Related papers: IOPathTune: Adaptive Online Parameter Tuning for P…
The proliferation of demanding applications and edge computing establishes the need for an efficient management of the underlying computing infrastructures, urging the providers to rethink their operational methods. In this paper, we…
Self-driving cars operate in constantly changing environments and are exposed to a variety of uncertainties and disturbances. These factors render classical controllers ineffective, especially for lateral control. Therefore, an adaptive MPC…
The rapid growth of AI training has dramatically increased datacenter traffic demand and energy consumption, which has motivated renewed interest in optical circuit switches (OCSes) as a high-bandwidth, energy-efficient alternative for AI…
Applications running on parallel systems often need to join a streaming relation or a stored relation with data indexed in a parallel data storage system. Some applications also compute UDFs on the joined tuples. The join can be done at the…
The field of parallel network simulation frameworks is evolving at a great pace. That is also because of the growth of Intelligent Transportation Systems (ITS) and the necessity for cost-effective large-scale trials. In this contribution,…
The prohibitive expense of automatic performance tuning at scale has largely limited the use of autotuning to libraries for shared-memory and GPU architectures. We introduce a framework for approximate autotuning that achieves a desired…
Prompt tuning is a parameter-efficient tuning (PETuning) method for utilizing pre-trained models (PTMs) that simply prepends a soft prompt to the input and only optimizes the prompt to adapt PTMs to downstream tasks. Although it is…
Transformer-based, pre-trained large language models (LLMs) have demonstrated outstanding performance across diverse domains, particularly in the emerging {\em pretrain-then-finetune} paradigm. Low-Rank Adaptation (LoRA), a…
We introduce a user mode file system, CannyFS, that hides latency by assuming all I/O operations will succeed. The user mode process will in turn report errors, allowing proper cleanup and a repeated attempt to take place. We demonstrate…
When training a neural network for a desired task, one may prefer to adapt a pre-trained network rather than starting from randomly initialized weights. Adaptation can be useful in cases when training data is scarce, when a single learner…
Pipeline parallelism is widely used to train large language models (LLMs). However, increasing heterogeneity in model architectures exacerbates pipeline bubbles, thereby reducing training efficiency. Existing approaches overlook the…
Most online service providers deploy their own data stream processing systems in the cloud to conduct large-scale and real-time data analytics. However, such systems, e.g., Apache Heron, often adopt naive scheduling schemes to distribute…
The rapid adoption of large language models and multimodal foundation models has made multimodal data preparation pipelines critical AI infrastructure. These pipelines interleave CPU-heavy preprocessing with accelerator-backed (GPU/NPU/TPU)…
Internet routing can often be sub-optimal, with the chosen routes providing worse performance than other available policy-compliant routes. This stems from the lack of visibility into route performance at the network layer. While this is an…
Federated Learning (FL) is a decentralized learning approach that protects sensitive information by utilizing local model parameters rather than sharing clients' raw datasets. While this privacy-preserving method is widely employed across…
Recently, Mixture-of-Experts (MoE) has become one of the most popular techniques to scale pre-trained models to extraordinarily large sizes. Dynamic activation of experts allows for conditional computation, increasing the number of…
Modern machine learning training is increasingly bottlenecked by data I/O rather than compute. GPUs often sit idle at below 50% utilization waiting for data. This paper presents a machine learning approach to predict I/O performance and…
When IP-packet processing is unconditionally carried out on behalf of an operating system kernel thread, processing systems can experience overload in high incoming traffic scenarios. This is especially worrying for embedded real-time…
Fine-tuning of self-supervised models is a powerful transfer learning method in a variety of fields, including speech processing, since it can utilize generic feature representations obtained from large amounts of unlabeled data.…
Due to the diversity and implicit redundancy in terms of processing units and compute kernels, off-the-shelf heterogeneous systems offer the opportunity to detect and tolerate faults during task execution in hardware as well as in software.…