Subjects: Astronomy >> Astrophysical processes submitted time 2022-06-28
Abstract：
After years of planning for the next generation of radio telescopes, the Square Kilometer Array (SKA), the construction of the SKA phase one (SKA1) had started in July 2021.After the formal operation of SKA1, it is expected that 750 petabytes of scientifically processed data will be generated every year. The data will be stored at SKA regional centers around the world for further analysis by researchers.In this paper, the models of SKA observation station, central signal processor, scientific data processing and regional center are quantitatively analyzed. Based on the high-priority scientific observation of SKA1, the data flow evaluation at each stage and the demand for computing power of scientific data processing are obtained. Taking the current SKA1-Low and SKA1-Mid arrays as examples, the key factors affecting the layout of interference arrays including resolution, sensitivity and UV coverage are summarized. Finally, OSKAR is used for data simulation of interference array. Through the simulation of SKA1-Mid, the scalability and stability of the system are obtained. Through the simulation of SKA1-Low on CSRC-P, it can be seen that the design of prototype SKA regional center in China has been fully optimized. And the detailed requirements of computing power and the detailed information of data volume are obtained. The SKA's demand for data processing, computing and storage also requires a combination of technologies and interdisciplinary efforts from areas such as electronics, communication, information technology and computer.
Peer Review Status:
Awaiting Review
Subjects: Astronomy >> Astrophysical processes submitted time 2022-06-28
Abstract：
The Square Kilometer Array (SKA) is the largest radio telescope, and the data generated by its observations will be transmitted from Australia and South Africa to the scientific data processing center about one hundred kilometers away at first, and then distributed to various SKA Regional Centres(SRC) with a distance of tens of thousands of kilometers through high-speed network.In the SKA Phase One (SKA1) stage with a scale of 10\% of SKA, it is estimated that about 750PB of data needs to be distributed to each SRC through a network of at least 100Gbps each year. Such high network bandwidth and data scale bring great challenges to data transmission and distribution. This paper analyzes different network protocols such as TCP/UDP/HTTP and uses different software in the field of radio astronomy for testing and research, and then the optimal transmission scheme parameters under the current infrastructure of 10Gbps network are obtained. In this paper, the factors affecting high-speed transmission are discussed, and the corresponding performance optimization strategies are given.Before the real observation data of SKA1 is generated, it will provide the technical foundation for the network construction and layout of China's SKA regional center. The technical details and methods described are available for reference and use in relevant scientific applications. Finally, the challenges of future SKA network requirements are discussed and prospected.
Peer Review Status:Awaiting Review
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-05-10 Cooperative journals: 《计算机应用研究》
Abstract： Named data networking (NDN) is a band new Internet architecture, which aims to cope with the increasing growth of data traffic. NDN brings native support for consumer mobility due to its receiver-driven content retrieval model. However, producer mobility is still a challenging issue, which needs additional mechanisms to improve the data availability during the producer’s movement. Therefore, this paper proposed a scalable mobility-management mechanism, which further extends the stateful forwarding plane of NDN to support producer mobility. The mechanism built a temporary forwarding path on the name-based NDN forwarding plane through network terminals and designed a buffering and retransmission mechanism to support both delay-tolerant and delay-sensitive data traffic. Finally, this paper set up a comprehensive simulation environment in ndnSIM for the proposed evaluation and comparison against the existing classes of solutions. The simulation results show that the proposed can fully support producer mobility in the wireless environment. When the speed is 30 m/s, the packet loss rate is only 3.0 %, and the average transmission delay is 352.1ms. Additionally, the additional consumption required to support producer mobility is reduced by 49.18 % compared to the comparison scheme.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-04-07 Cooperative journals: 《计算机应用研究》
Abstract： In order to solve the problem of the existing research work focuses on the compatibility of paired items, this paper proposed an outfit compatibility prediction model based on hypergraph representation. The model constructs a fashion hypergraph based on the category information of fashion items and the collocation relationship between different fashion items in the existing dataset, where each hypernode represents an item, and each hyperedge represents an outfit made up of multiple items. To better infer outfit compatibility from the hypergraph, the model converts hypergraphs into traditional graphs, and the graph neural network is used to simulate the complex interaction between nodes. Finally, the attention mechanism is introduced to calculate the outfit compatibility score to strengthen the predictive ability of the model. Experimental results show that in the outfit fill-in-the-blank task and the outfit compatibility prediction task, the model achieved an accuracy rate of 77.29% and 96.23% respectively, which was significantly improved compared with other baseline models.
Hits
Hits
Downloads
Comment