Hardware-aware performance optimization of YOLOV8 for real-time applications: gpu acceleration with tensorrt and an analytical fpga comparison

Tursunaliev, Ulugbek, Tulkinov, Bakhromjon

Innovation science and technologiy · 2026-yil

Annotatsiya

Real-time computer-vision systems deployed on edge platforms are constrained as much by the surrounding softwarepipeline and the host–accelerator interface as by raw compute. This paper presents a reproducible, hardware-aware optimization studyof the lightweight YOLOv8n object detector on an NVIDIA Tesla T4 GPU and contrasts the measured GPU behavior against ananalytical projection of a Xilinx Zynq UltraScale+ (ZU9EG) FPGA accelerator. Applying TensorRT graph optimization—verticalConv–BatchNorm–activation layer fusion together with FP16 precision—raised single-stream inference throughput on the T4 from aPyTorch FP32 baseline of 62.92 FPS to 72.72 FPS, a 1.16× (15.6%) speed-up, while mean average precision on COCO val2017 fellonly marginally (mAP@0.5 from 52.8% to 52.6%). A controlled batch-size sweep then exposes a host-side preprocessing bottleneck:as the batch grows to 16, host CPU utilization saturates to 98%, while GPU utilization decreases to 8%, and end-to-end throughputdegrades to 2.05 effective FPS. We show analytically and empirically that this is a data-ingest (decode/resize) limitation of the hostpipeline rather than a compute limit of the accelerator. Finally, using device datasheets and published FPGA-accelerator figures, weproject that a spatial INT8 dataflow accelerator would deliver lower peak throughput (≈45 FPS) but markedly better energy efficiencyand deterministic per-frame latency, characteristics desirable for latency-critical edge deployment. All numerical claims are clearlyseparated into measured (GPU) and projected (FPGA) categories.

Maqola ma’lumotlari
MualliflarTursunaliev, Ulugbek, Tulkinov, Bakhromjon
JurnalInnovation science and technologiy
NashriyotТошкент давлат иқтисодиёт университети
Nashr sanasi2026-06-01
Jild2
Son6
TilIngliz
DOI10.5281/zenodo.20781147
Indeksga qo‘shilgan2026-08-26

DOI: 10.5281/zenodo.20781147 · Maqolaning asl sahifasi

Kalit so‘zlar

object detection; YOLOv8; TensorRT; layer fusion; FP16/INT8 quantization; GPU; FPGA; edge inference; energy efficiency; deterministic latency.

Iqtibos keltirish

APATursunaliev, Ulugbek, & Tulkinov, Bakhromjon (2026). HARDWARE-AWARE PERFORMANCE OPTIMIZATION OF YOLOV8 FOR REAL-TIME APPLICATIONS: GPU ACCELERATION WITH TENSORRT AND AN ANALYTICAL FPGA COMPARISON. Innovation science and technologiy, 2(6). https://doi.org/10.5281/zenodo.20781147
GOST (OAK)Tursunaliev, Ulugbek, Tulkinov, Bakhromjon HARDWARE-AWARE PERFORMANCE OPTIMIZATION OF YOLOV8 FOR REAL-TIME APPLICATIONS: GPU ACCELERATION WITH TENSORRT AND AN ANALYTICAL FPGA COMPARISON // Innovation science and technologiy. – 2026. – T. 2, № 6. – DOI: 10.5281/zenodo.20781147.
MLATursunaliev, Ulugbek, and Tulkinov, Bakhromjon. “HARDWARE-AWARE PERFORMANCE OPTIMIZATION OF YOLOV8 FOR REAL-TIME APPLICATIONS: GPU ACCELERATION WITH TENSORRT AND AN ANALYTICAL FPGA COMPARISON.” Innovation science and technologiy, 2(6), 2026. https://doi.org/10.5281/zenodo.20781147
ChicagoTursunaliev, Ulugbek, and Tulkinov, Bakhromjon. “HARDWARE-AWARE PERFORMANCE OPTIMIZATION OF YOLOV8 FOR REAL-TIME APPLICATIONS: GPU ACCELERATION WITH TENSORRT AND AN ANALYTICAL FPGA COMPARISON.” Innovation science and technologiy, 2(6) (2026). https://doi.org/10.5281/zenodo.20781147
HarvardTursunaliev, Ulugbek, and Tulkinov, Bakhromjon (2026) ‘HARDWARE-AWARE PERFORMANCE OPTIMIZATION OF YOLOV8 FOR REAL-TIME APPLICATIONS: GPU ACCELERATION WITH TENSORRT AND AN ANALYTICAL FPGA COMPARISON’, Innovation science and technologiy, 2(6). https://doi.org/10.5281/zenodo.20781147
BibTeX
@article{tursunalievulugbek2026,
  title = {HARDWARE-AWARE PERFORMANCE OPTIMIZATION OF YOLOV8 FOR REAL-TIME APPLICATIONS: GPU ACCELERATION WITH TENSORRT AND AN ANALYTICAL FPGA COMPARISON},
  author = {Tursunaliev, Ulugbek and Tulkinov, Bakhromjon},
  journal = {Innovation science and technologiy},
  year = {2026},
  volume = {2},
  number = {6},
  doi = {10.5281/zenodo.20781147},
  url = {https://doi.org/10.5281/zenodo.20781147}
}

Boshqa maqola uchun iqtibos tayyorlash

Innovation science and technologiy jurnalidan boshqa maqolalar

Innovation science and technologiy — barcha maqolalar · 2026-yil arxivi

Tursunaliev, Ulugbek — boshqa maqolalari