相关论文: Pangloss: a novel Markov chain prefetcher

State-of-the-art optimization is steadily shifting towards massively parallel pipelines with extremely large batch sizes. As a consequence, CPU-bound preprocessing and disk/memory/network operations have emerged as new performance…

Paxos is a prominent theory of state machine replication. Recent data intensive Systems those implement state machine replication generally require high throughput. Earlier versions of Paxos as few of them are classical Paxos, fast Paxos…

Coflow provides a key application-layer abstraction for capturing communication patterns, enabling the efficient coordination of parallel data flows to reduce job completion times in distributed systems. Modern data center networks (DCNs)…

We present a general framework for accelerating a large class of widely used Markov chain Monte Carlo (MCMC) algorithms. Our approach exploits fast, iterative approximations to the target density to speculatively evaluate many potential…

We present the design of a new large scale orchestration layer for accelerators. Our system, Pathways, is explicitly designed to enable exploration of new systems and ML research ideas, while retaining state of the art performance for…

Stochastic network calculus is an evolving theory which accounts for statistical multiplexing and uses an envelope approach for probabilistic delay and backlog analysis of networks. One of the key ideas of stochastic network calculus is the…

To maximize the performance of concurrent data structures, researchers have often turned to highly complex fine-grained techniques, resulting in efficient and elegant algorithms, which can however be often difficult to understand and prove…

HipMCL is a high-performance distributed memory implementation of the popular Markov Cluster Algorithm (MCL) and can cluster large-scale networks within hours using a few thousand CPU-equipped nodes. It relies on sparse matrix computations…

The evolution of 5G and Beyond networks has enabled new applications with stringent end-to-end latency requirements, but providing reliable low-latency service with high throughput over public wireless networks is still a significant…

Caches only exploit spatial and temporal locality in a set of address referenced in a program. Due to dynamic construction of linked data-structures, they are difficult to cache as the spatial locality between the nodes is highly dependent…

Efficient parallelism is necessary for achieving low-latency, high-throughput inference with large language models (LLMs). Tensor parallelism (TP) is the state-of-the-art method for reducing LLM response latency, however GPU communications…

We consider scheduling in a quantum switch with stochastic entanglement generation, finite quantum memories, and decoherence. The objective is to design a scheduling algorithm with polynomial-time computational complexity that stabilizes a…

FPGA accelerators for lightweight neural convolutional networks (LWCNNs) have recently attracted significant attention. Most existing LWCNN accelerators focus on single-Computing-Engine (CE) architecture with local optimization. However,…

Consensus protocols are the foundation for building fault-tolerant, distributed systems, and services. They are also widely acknowledged as performance bottlenecks. Several recent systems have proposed accelerating these protocols using the…

Masked Diffusion Models (MDMs) offer a promising alternative to autoregressive language models by enabling parallel token generation and bidirectional context modeling. However, their inference speed is significantly limited by the…

AI infrastructures, predominantly GPUs, have delivered remarkable performance gains for deep learning. Conversely, scientific computing, exemplified by quantum chemistry systems, suffers from dynamic diversity, where computational patterns…

We study the optimal provision of information for two natural performance measures of queuing systems: throughput and makespan. A set of parallel links is equipped with deterministic capacities and stochastic travel times where the latter…

Increasingly stringent throughput and latency requirements in datacenter networks demand fast and accurate congestion control. We observe that the reaction time and accuracy of existing datacenter congestion control schemes are inherently…

Classifying network traffic according to their application-layer protocols is an important task in modern networks for traffic management and network security. Existing payload-based or statistical methods of application identification…

We develop a new bidirectional algorithm for estimating Markov chain multi-step transition probabilities: given a Markov chain, we want to estimate the probability of hitting a given target state in $\ell$ steps after starting from a given…