Amazon.com

About: Amazon.com is a company organization based out in Seattle, Washington, United States. It is known for research contribution in the topics: Computer science & Service (business). The organization has 13363 authors who have published 17317 publications receiving 266589 citations.

Frameworks and interfaces for offload device-based packet processing

Abstract: High-speed processing of packets to, and from, a virtualization environment can be provided while utilizing hardware-based segmentation offload and other such functionality. A hardware vendor such as a network interface card (NIC) manufacturer can enable the hardware to support open and proprietary stateless tunneling in conjunction with a protocol such as single root I/O virtualization (SR-IOV) in order to implement a virtualized overlay network. The hardware can utilize various rules, for example, that can be used by the NIC to perform certain actions, such as to encapsulate egress packets and decapsulate packets.

Vertical stratification of bat communities in primary forests of Central Amazon, Brazil

Abstract: The vertical stratification of bat communities in primary forests of the Central Amazon (80 km north of Manaus, Brazil) was investigated using capture nets in the canopy (17 to 30 m high) and in the understorey (from 0-2.5 m). Seventeen sites were sampled during one year (3398.5 mistnet-hours) and 936 individuals captured, belonging to 6 families, 29 genera and 51 species. Utilizing Non-Metric Multidimensional Scaling (NMMDS), a well-marked vertical stratification between the communities was verified, the canopy being the more utilized region. Fifteen species were exclusively captured in the canopy, 10 were predominantly captured in the canopy, and 12 species were exclus ively captured in ground nets. Species recorded and the communities they form were analysed using a matrix of guilds. The matrix obtained had 24 cells. A guild composed by background cluttered/gleaning frugivores was the richest in species (19), followed by background cluttered/gleaning insectivores (12 species). The results illustrate that when studying tropical forests it is highly desirable to involve both the lower and the upper part of the forests; otherwise the fauna would be merely subsampled, thus under-estimating the status and abundance of some species.

FetchSGD: Communication-Efficient Federated Learning with Sketching.

Abstract: Existing approaches to federated learning suffer from a communication bottleneck as well as convergence issues due to sparse client participation. In this paper we introduce a novel algorithm, called FetchSGD, to overcome these challenges. FetchSGD compresses model updates using a Count Sketch, and then takes advantage of the mergeability of sketches to combine model updates from many workers. A key insight in the design of FetchSGD is that, because the Count Sketch is linear, momentum and error accumulation can both be carried out within the sketch. This allows the algorithm to move momentum and error accumulation from clients to the central aggregator, overcoming the challenges of sparse client participation while still achieving high compression rates and good convergence. We prove that FetchSGD has favorable convergence guarantees, and we demonstrate its empirical effectiveness by training two residual networks and a transformer model.