You can outperform a cluster of high-end GPUs by intelligently mixing in your old, cheap hardware.
Tessera uses kernel-granularity disaggregation to distribute workloads across mismatched GPUs. This enables a heterogeneous mix of hardware to outperform uniform clusters, potentially ending the requirement for perfectly matched GPU pods for AI training.
Tessera: Unlocking Heterogeneous GPUs through Kernel-Granularity Disaggregation
arXiv · 2604.10180
Disaggregation maps parts of an AI workload to different types of GPUs, offering a path to utilize modern heterogeneous GPU clusters. However, existing solutions operate at a coarse granularity and are tightly coupled to specific model architectures, leaving much room for performance improvement. This paper presents Tessera, the first kernel disaggregation system to improve performance and cost efficiency on heterogeneous GPUs for large model inference. Our key insight is that kernels within a s