usenix atc '22 - serving heterogeneous machine learning models on multi-gpu servers...
Published 2 years ago • 656 plays • Length 16:39Download video MP4
Download video MP3
Similar videos
-
1:14:07
serving heterogeneous machine learning models on multi-gpu servers with spatio-temporal sharing
-
19:58
usenix atc '22 - memory harvesting in multi-gpu systems with hierarchical unified virtual memory
-
20:54
usenix atc '22 - whale: efficient giant model training over heterogeneous gpus
-
19:55
usenix atc '24 - metis: fast automatic distributed training on heterogeneous gpus
-
14:52
usenix atc '21 - infaas: automated model-less inference serving
-
19:44
usenix atc '22 - tetris: memory-efficient serverless inference through tensor sharing
-
19:18
usenix atc '19 - analysis of large-scale multi-tenant gpu clusters for dnn training workloads
-
25:10
usenix atc '23 - vectorvisor: a binary translation scheme for throughput-oriented gpu acceleration
-
18:26
usenix atc '23 - tc-gnn: bridging sparse gnn computation and dense tensor cores on gpus
-
1:24:55
gpu series: multi-gpu programming part 1
-
18:49
usenix atc '22 - crisp: critical path analysis of large-scale microservice architectures
-
20:30
usenix atc '24 - power-aware deep learning model serving with μ-serve
-
12:47
usenix atc '22 - sibylla: to retry or not to retry on deep learning job failure
-
20:51
usenix atc '24 - evaluating chiplet-based large-scale interconnection networks via cycle-accurate...
-
19:35
usenix atc '22 - zero-change object transmission for distributed big data analytics
-
15:16
usenix atc '21 - fair scheduling for avx2 and avx-512 workloads
-
12:15
usenix security '21 - gforce: gpu-friendly oblivious and rapid neural network inference
-
35:27
srecon22 asia/pacific - lessons learned building a global synthetic monitoring system