Related papers: Performance Isolation and Semantic Determinism in …
Cloud-based computing infrastructure provides an efficient means to support real-time processing workloads, e.g., virtualized base station processing, and collaborative video conferencing. This paper addresses resource allocation for a…
The development of composable systems architecture marks a significant shift in resource allocation and utilization within data centers. This paper presents a composable architecture scaling up to 32 GPUs on a single node, addressing the…
Chip multiprocessors (CMPs) are ubiquitous in most of today's computing fields. Although they provide noticeable benefits in terms of performance, cost and power efficiency, they also introduce some new issues. In this paper we analyze how…
Coded computation techniques provide robustness against straggling servers in distributed computing, with the following limitations: First, they increase decoding complexity. Second, they ignore computations carried out by straggling…
Large model training beyond tens of thousands of GPUs is an uncharted territory. At such scales, disruptions to the training process are not a matter of if, but a matter of when -- a stochastic process degrading training productivity.…
Future experiments in high-energy physics will pose stringent requirements to computing, in particular to real-time data processing. As an example, the CBM experiment at FAIR Germany intends to perform online data selection exclusively in…
Designing efficient and scalable sparse linear algebra kernels on modern multi-GPU based HPC systems is a daunting task due to significant irregular memory references and workload imbalance across the GPUs. This is particularly the case for…
Semantic communication (SC) is recognized as a promising approach for enabling reliable communication with minimal data transfer while maintaining seamless connectivity for a group of wireless users. Unlocking the advantages of SC for…
Cloud-based serverless computing is an increasingly popular computing paradigm. In this paradigm, different services have diverse computing requirements that justify deploying an inconsistently Heterogeneous Computing (HC) system to…
We introduce Capsule, a mechanism for seamlessly sharing datacenter resources across multiple players. It decouples player-local and global states to achieve isolation and to maximize cross-player sharing. Our evaluations show that Capsule…
To automatically tune configurations for the best possible system performance (e.g., runtime or throughput), much work has been focused on designing intelligent heuristics in a tuner. However, existing tuner designs have mostly ignored the…
Many-core accelerators, as represented by the XeonPhi coprocessors and GPGPUs, allow software to exploit spatial and temporal sharing of computing resources to improve the overall system performance. To unlock this performance potential…
The paradigm shift towards multi-core and heterogeneous computing, driven by the fundamental power and thermal limits of single-core processors, has established energy efficiency as a first-class design constraint in high-performance…
Large language models like GPT-4 are resource-intensive, but recent advancements suggest that smaller, specialized experts can outperform the monolithic models on specific tasks. The Collaboration-of-Experts (CoE) approach integrates…
Query co-processing on graphics processors (GPUs) has become an effective means to improve the performance of main memory databases. However, the relatively low bandwidth and high latency of the PCI-e bus are usually bottleneck issues for…
Reducing the average memory access time is crucial for improving the performance of applications running on multi-core architectures. With workload consolidation this becomes increasingly challenging due to shared resource contention.…
Training Graph Neural Networks (GNNs) on real-world graphs consisting of billions of nodes and edges is quite challenging, primarily due to the substantial memory needed to store the graph and its intermediate node and edge features, and…
Prevailing LLM serving engines employ expert parallelism (EP) to implement multi-device inference of massive MoE models. However, the efficiency of expert parallel inference is largely bounded by inter-device communication, as EP embraces…
Discrete optimization is a central problem in artificial intelligence. The optimization of the aggregated cost of a network of cost functions arises in a variety of problems including (W)CSP, DCOP, as well as optimization in stochastic…
OpenMP is a shared memory programming model which supports the offloading of target regions to accelerators such as NVIDIA GPUs. The implementation in Clang/LLVM aims to deliver a generic GPU compilation toolchain that supports both the…