Abstract:

Hadoop is a framework for processing large amount of data in parallel. It uses the Hadoop Distributed File System (HDFS) and MapReduce framework for processing. Job scheduling is an important process in Hadoop MapReduce. There are three types of schedulers mainly used in Hadoop Framework. They are namely FIFO, Fair and Capacity Scheduler. In Hadoop Framework the schedulers are a pluggable component now a day. There are jobs which have dependency on an external service like database or web service. This condition will lead to the failure of tasks due to overloading. In this case, Hadoop needs to re-run the entire tasks in another slots. To address this issue, the TaskTracker aware scheduling has been introduced. The TaskTracker scheduler enables users to configure a maximum load per TaskTracker in the Job Configuration itself. If the load of the TaskTracker reaches its maximum limit then the algorithm will not allow the task to run and stops the task from failing. Also, the TaskTracker aware scheduler allows the users to select the TaskTracker per Job which is configured in the Job configuration.

Keywords:

Hadoop, HDFS, BigData, MapReduce, JobTracker, TaskTracker, Scheduler

Abstract:

Social network has become a very popular way for internet users to communicate and interact in online. Users spend plenty of time on famous social networks (e.g., Facebook, Twitter, SinaWeibo, etc.), for reading news, discussing events and posting messages. The main aim of this work is to find out the spammers and non-spammers (good and bad friends or followers). For this work more than 1000 messages collected from different friends. Then a labeled dataset of users are constructed and applied into SVM (Support Vector Machines) based spammer detection algorithm. This algorithm gives the true positive rate of spammers and non-spammers reaching 99.1% and 99.9% respectively.

Keywords:

Support Vector Machines, Application Program Interface, Radial Basis Function, spammer, non-spammer

Abstract:

With the ever-increasing demands on multimedia applications, cloud computing, due to its economical but powerful resources, is becoming a natural platform to process, store, and share multimedia contents. However, the employment of cloud computing also brings new security and privacy issues as few public cloud servers can be fully trusted by users. In this paper, they focus on how to securely share video contents to a certain group of people during a particular time period in cloud-based multimedia systems, and propose a cryptographic approach, a provably secure time-domain attribute-based access control (TAAC) scheme, to secure the cloud-based video content sharing. Specifically, they first propose a provably secure time-domain attribute-based encryption scheme by embedding the time into both the cipher texts and the keys, such that only users who hold sufficient attributes in a specific time slot can decrypt the video contents. They also propose an efficient attribute updating method to achieve the dynamic change of users' attributes, including granting new attributes, revoking previous attributes, and regranting previously revoked attributes. They further discuss on how to control those video contents that can be commonly accessed in multiple time slots and how to make special queries on video contents generated in previous time slots. The security analysis and performance evaluation show that TAAC is provably secure in generic group model and efficient in practice.

Keywords:

Cloud Computing, Security, Access Control, Cryptographic

Abstract:

This paper, we present a new in networked system for efficient encrypted video delivery while preserving the benefits of in-network caching. As video chunks are encrypted before distribution, we first design a compact, efficient, yet encrypted video user index to empower the network with a fully controlled capability of locating the cached encrypted chunks for given encrypted requests. We then explain how to deploy the encrypted design in our proposed architecture, and present a secure redundancy elimination protocol to enable fast video delivery via leveraging cached encrypted chunks. We further discuss the full support of cache management, adaptive video delivery, and video access control. Rigorous analysis and prototype evaluations demonstrate the security, efficiency, and effectiveness of the design.

Keywords:

Video Delivery, Encryption, MD5 Public Key Algorithm