Sorting large datasets with heterogeneous CPU/GPU architectures

Keyphrases

• Heterogeneous CPU-GPU 100%
• Heterogeneous Architectures 100%
• Global Memory 50%
• Communication Overhead 50%
• GPU Platform 50%
• Memory Capacity 50%
• NVLink 50%
• Reference Implementation 50%
• Multi-GPU 50%
• Performance Gain 50%
• GPU Communication 50%
• End-to-End Response Time 50%
• Exascale 50%
• Timing Calculations 50%
• Communication Bottleneck 50%
• Current Path 50%

Computer Science

• Large Data Set 100%
• Heterogeneous Architecture 100%
• Response Time 50%
• Communication Overhead 50%
• Memory Capacity 50%
• Reference Implementation 50%
• Performance Gain 50%