2025
2025
3
SGDRC: Software-Defined Dynamic Resource Control for Concurrent DNN Inference on NVIDIA GPUs
PPoPP '25
2024
4
2024
5
2024
6
Missile: Fine-Grained, Hardware-Level GPU Resource Isolation for Multi-Tenant DNN Inference
arXiv:2407.13996
2024
7
2023
8
2023
9
Pond: The Case of CXL Memory Pooling for Cloud Datacenters
NVMW '23
2023
10
2023
2023
12
2023
13
2022
14
2021 (Pre-VT)
2020
17
2018
19
Fail-Slow at Scale: Evidence of Hardware Performance Faults in Large Production Systems
ACM TOS (Extended version of FAST '18)
Fast-Tracked
2018
21
Fail-Slow at Scale: Evidence of Hardware Performance Faults in Large Production Systems
FAST '18
Best Paper Nominee
2017
22
2017
23
Tiny-Tail Flash: Near-Perfect Elimination of Garbage Collection Tail Latencies in NAND SSDs
ACM TOS (Extended version of FAST '17)
Fast-Tracked