Memory Optimizations in an Array Language
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Standard
Memory Optimizations in an Array Language. / Munksgaard, Philip; Henriksen, Troels; Sadayappan, Ponnuswamy; Oancea, Cosmin.
Proceedings of SC 2022: International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE Computer Society Press, 2022. s. 1-15 31.Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Memory Optimizations in an Array Language
AU - Munksgaard, Philip
AU - Henriksen, Troels
AU - Sadayappan, Ponnuswamy
AU - Oancea, Cosmin
N1 - Publisher Copyright: © 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - We present a technique for introducing and op-timizing the use of memory in a functional array language, aimed at GPU execution, that supports correct-by-construction parallelism. Using linear memory access descriptors as building blocks, we define a notion of memory in the compiler IR that enables cost-free change-of-layout transformations (e.g., slicing, transposition), whose results can even be carried across control flow such as ifs/loops without manifestation in memory. The memory notion allows a graceful transition to an unsafe IR that is automatically optimized (1) to mix reads and writes to the same array inside a parallel construct, and (2) to map semantically different arrays to the same memory block. The result is code similar to what imperative users would write. Our evaluation shows that our optimizations have significant impact (1.1 x -2 x) and result in performance competitive to hand-written code from challenging benchmarks, such as Rodinia's NW, LUD, Hotspot.
AB - We present a technique for introducing and op-timizing the use of memory in a functional array language, aimed at GPU execution, that supports correct-by-construction parallelism. Using linear memory access descriptors as building blocks, we define a notion of memory in the compiler IR that enables cost-free change-of-layout transformations (e.g., slicing, transposition), whose results can even be carried across control flow such as ifs/loops without manifestation in memory. The memory notion allows a graceful transition to an unsafe IR that is automatically optimized (1) to mix reads and writes to the same array inside a parallel construct, and (2) to map semantically different arrays to the same memory block. The result is code similar to what imperative users would write. Our evaluation shows that our optimizations have significant impact (1.1 x -2 x) and result in performance competitive to hand-written code from challenging benchmarks, such as Rodinia's NW, LUD, Hotspot.
KW - functional programming
KW - GPU
KW - op-timizing compiler
KW - parallelism
UR - http://www.scopus.com/inward/record.url?scp=85149323721&partnerID=8YFLogxK
U2 - 10.1109/SC41404.2022.00036
DO - 10.1109/SC41404.2022.00036
M3 - Article in proceedings
AN - SCOPUS:85149323721
SP - 1
EP - 15
BT - Proceedings of SC 2022
PB - IEEE Computer Society Press
T2 - 2022 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2022
Y2 - 13 November 2022 through 18 November 2022
ER -
ID: 341477319