Characterizing & Improving the General Performance of Apache Zookeeper : Sub-project of Apache Hadoop

ISBN-13: 9783659823558 / Angielski / Miękka / 2016 / 108 str.

Chandan Bagai

Opis

Coordination has become one of the most vital requirements in a scenario where a group of systems are communicating with each other over a network trying to accomplish goals. Most of the distributed applications require sophisticated coordination primitives such as leader election. Apache ZooKeeper is used for cloud computing in order to provide essential synchronization and group services in a distributed environment. ZooKeeper not only possesses a wait free property, but also guarantees FIFO client ordering and linearizable writes. ZooKeeper achieves FIFO ordering by using a pipelined architecture allowing thousands of outstanding requests while achieving a low latency and high service rate. ZooKeeper provides a client library through which clients connect and submit their requests to the server. The server on the other hand intercepts the request, processes it, and responds back to the client. The read requests are serviced from the local database which have a replica of the ZooKeeper state. Write requests are transformed from ZooKeeper request. This is where we can walk an extra mile in order to boost performance of our application.

O autorze

Chandan Bagai has obtained his Masters in Computer Science from University of New Brunswick. He has keen interest & hands on experience in multiple domains of computer science such as Software Development & Security. He holds global certifications such as GWAPT & C EH. He believes that Knowledge & Wisdom should go hand in hand for achieving success