The paper "It’s Not Easy Being Green" by Peter Xiang Gao, Andrew R. Curtis, Bernard Wong, and S. Keshav from the Cheriton School of Computer Science at the University of Waterloo, addresses the environmental impact of large-scale Internet applications, particularly content distribution networks (CDNs). These applications consume significant amounts of electricity and generate substantial carbon emissions, which vary by location due to differences in electricity sources and usage patterns. The authors introduce FORTE (Flow Optimization based framework for request-Routing and Traffic Engineering), a framework that optimizes the allocation of user traffic to datacenters to balance access latency, electricity costs, and carbon footprint. FORTE dynamically adjusts the fraction of user traffic directed to each datacenter in response to changes in request workload and carbon footprint. The paper demonstrates that carbon taxes or credits are impractical for incentivizing carbon reduction by providers of large-scale Internet applications, but FORTE can reduce carbon emissions by 10% without increasing electricity costs or access latency. Additionally, FORTE can help determine optimal datacenter upgrade plans to further reduce carbon emissions over time. The authors also introduce Fast-FORTE, a heuristic that closely approximates FORTE but runs significantly faster, making it suitable for large-scale applications. The paper evaluates FORTE using a custom discrete-time simulator with a 24-day workload from Akamai, a major CDN provider. The results show that FORTE can achieve substantial carbon emission reductions, with the potential to reduce emissions by over 38% if datacenters have a PUE of 1.2. The paper concludes by discussing the practical implications for different types of application service providers and the trade-offs between carbon reduction, latency, and electricity costs.The paper "It’s Not Easy Being Green" by Peter Xiang Gao, Andrew R. Curtis, Bernard Wong, and S. Keshav from the Cheriton School of Computer Science at the University of Waterloo, addresses the environmental impact of large-scale Internet applications, particularly content distribution networks (CDNs). These applications consume significant amounts of electricity and generate substantial carbon emissions, which vary by location due to differences in electricity sources and usage patterns. The authors introduce FORTE (Flow Optimization based framework for request-Routing and Traffic Engineering), a framework that optimizes the allocation of user traffic to datacenters to balance access latency, electricity costs, and carbon footprint. FORTE dynamically adjusts the fraction of user traffic directed to each datacenter in response to changes in request workload and carbon footprint. The paper demonstrates that carbon taxes or credits are impractical for incentivizing carbon reduction by providers of large-scale Internet applications, but FORTE can reduce carbon emissions by 10% without increasing electricity costs or access latency. Additionally, FORTE can help determine optimal datacenter upgrade plans to further reduce carbon emissions over time. The authors also introduce Fast-FORTE, a heuristic that closely approximates FORTE but runs significantly faster, making it suitable for large-scale applications. The paper evaluates FORTE using a custom discrete-time simulator with a 24-day workload from Akamai, a major CDN provider. The results show that FORTE can achieve substantial carbon emission reductions, with the potential to reduce emissions by over 38% if datacenters have a PUE of 1.2. The paper concludes by discussing the practical implications for different types of application service providers and the trade-offs between carbon reduction, latency, and electricity costs.