http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
An experimental analysis of limitations of MapReduce for iterative algorithms on Spark
Kang, Minseo,Lee, Jae-Gil Baltzer Science Publishers 2017 CLUSTER COMPUTING Vol.20 No.4
<P>MapReduce is the most popular framework for distributed processing. Recently, the scalability of data mining and machine learning algorithms has significantly improved with help from MapReduce. However, MapReduce does not handle iterative algorithms very efficiently. The problem is that many data mining and machine learning algorithms are iterative by nature. In order to overcome the limitations of MapReduce, many advanced distributed systems have been developed, including HaLoop, iMapReduce, Twister, and Spark. In this paper, we identify and categorize the limitations of MapReduce in handling iterative algorithms, and then, experimentally investigate the consequences of these limitations by using the most flexible and stable distributed system, Spark. According to our experiment results, the network I/O overhead was the primary factor that affected system performance the most. The disk I/O overhead also affected system performance, but it was less significant than the network I/O overhead. For the synchronization overhead, it affected system performance only when the static data was not cached.</P>
Baltzer Science Publishers 2017 WIRELESS NETWORKS Vol.23 No.4
<P>Even though some velocity based routing protocols for (m,k)-firm stream have been recently proposed in multimedia wireless sensor networks, there are still many perspective parameters to be considered for forwarding procedure. Moreover, since they are not correlated with each other, multi-criteria system for forwarding is desirable to select next hop. However, current existing protocols apply these parameters sequentially without any prioritization. To address this issue, in this paper, we propose two (m,k)-firm specific routing protocols based on fuzzy interference system and analytical hierarchical process in conjunction with the gray relational analysis. In each protocol, delivery ratio, energy, speed, (m,k)-firm stream requirement as well as current stream status are used to select the best appropriate next hop while considering given node's constraints. Through the simulation results, we demonstrate that the proposed scheme gracefully maintains (m,k)-firm requirement while extending the network lifetime. Finally, superiority of the proposed approach to existing velocity based routing protocols is also proven through diverse simulation scenarios.</P>
Jiang, J.,Rajendiran, S.,Piao, L.,Yoon, S. Baltzer Science Publishers 2017 Topics in catalysis Vol.60 No.9
<P>Poly (3-hydroxybutyrate) (PHB) is a naturally occurring biodegradable and biocompatible polyester and considered as a class of potentially thermoplastic biopolymers in industry. The alternating copolymerization of cheap and available propylene oxide and carbon monoxide may be an efficient synthetic route for PHB. Here, we studied the base effects on the synthesis of PHB using a well-defined [(salph)Cr(THF)(2)](+)[Co(CO)(4)](-) catalyst combined with various bases. It was found that base not only affects the yield of PHB, but also the molar ratio of each unit in the product.</P>
Joint resource allocation, routing and CAC for uplink OFDMA networks with cooperative relaying
Baltzer Science Publishers 2016 WIRELESS NETWORKS Vol.22 No.5
<P>In this paper, we propose a joint resource allocation, routing, and connection admission control (CAC) scheme for uplink transmission in orthogonal frequency division multiple access (OFDMA) relay networks with cooperative relaying. For cooperative relaying, relay station can relay uplink data from mobile station (MS) to base station with cooperation of the MS using transmit diversity. Transmit diversity can be achieved by virtual MISO via distributed space-time coding. The proposed scheme jointly allocates OFDMA resources and selects path for each user with CAC to maximize the upink throughput of cooperative OFDMA relay networks. The basic OFDMA resource unit is considered as a resource element which is one subcarrier over one OFDMA symbol. An efficient multi-choice multi-dimensional knapsack (MMKP) algorithm is presented for the proposed scheme. The proposed MMKP algorithm provides a unified framework which is applicable to OFDMA networks with and without cooperative relaying. We evaluate the performance of the proposed scheme with and without cooperative relaying in a hilly terrain with heavy tree density by using OPNET-based simulation. We show that the cooperative relaying improve the uplink system throughput compared with non-cooperative relaying, and the proposed scheme outperforms the conventional link quality-based scheme in both cooperative and non-cooperative relay networks.</P>
CO and CO2 Methanation Over Supported Cobalt Catalysts
Le, T. A.,Kim, M. S.,Lee, S. H.,Park, E. D. Baltzer Science Publishers 2017 Topics in catalysis Vol.60 No.9
<P>CO and CO2 methanation was investigated over cobalt catalysts supported on different supports such as gamma-Al2O3, SiO2, TiO2, CeO2, and ZrO2. Among them, the cobalt catalyst supported on the high-surface-area CeO2 was determined to be the most active for both CO and CO2 methanation. These catalytic activities increased with increasing surface area of CeO2. To increase the specific catalytic activity for CO and CO2 methanation, various Co-CeO2 catalysts with different cobalt contents were prepared with co-precipitation method. The optimum cobalt content was determined for both reactions. The prepared catalysts were characterized with N-2 physisorption, temperature-programmed reduction with H-2, pulsed-chemisorption of CO2, temperature-programmed desorption of CO2, and X-ray diffraction. The high cobalt dispersion and strong CO2 adsorption appeared to be responsible for the high catalytic activity for CO and CO2 methanation, respectively. This Co-CeO2 also showed the stable catalytic activity even after an exposure to high-temperature reaction conditions.</P>
IRIS-HiSA: Highly Scalable and Available Carrier-Grade SDN Controller Cluster
Shin, Jisoo,Kim, Taehong,Lee, Byungjoon,Yang, Sunhee BALTZER SCIENCE PUBLISHERS 2017 MOBILE NETWORKS AND APPLICATIONS Vol.22 No.5
<P>As software defined networking (SDN) extends its applications to carrier-grade networks, the need for high scalability and availability of a SDN controller is becoming increasingly important. Although existing works have shown the feasibility of a distributed controller, the switches in the data plane are required to know some of the internal specifics such as the IP addresses of the individual controller instances. This constraint increases the operational complexity as the number of controller instances increases. In this paper, we propose a distributed controller cluster architecture called IRIS-HiSA. The main goal is to support seamless load balancing and failover with horizontal scalability, as is done in existing works, but one of the distinguishing features of IRIS-HiSA is to provide transparency to the switches in the data plane. Thus, the switches do not need to know the internal details of the controller cluster, and they simply access it in the same way a single controller is accessed. In addition to proving seamless load balancing and a failover, a performance evaluation is conducted to analyze the high scalability in which the throughput of the flow setup is proportionally increased with the number of controller instances.</P>