Optimization of Edge Resources for Deep Learning Application with Batch and Model Management.

As deep learning technology paves its way, real-world applications that make use of it become popular these days. Edge computing architecture is one of the service architectures to realize the deep learning based service, which makes use of the resources near the data source or client. In Edge computing architecture it becomes important to manage resource usage, and there is research on optimization of deep learning, such as pruning or binarization, which makes deep learning models more lightweight, along with the research for the efficient distribution of workloads on cloud or edge resources. Those are to reduce the workload on edge resources. In this paper, a usage optimization method with batch and model management is proposed. The proposed method is to increase the utilization of GPU resource by modifying the batch size of the input of an inference application. To this end, the inference pipelines are identified to see how the different kinds of resources are used, and then the effect of batch inference on GPU is measured. The proposed method consists of a few modules, including a tool for batch size management which is able to change a batch size with respect to the available resources, and another one for model management which supports on-the-fly update of a model. The proposed methods are implemented on a real-time video analysis application and deployed in the Kubernetes cluster as a Docker container. The result shows that the proposed method can optimize the usage of edge resources for real-time video analysis deep learning applications.

A Heterogeneous IoT Data Analysis Framework with Collaboration of Edge-Cloud Computing: Focusing on Indoor PM10 and PM2.5 Status Prediction.

The edge platform has evolved to become a part of a distributed computing environment. While typical edges do not have enough processing power to train machine learning models in real time, it is common to generate models in the cloud for use on the edge. The pattern of heterogeneous Internet of Things (IoT) data is dependent on individual circumstances. It is not easy to guarantee prediction performance when a monolithic model is used without considering the spatial characteristics of the space generating those data. In this paper, we propose a collaborative framework using a new method to select the best model for the edge from candidate models of cloud based on sample data correlation. This method lets the edge use the most suitable model without any training tasks on the edge side, and it also minimizes privacy issues. We apply the proposed method to predict future fine particulate matter concentration in an individual space. The results suggest that our method can provide better performance than the previous method.

FANCA Promotes DNA Double-Strand Break Repair by Catalyzing Single-Strand Annealing and Strand Exchange.

FANCA is a component of the Fanconi anemia (FA) core complex that activates DNA interstrand crosslink repair by monoubiquitination of FANCD2. Here, we report that purified FANCA protein catalyzes bidirectional single-strand annealing (SA) and strand exchange (SE) at a level comparable to RAD52, while a disease-causing FANCA mutant, F1263Δ, is defective in both activities. FANCG, which directly interacts with FANCA, dramatically stimulates its SA and SE activities. Alternatively, FANCB, which does not directly interact with FANCA, does not stimulate this activity. Importantly, five other patient-derived FANCA mutants also exhibit deficient SA and SE, suggesting that the biochemical activities of FANCA are relevant to the etiology of FA. A cell-based DNA double-strand break (DSB) repair assay demonstrates that FANCA plays a direct role in the single-strand annealing sub-pathway (SSA) of DSB repair by catalyzing SA, and this role is independent of the canonical FA pathway and RAD52.

Simple and generalized synthesis of oxide-metal heterostructured nanoparticles and their applications in multimodal biomedical probes.

Heterostructured nanoparticles composed of metals and Fe3O4 or MnO were synthesized by thermal decomposition of mixtures of metal-oleate complexes (for the oxide component) and metal-oleylamine complexes (for the metal component). The products included flowerlike-shaped nanoparticles of Pt-Fe3O4 and Ni-Fe3O4 and snowmanlike-shaped nanoparticles of Ag-MnO and Au-MnO. Powder X-ray diffraction patterns showed that these nanoparticles were composed of face-centered cubic (fcc)-structured Fe3O4 or MnO and fcc-structured metals. The relaxivity values of the Au-MnO and Au-Fe3O4 nanoparticles were similar to those of the MnO and Fe3O4 nanoparticles, respectively. Au-Fe3O4 heterostructured nanoparticles conjugated with two kinds of 12-base oligonucleotide sequences were able to sense a complementary 24-mer sequence, causing nanoparticle aggregation. This hybridization-mediated aggregation was detected by the overall size increase indicated by dynamic light scattering data, the red shift of the surface plasmon band of the Au component, and the enhancement of the signal intensity of the Fe3O4 component in T2-weighted magnetic resonance imaging.

Synthesis of uniform goethite nanotubes with parallelogram cross section.

Uniform goethite nanotubes were synthesized from the reaction of hydrazine with Fe(III)-oleate complex immobilized in reverse micelles. The nanotubes have interesting parallelogram cross section with uniform edge dimension of as small as 7 nm. The edge dimensions and lengths of the nanotubes were easily controlled by varying the reaction conditions.