Finding NEMO: Nimble and Expressive Memory Observability

Beyond Monoliths: Enabling Flexible and Composable AI Systems via Memory Disaggregation

CAISA is introduced, a composable AI systems architecture that enables disaggregated memory expansion for large-scale AI workloads using CXL-based shared memory and coupling memory isolation with software-managed data orchestration decouples compute and memory resources while preserving efficient data movement, providi...

xTier: Intelligent Tiering for CXL-Enabled Memory

CXL-enabled memory expands server memory capacity, but introduces a page-placement problem: the operating system must decide which pages should reside in DRAM and which should reside on slower CXL memory. Existing systems make this tradeoff in one of two ways. Userspace controllers support flexible policies, but expose...

HostPIMSim: A High-Productivity Framework for Host-Side Processing-in-Memory Simulation

of the 24th USENIX Conference on File and Storage Technologies

MCSched: Memory-Controller-Aware Scheduling for Embodied LLM Workloads on NVIDIA Jetson

Embodied LLM systems increasingly co-locate latency-critical robotics pipelines with compute- and memory-intensive language-model inference on edge platforms such as NVIDIA Jetson. This co-location avoids cloud round trips and enables privacy-preserving, low-latency interaction, but it also creates a new source of reso...

Dorado: Scaling SmartNIC Session Tables on Commodity DDRs

Dorado is a novel design that scales SmartNIC session tables entirely on inexpensive DDR modules and uses three new techniques that extract commodity DDR performance by restructuring session table layout, decomposing processing pipelines to reduce locking, and scheduling memory accesses to minimize stalls.