#machine learning
Apr 2026
Tessera: Unlocking Heterogeneous GPUs through Kernel-Granularity Disaggregation
Tessera is presented, the first kernel disaggregation system to improve performance and cost efficiency on heterogeneous GPUs for large model inference, and shows that a heterogeneous GPU pair under Tessera can even exceed the throughput of two homogeneous high-end GPUs at a lower cost.
Tian-Cheng Hu, Jin Qin, Zheng Wang et al.
· arXiv.org · 0 citations