Related papers: Torrent: A Distributed DMA for Efficient and Flexi…
As the needs of Internet users and applications significantly changed over the last decade, inter-domain routing became more important to fulfill these needs. The ways how data flows over the Internet are still completely in the hand of…
The tensor-vector contraction (TVC) is the most memory-bound operation of its class and a core component of the higher-order power method (HOPM). This paper brings distributed-memory parallelization to a native TVC algorithm for dense…
The exponential increase in Machine Learning (ML) model size and complexity has driven unprecedented demand for high-performance acceleration systems. As technology scaling enables the integration of thousands of computing elements onto a…
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…
With developments of real-time applications into data centers, the need for alternatives of the standard TCP protocol has been prime demand in several applications of data centers. The several alternatives of TCP protocol has been proposed…
Datacenter network design plays a critical role in AI training by supporting scaling to thousands of accelerators. An open problem, designing a near-optimal throughput oriented network-topology, routing, and collectives-has not been…
This study proposes a new router architecture to improve the performance of dynamic allocation of virtual channels. The proposed router is designed to reduce the hardware complexity and to improve power and area consumption, simultaneously.…
Transformers achieve state-of-the-art results across many tasks, but their uniform application of quadratic self-attention to every token at every layer makes them computationally expensive. We introduce DTRNet (Dynamic Token Routing…
Distributed machine learning is becoming increasingly popular for geo-distributed data analytics, facilitating the collaborative analysis of data scattered across data centers in different regions. This paradigm eliminates the need for…
A well-known technique for enhancing the performance and stability of content distribution is the use of multiple dissemination flows. Multipath TCP (MPTCP), the most popular multiflow protocol on the Internet, allows receivers to exploit…
We present and analyze a simple and general scheme to build a churn (fault)-tolerant structured Peer-to-Peer (P2P) network. Our scheme shows how to "convert" a static network into a dynamic distributed hash table(DHT)-based P2P network such…
Engaging mobility with file sharing is considered very promising in today's run Anywhere, Anytime, Anything (3As) environments. The Bittorrent file sharing protocol can be rarely combined with the mobility scenario framework since resources…
The new generation of domain-specific AI accelerators is characterized by rapidly increasing demands for bulk data transfers, as opposed to small, latency-critical cache line transfers typical of traditional cache-coherent systems. In this…
nTorrent is a BitTorrent-like application that is based on NDN (Named Data Networking). Ad hoc environments introduce additional challenges to the dissemination of files among peers. Some issues that we encounter are that not all peers in…
Networks on Chip is a recent solution paradigm adopted to increase the performance of Multicore designs. The key idea is to interconnect various computation modules (IP cores) in a network fashion and transport packets simultaneously across…
The rapid growth of the internet in general and of bandwidth capacity at internet clients in particular poses increasing computation and bandwidth demands on internet servers. Internet access technologies like ADSL [DSL], Cable Modem and…
Memory management is necessary with the increasing number of multi-connected AI devices and data bandwidth issues. For this purpose, high-speed multi-port memory is used. The traditional multi-port memory solutions are hard-bounded to a…
Discrete multi-tone transmission (DMT) is a promising candidate for future 400G data center interconnects. Eight channels, each carrying 56 Gb/s of data can be combined in a 50-GHz channel grid to form a 400 Gb/s superchannel. For a fully…
This paper introduces a novel physical-layer method labelled as Multi-Modal Concurrent Transmission (MMCT) for efficient transmission of multiple data streams with different reliability-latency performance requirements. The MMCT arranges…
Distributed computing frameworks such as MapReduce and Spark are often used to process large-scale data computing jobs. In wireless scenarios, exchanging data among distributed nodes would seriously suffer from the communication bottleneck…