[spcl_bcast] co-optimization of computation and data layout to optimize data movement
Published 2 years ago • 389 plays • Length 41:50Download video MP4
Download video MP3
Similar videos
-
![[spcl_bcast] optimization of data movement for convolutional neural networks](https://i.ytimg.com/vi/t5kENP4oehk/mqdefault.jpg)
51:18
[spcl_bcast] optimization of data movement for convolutional neural networks
-
![[spcl_bcast] self-adjusting networks](https://i.ytimg.com/vi/Npz0MsnN3qQ/mqdefault.jpg)
40:27
[spcl_bcast] self-adjusting networks
-
![[spcl_bcast] scalable graph machine learning](https://i.ytimg.com/vi/gOD24-HX5Vg/mqdefault.jpg)
59:51
[spcl_bcast] scalable graph machine learning
-
17:11
data movement is all you need: a case study on optimizing transformers
-
![[spcl_bcast] capturing computation with algorithmic alignment](https://i.ytimg.com/vi/Kqc4OmiTwCM/mqdefault.jpg)
1:01:01
[spcl_bcast] capturing computation with algorithmic alignment
-
14:05
hexamesh: scaling to hundreds of chiplets with an optimized chiplet arrangement
-
35:45
telos fireside chat: exploring zero-knowledge technology at ethcc brussels
-
24:28
using functional reactive programming with js-libp2p and cells - henri binszstok
-
![[spcl_bcast] a chiplet based generative inference architecture with block floating point datatypes](https://i.ytimg.com/vi/y0WqPI3A3Mw/mqdefault.jpg)
1:06:21
[spcl_bcast] a chiplet based generative inference architecture with block floating point datatypes
-
44:49
ai-driven performance metaprogramming
-
![[spcl_bcast] automating distributed heterogeneous computing for domain experts](https://i.ytimg.com/vi/8-oywYBHyM8/mqdefault.jpg)
55:50
[spcl_bcast] automating distributed heterogeneous computing for domain experts
-
![[spcl_bcast] light-weight performance analysis for next-generation hpc systems](https://i.ytimg.com/vi/Tt-7JfvXSog/mqdefault.jpg)
53:28
[spcl_bcast] light-weight performance analysis for next-generation hpc systems
-
![[spcl_bcast] next-generation networks for machine learning](https://i.ytimg.com/vi/r5bQzW8ZUUw/mqdefault.jpg)
49:57
[spcl_bcast] next-generation networks for machine learning
-
![[spcl_bcast] parallel sparse matrix algorithms for data analysis and machine learning](https://i.ytimg.com/vi/8ZoAZzkhjkc/mqdefault.jpg)
57:45
[spcl_bcast] parallel sparse matrix algorithms for data analysis and machine learning
-
![[spcl_bcast] high-performance sparse tensor operations in hiparti library](https://i.ytimg.com/vi/_V0fhkbnMmk/mqdefault.jpg)
41:44
[spcl_bcast] high-performance sparse tensor operations in hiparti library
-
![[spcl_bcast] heterogeneous serverless computing](https://i.ytimg.com/vi/9NCJt0aAgnM/mqdefault.jpg)
47:17
[spcl_bcast] heterogeneous serverless computing
-
![[spcl_bcast] evaluating modern programming models using the parallel research kernels](https://i.ytimg.com/vi/bXeDfA21-VA/mqdefault.jpg)
57:28
[spcl_bcast] evaluating modern programming models using the parallel research kernels
-
![[spcl_bcast] innovating the next discontinuity](https://i.ytimg.com/vi/KCrQtpx31CQ/mqdefault.jpg)
1:03:56
[spcl_bcast] innovating the next discontinuity
-
![[spcl_bcast] challenges of scaling deep learning on hpc systems](https://i.ytimg.com/vi/Hd2cTcmWkvM/mqdefault.jpg)
59:16
[spcl_bcast] challenges of scaling deep learning on hpc systems