Related papers: Agora: Bridging the GPU Cloud Resource-Price Disco…
Recent developments in Machine Learning and Deep Learning depend heavily on cloud computing and specialized hardware, such as GPUs and TPUs. This forces those using those models to trust private data to cloud servers. Such scenario has…
In the near future FPGAs will be available by the hour, however this new Infrastructure as a Service (IaaS) usage mode presents both an opportunity and a challenge: The opportunity is that programmers can potentially trade resources for…
The rise of disaggregated AI GPUs has exposed a critical bottleneck in large-scale attention workloads: non-uniform memory access (NUMA). As multi-chiplet designs become the norm for scaling compute capabilities, memory latency and…
Language models have seen enormous progress on advanced benchmarks in recent years, but much of this progress has only been possible by using more costly models. Benchmarks may therefore present a warped picture of progress in practical…
Cloud computing as a fairly new commercial paradigm, widely investigated by different researchers, already has a great range of challenges. Pricing is a major problem in Cloud computing marketplace; as providers are competing to attract…
The rapid growth of large language model (LLM) services imposes increasing demands on distributed GPU inference infrastructure. Most existing scheduling systems follow a reactive paradigm, relying solely on the current system state to make…
The explosive growth of AI applications has created unprecedented demand for GPU resources. Cloud providers meet this demand through GPU-as-a-Service platforms that offer rentable GPU resources for running AI workloads. In this context, the…
The congestion pricing is an efficient allocation approach to mediate demand and supply of network resources. Different from the previous pricing using Affine Marginal Cost (AMC), we focus on studying the game between network coding and…
The application resource specification--a static specification of several parameters such as the number of threads and the scratchpad memory usage per thread block--forms a critical component of modern GPU programming models. This…
We empirically characterise the cost-efficiency deficit between cloud Tensor Processing Units and GPUs for finite-field cryptography. Against A100 GPU baselines (cuZK), we measure a $[5{,}558\times, 6{,}908\times]$ deficit across v5p and v4…
The latest Graphics Processing Units (GPUs) are reported to reach up to 200 billion floating point operations per second (200 Gflops) and to have price performance of 0.1 cents per M flop. These facts raise great interest in the…
Modern Graphics Processing Units (GPUs) are well provisioned to support the concurrent execution of thousands of threads. Unfortunately, different bottlenecks during execution and heterogeneous application requirements create imbalances in…
Dynamic optimization is currently limited by sensitivity computations that require information from full forward and adjoint wave fields. Since the forward and adjoint solutions are computed in opposing time directions, the forward solution…
The usage of large language models (LLMs) has grown increasingly fragmented, with no single model dominating. Meanwhile, cloud providers offer a wide range of mid-tier and older-generation GPUs that enjoy better availability and deliver…
Trends in hardware, the prevalence of the cloud, and the rise of highly demanding applications have ushered an era of specialization that quickly changes how data is processed at scale. These changes are likely to continue and accelerate in…
Distributed dataflow systems like Spark and Flink enable data-parallel processing of large datasets on clusters of cloud resources. Yet, selecting appropriate computational resources for dataflow jobs is often challenging. For efficient…
In the post-Moore's Law era, relying solely on hardware advancements for automatic performance gains is no longer feasible without increased energy consumption, due to the end of Dennard scaling. Consequently, computing accounts for an…
Matrix factorization (MF) is employed by many popular algorithms, e.g., collaborative filtering. The emerging GPU technology, with massively multicore and high intra-chip memory bandwidth but limited memory capacity, presents an opportunity…
We exploit floating-point DSPs in the Arria10 FPGA and multi-pumping feature of the M20K RAMs to build a dataflow-driven soft processor fabric for large graph workloads. In this paper, we introduce the idea of out-of-order node scheduling…
Structural parameters are normally extracted from observed galaxies by fitting analytic light profiles to the observations. Obtaining accurate fits to high-resolution images is a computationally expensive task, requiring many model…