The field of sustainable computing and data center management is moving towards the development of innovative solutions that optimize energy efficiency, reduce carbon footprint, and improve network performance. Researchers are focusing on creating benchmarks and simulation environments that can accurately model the complex interactions between data center operations, network dynamics, and environmental factors. This has led to the development of new scheduling algorithms and optimization techniques that can dynamically reassign tasks and resources to minimize energy consumption and carbon emissions. Additionally, there is a growing interest in improving congestion control mechanisms in data center networks to reduce latency and improve responsiveness. Noteworthy papers in this area include: DCcluster-Opt, which presents a high-fidelity simulation benchmark for sustainable geo-temporal task scheduling. FREESH, which proposes a fair and energy-efficient scheduling algorithm for LLM serving on heterogeneous GPUs. Improving dynamic congestion isolation in data-center networks, which introduces a new CI mechanism that efficiently combines CI and DCQCN to reduce false-positive congestion detection and improve responsiveness.