相关论文: Zero-CPU Collection with Direct Telemetry Access
In the realm of neural architecture design, achieving high performance is largely reliant on the manual expertise of researchers. Despite the emergence of Neural Architecture Search (NAS) as a promising technique for automating this…
Coreset selection compresses large datasets into compact, representative subsets, reducing the energy and computational burden of training deep neural networks. Existing methods are either: (i) DNN-based, which are tied to model-specific…
Models trained on data composed of different groups or domains can suffer from severe performance degradation under distribution shifts. While recent methods have largely focused on optimizing the worst-group objective, this often comes at…
Requests arriving at main memory are often different from what programmers can observe or estimate by using CPU-based monitoring. Hardware cache prefetching, memory request scheduling and interleaving cause a loss of observability that…
High-performance clusters and datacenters pose increasingly demanding requirements on storage systems. If these systems do not operate at scale, applications are doomed to become I/O bound and waste compute cycles. To accelerate the data…
Bayesian Additive Regression Trees (BART) is a fully Bayesian approach to modeling with ensembles of trees. BART can uncover complex regression functions with high dimensional regressors in a fairly automatic way and provide Bayesian…
Byte-addressable persistent memory (PM) brings hash tables the potential of low latency, cheap persistence and instant recovery. The recent advent of Intel Optane DC Persistent Memory Modules (DCPMM) further accelerates this trend. Many new…
Smart meters provide relevant information for impedance identification, but they lack global phase alignment and internal network nodes are often unobserved. A few methods for this setting were developed, but they have requirements on data…
A cornerstone of the smart grid is the advanced monitorability on its assets and operations. Increasingly pervasive installation of the phasor measurement units (PMUs) allows the so-called synchrophasor measurements to be taken roughly 100…
Dynamic voltage and frequency scaling (DVFS) and task-to-core allocation are critical for thermal management and balancing energy and performance in embedded systems. Existing approaches either rely on utilization-based heuristics that…
Phasor Measurement Units (PMUs) are placed at strategic vertices in an electrical power network to monitor the flow of power. Determining the minimum number and optimal placement of PMUs is modeled by the graph theoretic process called…
Deformable Attention Transformers (DAT) have shown remarkable performance in computer vision tasks by adaptively focusing on informative image regions. However, their data-dependent sampling mechanism introduces irregular memory access…
Deep neural networks are an extremely successful and widely used technique for various pattern recognition and machine learning tasks. Due to power and resource constraints, these computationally intensive networks are difficult to…
Datacenter networks are currently locked in a "tyranny of the single objective". While modern workloads demand diverse performance goals, ranging from coflow completion times, per-flow fairness, short-flow latencies, existing fabrics are…
Networking is considered a datacenter tax, and hyperscalers push hard to provide high-performance networking with minimal resource expenditure. To keep up with the ever-increasing network rates, many CPU cycles are spent on the networking…
Conventional wisdom holds that an efficient interface between an OS running on a CPU and a high-bandwidth I/O device should use Direct Memory Access (DMA) to offload data transfer, descriptor rings for buffering and queuing, and interrupts…
Field-aware Factorization Machines (FFMs) have emerged as a powerful model for click-through rate prediction, particularly excelling in capturing complex feature interactions. In this work, we present an in-depth analysis of our in-house,…
This work presents P4TE, an in-band traffic monitoring, load-aware packet forwarding, and flow rate controlling mechanism for traffic engineering in fat-tree topology-based data center networks using PISA switches. It achieves sub-RTT…
We address the problem of efficiently gathering correlated data from a wired or a wireless sensor network, with the aim of designing algorithms with provable optimality guarantees, and understanding how close we can get to the known…
Non-linear detection schemes can substantially improve the achievable throughput and connectivity capabilities of uplink MU-MIMO systems that employ linear detection. However, the complexity requirements of existing non-linear soft…