, ACO is superior to the B&B algorithm in a large number of hard scheduling areas [8]. In order to eliminate vegetation interference, Xia et al established a spectral mixture analysis model based on ACO.…”

“…In formula (8), V i represents the arrangement number of the vertices of the curve; therefore, the vertices of the k last B-spline curve form a characteristic polygon matrix. Let k = 3 and then write the expression of the basis function as…”