Journal of Network Communications and Emerging Technologies (JNCET) www.jncet.org Volume 7, Issue 3, March (2017) ISSN: 2395-5317 ©EverScience Publications 41 Survey on the Optimal Placement of Secure Data Objects over Internet Jayshri Singh Computer Science and Engineering Deptt., CSIT-Durg, India. Keshav Kori Computer Science and Engineering Deptt., CSIT-Durg, India. Abstract – Outsourcing information to an outsider authoritative control, as is done in distributed computing, offers ascend to security concerns. The information trade off may happen because of assaults by different clients and hubs inside of the cloud. Hence, high efforts to establish safety are required to secure information inside of the cloud. On the other hand, the utilized security procedure should likewise consider the advancement of the information recovery time. In this paper, we propose Division and Replication of Data in the Cloud for Optimal Performance and Security (DROPS) that all in all methodologies the security and execution issues. In the DROPS procedure, we partition a record into sections, and reproduce the divided information over the cloud hubs. Each of the hubs stores just a itary part of a specific information record that guarantees that even in the event of a fruitful assault, no important data is uncovered to the assailant. Additionally, the hubs putting away the sections are isolated with certain separation by method for diagram T-shading to restrict an assailant of speculating the areas of the sections. Moreover, the DROPS procedure does not depend on the customary cryptographic procedures for the information security; in this way alleviating the arrangement of computationally costly approaches. We demonstrate that the likelihood to find and bargain the greater part of the hubs putting away the sections of a solitary record is to a great degree low. We likewise analyze the execution of the DROPS system with ten different plans. The more elevated amount of security with slight execution overhead was watched. Index Terms – Centrality, cloud security, fragmentation, replication, performance. I. INTRODUCTION The distributed computing worldview has changed the use and administration of the data innovation foundation. Distributed computing is portrayed by on-interest self- administrations; universal system gets to, asset pooling, flexibility, and measured administrations. The previously stated qualities of distributed computing make it a striking possibility for organizations, associations, and individual clients for reception. In any case, the advantages of minimal effort, insignificant administration (from a clients point of view), and more prominent adaptability accompany expanded security concerns. Security is a standout amongst the most pivotal perspectives among those forbidding the far reaching reception of distributed computing. Cloud security issues might stem because of the centre technology execution (virtual machine (VM) escape, session riding, and so forth.), cloud administration offerings (organized question dialect infusion, feeble confirmation plans, and so on.), and emerging from cloud attributes (information recuperation weakness, Internet convention powerlessness, and so on.) For a cloud to be secure, the majority of the taking part substances must be secure. In any given framework with numerous units, the largest amount of the systems security is equivalent to the security level of the weakest element. In this way, in a cloud, the security of the benefits does not exclusively rely on upon an individual's efforts to establish safety. The neighbouring elements might give a chance to an assailant to sidestep the client’s protections. The off-site information stockpiling cloud utility obliges clients to move information in cloud's virtualized and shared environment that may bring about different security concerns. Pooling and flexibility of a cloud, permits the physical assets to be shared among numerous clients. Also, the common assets may be reassigned to different clients at some occurrence of time that may bring about information trade off through information recuperation systems. Moreover, a multi-occupant virtualized environment might bring about a VM to get away from the limits of virtual machine screen (VMM). They got away VM can meddle with different VMs to have entry to unapproved information. Additionally, cross-occupant virtualized system access might likewise trade off information protection and trustworthiness. Shameful media disinfection can likewise spill customers private information. The information outsourced to an open cloud must be secured. Unapproved information access by different clients and forms (whether coincidental or purposeful) must be avoided. As examined over, any frail substance can put the entire cloud at danger. In such a situation, the security system should significantly build an assailant's push to recover a sensible measure of information even after an effective interruption in the cloud. In addition, the plausible measure of misfortune (as an after effect of information spillage) should likewise be minimized.