May 21, 2024 · The Droplet Search is a greedy heuristic to explore the kernel transformation space. The proposed technique is a variation of the Coordinate ...

May 23, 2024 · Kernel scheduling is the problem of finding the most efficient implementation for a computational kernel. Identifying this implementation ...

Identifying this implementation involves experimenting with the parameters of compiler optimizations, such as the size of tiling windows and unrolling factors.

TL;DR: The Droplet Search algorithm efficiently finds the optimal kernel configuration for deep learning models, significantly outperforming other AutoTVM ...

Mar 2, 2024 · Based on this observation, Droplet Search tries to find the optimal optimization parameters through the coordinate descent search technique ( ...

Mar 2, 2024 · The paper "The Droplet Search Algorithm for Kernel Scheduling", describing an optimization technique for tensor compilers, has been accepted ...

Fast and low-cost search for efficient cloud configurations for hpc workloads ... The Droplet Search Algorithm for Kernel Scheduling. M Canesche, V Rosário, E ...

Rosário, Edson Borin, Fernando Magno. Quintão Pereira: The Droplet Search Algorithm for Kernel Scheduling. ACM. Transactions on Architecture and Code ...

Jun 28, 2024 · In this work, we demonstrate the potential to reduce Ansor's search time while enhancing kernel quality by incorporating Droplet Search, an AutoTVM algorithm.

The Droplet Search kernel scheduling algorithm. This pseudo-code is a simplified version of the original presentation of the algorithm, taken from [Canesche ...