Comparing datasets of volume servers to illuminate their energy use in data centers

摘要: As data centers proliferate, their energy intensity deserves close attention. Always-on operations and growing usage for cloud other backend processes make servers the fundamental driver of center use. Yet servers’ power draw under real-world conditions is poorly understood. This paper explores characteristics volume that affect use, quantifying differences in between higher-performing Standard Performance Evaluation Corporation (SPEC) ENERGY STAR a typical server. First, we establish general US installed base, before reporting hardware configurations from major online retail website. We then compare idle across three datasets (one unique to this paper) explain via which most sensitive. find server demand be significantly higher than benchmarks industry-released SPEC database, configurations—and likely scaling—to atypical servers. Next, examine trends among high-performing load range consider whether these are representative servers, inputting average values into recent national use model. Lastly, results two surveys IT professionals illustrate incidence more efficient equipment operational practices rooms/closets. Future work should include field measurements different sizes, accounting variations setting changes post-purchase, as well investigating linkage time efficiency.