16:49ASPLOS'25 - Session 10A - Aqua: Network-Accelerated Memory Offloading for LLMs in Scale-Up GPU
Advanced
A 17-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10A - Aqua: Network-Accelerated Memory Offloading for LLMs in Scale-Up GPU”.
2.81K subscribers on YouTube
Shorts, livestreams, trailers and promos are never listed. Unless a curator approved this channel in full, only individually picked videos appear.
16:49Advanced
A 17-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10A - Aqua: Network-Accelerated Memory Offloading for LLMs in Scale-Up GPU”.
20:17Advanced
A 20-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10A - Optimizing Datalog for the GPU”.
12:25Advanced
A 12-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10A - Towards Unified Analysis of GPU Consistency”.
14:33Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10A - Virgo: Cluster-level Matrix Unit Integration in GPUs for Scalability and”.
14:33Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10B - A Software Caching Runtime for Embedded NVRAM Systems”.
17:13Advanced
A 17-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10B - AnyKey: A Key-Value SSD for All Workload Types”.
15:45Advanced
A 16-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10B - MaxEmbed: Maximizing SSD bandwidth utilization for huge embedding models”.
14:49Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10B - Simplifying and Accelerating NOR Flash I/O Stack for RAM-Restricted”.
14:17Advanced
A 14-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10C - Concurrency-Informed Orchestration for Serverless Functions”.
14:01Advanced
A 14-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10C - Dilu: Enabling GPU Resourcing-on-Demand for Serverless DL Serving via”.
16:01Advanced
A 16-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10C - Litmus: Fair Pricing for Serverless Computing”.
12:09Advanced
A 12-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10C - Medusa: Accelerating Serverless LLM Inference with Materialization”.
14:57Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10D - Instruction-Aware Cooperative TLB and Cache Replacement Policies”.
16:09Advanced
A 16-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10D - Velosiraptor: Code Synthesis for Memory Translation”.
17:45Advanced
A 18-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10D - Virtuoso: Enabling Fast and Accurate Virtual Memory Research via an”.
14:41Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 10D - vAttention: Dynamic Memory Management for Serving LLMs without”.
16:25Advanced
A 16-minute video from ACM SIGARCH: “ASPLOS'25 - Session 7D - KernelGPT: Enhanced Kernel Fuzzing via Large Language Models”.
21:05Advanced
A 21-minute video from ACM SIGARCH: “ASPLOS'25 - Session 7D - Snowplow: Effective Kernel Fuzzing with a Learned White-box Test Mutator”.
14:57Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8A - Cascade: A Dependency-aware Efficient Training Framework for Temporal”.
12:57Advanced
A 13-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8A - FastGL: A GPU-Efficient Framework for Accelerating Sampling-Based GNN”.
19:45Advanced
A 20-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8B - EDM: An Ultra-Low Latency Ethernet Fabric for Memory Disaggregation”.
15:34Advanced
A 16-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8B - Gigaflow: Pipeline-Aware Sub-Traversal Caching for Modern SmartNICs”.
17:05Advanced
A 17-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8B - Performance Prediction of On-NIC Network Functions with Multi-Resource”.
15:05Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8B - TNIC: A Trusted NIC Architecture”.
15:21Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8C - Extended User Interrupts (xUI): Fast and Flexible Notification without”.
19:37Advanced
A 20-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8C - H-Houdini: Scalable Invariant Learning”.
15:05Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8C - Stramash: A Fused-kernel Operating System For Cache-Coherent,”.
12:49Advanced
A 13-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8C - Target-Aware Implementation of Real Expressions”.
14:25Advanced
A 14-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8D - Pave: Information Flow Control for Privacy-preserving Online Data”.
14:33Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8D - Segue & ColorGuard: Optimizing SFI Performance and Scalability on Modern”.
13:21Advanced
A 13-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8D - Sharing is leaking: blocking transient-execution attacks with core-gapped”.
15:13Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 8D - TaintEMU: Decoupling Tracking from Functional Domains for”.
14:33Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 9A - Einsum Trees: An Abstraction for Optimizing the Execution of Tensor”.
15:21Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 9A - Optimizing Deep Learning Inference Efficiency through Block Dependency”.
15:13Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 9A - Pruner: A Draft-then-Verify Exploration Mechanism to Accelerate Tensor”.
15:29Advanced
A 15-minute video from ACM SIGARCH: “ASPLOS'25 - Session 9A - Relax: Composable Abstractions for End-to-End Dynamic Machine Learning”.