Related papers: A Novel Extensible Simulation Framework for CXL-En…

Tensor network methods are incredibly effective for simulating quantum circuits. This is due to their ability to efficiently represent and manipulate the wave-functions of large interacting quantum systems. We describe the challenges faced…

Large Language Model (LLM) inference uses an autoregressive manner to generate one token at a time, which exhibits notably lower operational intensity compared to earlier Machine Learning (ML) models such as encoder-only transformers and…

Modern distributed file systems rely on uncoordinated, per node page caches that replicate hot data locally across the cluster. While ensuring fast local access, this architecture underutilizes aggregate cluster DRAM capacity through…

Large Language Model (LLM) inference is growing increasingly complex with the rise of Mixture-of-Experts (MoE) models and disaggregated architectures that decouple components like prefill/decode (PD) or attention/FFN (AF) for heterogeneous…

PCIe devices, such as SSDs and GPUs, are pivotal in modern data centers, and their value is set to grow amidst the emergence of AI and large models. However, these devices face onboard DRAM shortage issue due to internal space limitation,…

FPGA accelerators on the NIC enable the offloading of expensive packet processing tasks from the CPU. However, FPGAs have limited resources that may need to be shared among diverse applications, and programming them is difficult. We present…

A major challenge in light-matter simulations is bridging the disparate time and length scales of electrodynamics and molecular dynamics. Current computational approaches often rely on heuristic approximations of either the electromagnetic…

Recent feature matching methods have achieved remarkable performance but lack efficiency consideration. In this paper, we revisit the mainstream detector-free matching pipeline and improve all its stages considering both accuracy and…

As customized accelerator design has become increasingly popular to keep up with the demand for high performance computing, it poses challenges for modern simulator design to adapt to such a large variety of accelerators. Existing…

We present a unified programming model for heterogeneous computing systems. Such systems integrate multiple computing accelerators and memory units to deliver higher performance than CPU-centric systems. Although heterogeneous systems have…

The continuing advancement of memory technology has not only fueled a surge in performance, but also substantially exacerbate reliability challenges. Traditional solutions have primarily focused on improving the efficiency of protection…

The A64FX CPU powers the current number one supercomputer on the Top500 list. Although it is a traditional cache-based multicore processor, its peak performance and memory bandwidth rival accelerator devices. Generating efficient code for…

The ccns3Sim project is an open source implementation of the CCNx 1.0 protocols for the NS3 simulator. We describe the implementation and several important features including modularity and process delay simulation. The ccns3Sim…

The rapid advancement of neural network applications necessitates hardware that not only accelerates computation but also adapts efficiently to dynamic processing requirements. While processing-in-pixel has emerged as a promising solution…

Large Language Models are increasingly being deployed in datacenters. Serving these models requires careful memory management, as their memory usage includes static weights, dynamic activations, and key-value caches. While static weights…

Large Language Models (LLMs) are increasingly deployed on converged Cloud and High-Performance Computing (HPC) infrastructure. However, as LLMs handle confidential inputs and are fine-tuned on costly, proprietary datasets, their heightened…

The widening gap between processor speed and storage latency has made data movement a dominant bottleneck in modern systems. Two lines of storage-layer innovation attempted to close this gap: persistent memory shortened the latency…

We present a thorough analysis of the use of modern heterogeneous systems interconnected by various cachecoherent links, including CXL, NVLink-C2C, and Infinity Fabric. We studied a wide range of server systems that combined CPUs from…

Performance in modern GPU-centric systems increasingly depends on resource management policies, including memory placement, scheduling, and observability. However, uniform policies typically yield suboptimal performance across diverse…

The rapid increase in LLM model sizes and the growing demand for long-context inference have made memory a critical bottleneck in GPU-accelerated serving systems. Although high-bandwidth memory (HBM) on GPUs offers fast access, its limited…