ATC/OSDI 2023 BoF Roller

Reproducing Roller paper from OSDI'22. https://www.usenix.org/conference/osdi22/presentation/zhu

Hongyu Zhu, Ruofan Wu, Yijia Diao, Shanbin Ke, Haoyu Li, Chen Zhang, Jilong Xue, Lingxiao Ma, Yuqing Xia, Wei Cui, Fan Yang, Mao Yang, Lidong Zhou, Asaf Cidon and Gennady Pekhimenko (2022). ROLLER: Fast and Efficient Tensor Compilation for Deep Learning. In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22) (pp. 233-248).

It's used as an example in ATC/OSDI 2023 BoF.

The artifact reproduces Figure 16 of the paper.

Reproducibility status:

Reproduced for demoing purpose during ATC/OSDI 2023 BOF. YouTube video available: https://youtu.be/5hZDU1IFNXY

Support:

Best effort, ruidanli@uchicago.edu

Reproducibility condition:

Observe a graph similar to Figure 16 of the Roller paper.

Requirements

This experiment requires a Chameleon account with an active project allocation.

Estimated Time

Around 2 hours

35 12 3 8 Jul. 31, 2023, 6:36 PM

Authors