changeset 107:7a4d101fd452

Modified abstract
author Nobuyasu Oshiro Tue, 18 Feb 2014 22:54:11 +0900 ac4ffd6c59de 2b88e59f8ae6 paper/abstract_eng.tex paper/gnuplot/distributed/distributed_read_bench.pdf paper/gnuplot/distributed/distributed_write_bench.pdf paper/master_paper.pdf 4 files changed, 11 insertions(+), 16 deletions(-) [+]
line wrap: on
line diff
--- a/paper/abstract_eng.tex	Tue Feb 18 17:52:54 2014 +0900
+++ b/paper/abstract_eng.tex	Tue Feb 18 22:54:11 2014 +0900
@@ -1,23 +1,18 @@
\begin{abstract_eng}
-Smartphone and tablet pc are widely used, thereby users of the Web services is increasing.
-It has caused the webserver is down.
-Therefore, scalability is important software factor today.
+Users of the Wev services is increasing by the diffusion of smartphone and tablet pc.
+However it caused the webserver down.
+Therefore, scalability is the important software factor for Web services today.
Scalability in distributed system is able to increase performance linearly when just added new node to system.
-In order to make provide scalability to Web services, database must have scalability.
+In order to provide scalability to Web services, database must have scalability.

-We are designing and developing a database Jungle for study of scalable database.
-It is use non-destructive tree structure.
-Non-destructive tree structure is not the destruction of data.
-Editing of data is done creating by new tree.
+For study of scalable database, we are designing and developing a database Jungle which has non-destructive tree structure.
+Non-destructive tree structure dose not destruct the data when data editing by creating new tree.

-Jungle was designed as a distributed database.
-But data distribution and persistent has not yet been implemented in the Jungle.
-
-In this paper, we develope persistent and distributed database on jungle.
-Distributed data on Jungle is developing using parallel distributed framework Alice.
-We implemented distributed on Jungle and confirmed with our cluster system that data is distributed between the server node.
+In this paper, we implemented persistent and distributed database on jungle.
+Distributed data on Jungle is developed by using Alice which is parallel distributed framework.
+We confirm the data distribution between the server nodes on Jungle with our cluster system.
Also, we developed simple bulletinboard system with Jungle and key-value store database Cassandra.
-We compared Jungle and Cassandra using simple bulletinboard.
-As a result, we got better performance then Cassandra.
+We compared response time of Jungle and Cassandra using simple bulletinboard.
+As a result, Jungle got better performance than Cassandra.

\end{abstract_eng}
Binary file paper/gnuplot/distributed/distributed_read_bench.pdf has changed
Binary file paper/gnuplot/distributed/distributed_write_bench.pdf has changed
Binary file paper/master_paper.pdf has changed