A Holistic Framework for Automated Configuration Recommendation for Cloud Service Monitoring

This RSS batch announces multiple new arXiv papers (systems, ML, and applied AI) focused on operational efficiency, performance optimization, and privacy-preserving learning. Highlights include: a Microsoft study and modular framework for automated cloud service monitor configuration; OpenDC-STEAM, an open-source datacenter simulator for evaluating sustainability techniques and trade-offs; KernelFoundry, an evolutionary, hardware-aware GPU kernel optimizer (SYCL/CUDA); TaxBreak, a trace-driven decomposition of host-side LLM inference overhead and the Host-Device Balance Index (HDBI); a memory-