Register Efficient Dynamic Memory Allocator for GPUs. (2nd July 2015)

• Record Type: Journal Article
• Authors: Vinkler, M.; Havran, V.
• Is Part Of: Computer graphics forum. Volume 34:Number 8(2015)
• Page Start: 143
• Page End: 154
• Publication Date: 2015-07-02
• Subjects: dynamic memory allocation -- many‐core architecture -- GPU -- CUDA -- D.1.3 [Programming Techniques]: Concurrent Programming Parallel programming D.3.3 [Programming Languages]: Language Constructs and Features Dynamic storage management D.4.2 [Operating Systems]: Storage Management Allocation/deallocation strategies; Computer graphics -- Periodicals; 006.605
• DOI: 10.1111/cgf.12666
• Languages: English
• ISSNs: 0167-7055

Abstract

Abstract : We compare five existing dynamic memory allocators optimized for GPUs and show their strengths and weaknesses. In the measurements, we use three generic evaluation tests proposed in the past and we add one with a real workload, where dynamic memory allocation is used in building the k ‐d tree data structure. Following the performance analysis we propose a new dynamic memory allocator and its variants that address the limitations of the existing dynamic memory allocators. The new dynamic memory allocator uses few resources and is targeted towards large and variably sized memory allocations on massively parallel hardware architectures. Abstract: We compare five existing dynamic memory allocators optimized for GPUs and show their strengths and weaknesses. In the measurements, we use three generic evaluation tests proposed in the past and we add one with a real workload, where dynamic memory allocation is used in building the k ‐d tree data structure. Following the performance analysis we propose a new dynamic memory allocator and its variants that address the limitations of the existing dynamic memory allocators. The new dynamic memory allocator uses few resources and is targeted towards large and variably sized memory allocations on massively parallel hardware architectures.