110 likes | 217 Vues
This presentation explores current issues in Distributed Computing Systems (DCS), focusing on the advancements and challenges associated with technologies like Hadoop, Grid Computing, and Cloud Computing. The authors, Dr. Sunny Jeong and Mr. Colin Zhang, discuss the architecture of Hadoop, including HDFS and MapReduce, along with the capabilities for handling big data and node failures. The talk also contrasts Grid and Cloud Computing, addressing concepts such as CPU scavenging and data management in distributed environments. The session highlights critical projects in the field and the barriers faced by emerging technologies.
E N D
Distributed Computing Systems Current Issues in DCS Dr. Sunny Jeong. spjeong@uic.edu.hk Mr. Colin Zhang colinzhang@uic.edu.hk With Thanks to Prof. G. Coulouris, Prof. A.S. Tanenbaum and Prof. S.C Joo
Cluster, Grid, Cloud …Now…Hadoop? • To Support hadling Big data • Hadoop HDFS + MapReduce • designed so that any node failures are automatically handled by the framework. • enables applications to work with thousands of computation-independent computers and petabytes of data. • Facebook Yahoo..of course GOOGLE
Overview • Hadoop • Computer Cluster • Grid Computing • Grid Computing Projects 3
Hadoop Architecture • Wikipedia
MapReduce Architecture • Yuan Luo
HDFS Architecture • Dhruba Borthakur
Clustering Computing • Consists of a set of loosely connected computers that work together as a single system. • The components of a cluster are usually connected to each other through fast LAN, each node with own OS • Availability of low cost microprocessors, high speed networks, and software for high performance distributed computing 7
Grid Computing • Computational Grid: • Focus on CPU scavenging/stealing technology for complicated operation • Data Grid): Managing and sharing distributed intensive data • Access Grid: For Cooperation between remote computer using audio or video applications • Equipment Grid : Controlling remote machine and analyzing data from the machine.
Grid Project • SETI@home • BOINC • holding@home • korea@home(stopped in 2011)
Cloud Computing( comparison with Grid) • Service Oriented • Transaction oriented • No standardization Organization • No Standard Technology • More less homogeneous gathering (compare to Grid) • Loosely connected 10
Barriers of the Concepts • Shimyoung cheol