Related papers: Thallus: An RDMA-based Columnar Data Transport Pro…
This paper considers a relay-assisted bidirectional cellular network where the base station (BS) communicates with each mobile station (MS) using OFDMA for both uplink and downlink. The goal is to improve the overall system performance by…
Data processing pipelines represent an important slice of the astronomical software library that include chains of processes that transform raw data into valuable information via data reduction and analysis. In this work we present Corral,…
Modern microservices increasingly depend on high-performance remote procedure calls (RPCs) to coordinate fine-grained, distributed computation. As network bandwidths continue to scale, the CPU overhead associated with RPC processing,…
In the digital era, data spaces are emerging as key ecosystems for the secure and controlled exchange of information among participants. To achieve this, components such as metadata catalogs and data space connectors are essential. This…
Cloud-native containerized applications constantly seek high-performance and easy-to-operate container network solutions. RDMA network is a potential enabler with higher throughput and lower latency than the standard TCP/IP network stack.…
We consider elastic resource provisioning in the cloud, focusing on in-memory key-value stores used as caches. Our goal is to dynamically scale resources to the traffic pattern minimizing the overall cost, which includes not only the…
Serverless computing promises enhanced resource efficiency and lower user costs, yet is burdened by a heavyweight, CPU-bound data plane. Prior efforts exploiting shared memory reduce overhead locally but fall short when scaling across…
Data streaming relies on continuous queries to process unbounded streams of data in a real-time fashion. It is commonly demanding in computation capacity, given that the relevant applications involve very large volumes of data. Data…
Public blockchains have spurred the growing popularity of decentralized transactions and smart contracts, but they exhibit limitations on the transaction throughput, storage, and computation. To avoid transaction gridlock, public…
Existing large language model (LLM) serving systems typically employ Prefill-Decode disaggregated architecture to prevent computational interference between the prefill and decode phases. However, in real-world LLM serving scenarios,…
Tail latency dominates the performance of synchronous pretraining jobs when running at very large scales. We describe a three-pronged approach: (1) a new RDMA-based transport protocol, MRC, sprays across many paths and actively…
The Data Science domain has expanded monumentally in both research and industry communities during the past decade, predominantly owing to the Big Data revolution. Artificial Intelligence (AI) and Machine Learning (ML) are bringing more…
RDMA is vital for efficient distributed training across datacenters, but millisecond-scale latencies complicate the design of its reliability layer. We show that depending on long-haul link characteristics, such as drop rate, distance and…
The DataFlow is sub-system of the ATLAS data acquisition responsible for the reception, buffering and subsequent movement of partial and full event data to the higher level triggers: Level 2 and Event Filter. The design of the software is…
An essential part of building a data-driven organization is the ability to handle and process continuous streams of data to discover actionable insights. The explosive growth of interconnected devices and the social Web has led to a large…
To fulfill the low latency requirements of today's applications, deployment of RDMA in datacenters has become prevalent over the recent years. However, the in-order delivery requirement of RDMAs prevents them from leveraging powerful…
Traditional data centers are designed with a rigid architecture of fit-for-purpose servers that provision resources beyond the average workload in order to deal with occasional peaks of data. Heterogeneous data centers are pushing towards…
Realizing that it is inherently difficult to precisely match the sending rates at the endhost with the available capacity on dynamic cellular links, we build a system, Octopus, that sends real-time data streams over cellular networks using…
Reinforcement learning (RL) post-training for Large Language Models (LLMs) is now scaling to large clusters and running for extended durations to enhance model reasoning performance. However, the scalability of existing RL frameworks is…
Blockchain technology, with implications in the financial domain, offers data in the form of large-scale transaction networks. Analyzing transaction networks facilitates fraud detection, market analysis, and supports government regulation.…