e-journal

On Optimal Kernel Size for Integrated CPU-GPUs – A Case Study

---

Integrated CPU-GPU architectures with a fully addressable shared memory completely eliminate any CPU-GPU data transfer overhead. Since such architectures are relatively new, it is unclear what level of interaction between the CPU and GPU attains the best energy efficiency. Too coarse grained or larger kernels with fairly low CPU - GPU interaction could cause poor utilization of the shared resources while too fine grained kernels could cause frequent interrupts of GPU Computation and performance degradation. Also larger kernels require larger shared resources causing increase in area and parasitics which affect the latency sensitive CPU cores. In this paper, we show the effect of granularity on the overall system’s energy efficiency using a synthetic workload. We describe how our framework models a truly unified shared memory in integrated architectures with frequent CPU - GPU communication.

---

Ketersediaan

Tidak ada salinan data

Informasi Detail

Judul Seri

COMPUTER ARCHITECTURE

No. Panggil

-

Penerbit

Sweden : IEEE Computer Architecture Letters., 2014

Deskripsi Fisik

IEEE COMPUTER ARCHITECTURE LETTERS, VOL. 13, NO. 1, JULY-DESEMBER 2014

Bahasa

English

ISBN/ISSN

Digital Object Ident

Klasifikasi

-

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

VOL.13 NO 2 JULY-DESEMBER 2014

Subjek

ARSITEKTUR KOMPUTER

Info Detail Spesifik

-

Pernyataan Tanggungjawab

deliza

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar