usenix atc '23 - smartmoe: efficiently training sparsely-activated models through combiningand...
Published 10 months ago • 285 plays • Length 11:58Download video MP4
Download video MP3
Similar videos
-
24:31
usenix atc '23 - efficient memory overcommitment for i/o passthrough enabled vms via fine-grained...
-
17:29
usenix atc '23 - bridging the gap between qoe and qos in congestion control: a large-scale mobile...
-
22:59
usenix atc '23 - secretflow-spu: a performant and user-friendly framework for privacy-preserving...
-
19:33
usenix atc '23 - accelerating distributed moe training and inference with lina
-
22:30
usenix atc '23 - rubbledb: cpu-efficient replication with nvme-of
-
19:52
usenix atc '23 - confidential computing within an ai accelerator
-
22:14
usenix atc '23 - portunus: re-imagining access control in distributed systems
-
25:51
usenix atc '23 - msrl: distributed reinforcement learning with dataflow fragments
-
3:04
hands-on with the funtech innex c570: ai autoframing & manual eptz demo
-
19:42
usenix atc '23 - adaptive online cache capacity optimization via lightweight working set size...
-
12:28
usenix security '22 - thehuzz: instruction fuzzing of processors using golden-reference models for
-
20:50
usenix atc '23 - envpipe: performance-preserving dnn training framework for saving energy
-
17:50
usenix atc '23 - sowalker: an i/o-optimized out-of-coregraph processing system for second-order...
-
20:19
usenix atc '20 - hetpipe: enabling large dnn training on (whimpy) heterogeneous gpu clusters through
-
19:57
usenix atc '23 - tidedup: a new distributed deduplication architecture for ceph4
-
19:32
usenix atc '23 - autoarts: taxonomy, insights and tools for root cause labelling of incidents in...
-
20:54
usenix atc '22 - whale: efficient giant model training over heterogeneous gpus
-
13:55
usenix atc '23 - zhuque: failure is not an option, it’s an exception
-
16:32
usenix atc '23 - avoiding the ordering trap in systems performance measurement
-
13:02
usenix atc '21 - palleon: a runtime system for efficient video processing toward dynamic class skew
-
19:14
usenix atc '23 - llfree: scalable and optionally-persistent page-frame allocation