Deep Q-network-based auto scaling for service in a multi-access edge computing environment

In 5G networks, it is necessary to provide services while meeting various service requirements, such as high data rates and low latency, in response to dynamic network conditions. Multi-access edge computing (MEC) is a promising concept to meet these requirements. The MEC environment enables service providers to deploy their low latency services that are composed of multiple components. However, operating a service manually and attempting to satisfy the quality of service (QoS) requirements is difficult because many factors need to be considered in an MEC scenario. In this paper, we propose an auto-scaling method using deep Q-networks (DQN), which is a reinforcement learning algorithm, to resize the number of instances assigned to service. In our evaluation, compared to other baseline methods, the proposed approach maintains the appropriate number of instances effectively in response to dynamic traffic change while satisfying QoS and minimizing the cost of operating the service in the MEC environment. The proposed method was implemented as a module running in OpenStack and published as open-source software.     

Adaptive path finding algorithm in dynamic environment for warehouse robot

Neural Computing and Applications - Warehouse robots have been widely used by manufacturers and online retailer to automate good delivery process. One of the fundamental components when designing a...     

How latecomers catch up to leaders in high-energy physics as Big Science: transition from national system to international collaboration

Scientometrics - This study describes the increase of research productivity of latecomer countries (latecomers) in the high-energy physics (HEP) community by research strategies based on a national...     

Developing a custom DSP for vision based human computer interaction applications

Multimedia Tools and Applications - As the computing power of modern devices become greater, computer vision is increasingly adopted as the means of human-computer interaction. The industry is...     

Low temperature bonding technology for 3D integration

3D integration provides a promising solution to achieve system level integration with high function density, small form factor, enhanced transmission speed and low power consumption. Stacked bonding is the key technology to enable the communication between different strata of the 3D integration system. Low temperature bonding approaches are explored in industry to solve the performance degradation issue of the integrated devices. In this paper, various low temperature bonding technologies are reviewed and introduced, as well as the latest developments in world-wide companies and research institutes. The outlook for industrial application is also addressed in the paper.     

Effect of green mate in overweight volunteers: A randomized placebo-controlled human study

Overweight and obesity have become a global epidemic and they may impair health. Traditional use and growing evidence indicate that mate (Ilex paraguariensis A. St.-Hil.) may be helpful in losing excessive weight and fat. The present randomized, double-blind, placebo-controlled study evaluated efficacy and safety of an extract from green mate in 60 overweight subjects aged 20–39 years during 6-weeks. Body composition was measured by Dual-Energy X-ray Absorptiometry (DEXA) at baseline and after 6 weeks. Body weight, body mass index (BMI), waist circumference (WC) and various safety parameters were monitored. After 6 weeks, subjects taking mate experienced a significantly greater reduction of percent of body fat (?0.3% vs. +0.6%, p = 0.04) and fat mass (?0.5 kg vs. +0.2 kg, p = 0.04) than placebo. No significant differences were observed in other measurements. No adverse events occurred and all safety parameters were within normal ranges during the study in both groups. Thus, taking green mate extract reduced body fat after 6 weeks, while the treatment was safe and well tolerated.     

Comparison of the Magnetic Ordering in Sr2Yb1Ru0.9Cu0.1O6 and Ru1Sr2Gd1Cu2O8 Superconductors

Ru₁Sr₂Gd₁Cu₂O₈ and Sr₂Yb₁Ru_0.9Cu_0.1O₆ were synthesized using a solid-state reaction. X-ray diffraction showed that Ru₁Sr₂Gd₁Cu₂O₈ crystallized into the tetragonal Ru-1212 structure (space group P4/mmm), whereas the Sr₂Yb₁Ru_0.9Cu_0.1O₆ had a monoclynic structure (space group P2₁/n). Sr₂Yb₁Ru_0.9Cu_0.1O₆ showed a possible magnetic transition ( T _{m a g} ) at 44 K with a superconducting transition temperature T _c at 30 K. In contrast, Ru₁Sr₂Gd₁Cu₂O₈ showed a magnetic transition at 140 K with a T _c at 25 K. The magnetization data for Sr₂Yb₁Ru_0.9Cu_0.1O₆ suggests that this sample exhibits antiferromagnetic ordering in zero-field-cooled (ZFC) magnetization, whereas it reveals a weak ferromagnetic component during field-cooled (FC) cycles. In addition, larger diamagnetic signals were observed during FC cycle compared to the ZFC. These results suggest that the valence state of the Yb ion have a significant effect on the magnetic state of the Sr₂Yb₁Ru_0.9Cu_0.1O₆ superconductor.