xPUBench: Scalable and Energy-Efficient GPU and DPU-Accelerated Network Functions

Key Points

• Energy efficiency is a key outcome, showing significant reductions in power consumption during processing.
• The analysis reveals enhancements in scalability, facilitating broader applications in various network settings.
• Assessment using GPU and DPU acceleration methods indicates improved performance metrics in network functions.
• This work supports the need for more energy-efficient solutions in computational environments, calling for further exploration.