@article{isthegpuhalfemptyorhalffullpractical, title = {Is the GPU Half-Empty or Half-Full? Practical Scheduling Techniques for LLMs}, author = {Ferdi Kossmann and Bruce Fontaine and Daya Khudia and Michael Cafarella and Samuel Madden}, year = {2024}, eprint = {2410.17840}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2410.17840v2}, }