Improvement of Tenant Placement through In-memory Database

  • Arpita Shah, Amrinbanu Shaikh

Abstract

Cloud computing is an innovation which provides enterprise-grade computing resources as services to clients through web. One of the famously accessible services is Database-as-a-Service (DBaaS) and Software-as-a-Service (SaaS). A large number of the DBaaS/SaaS suppliers make utilization of a multitenant model to host their applications. “In-Memory Database” (IMDB) technology is a standout amongst the most dynamic information administration programming software categories in recent past. Databases have utilized in-memory information systems for many years to avoid the incredibly sluggish overall performance of analyzing and writing immediately to disk drives. As an end result, many startup databases or caching merchandise today claim to be a part of the “in-memory” category. In particular, we proposed the placement of tenant towards the in-memory database and highlight few decisions to sustenance multi-tenancy approach over in-memory database for on-demand application. Our examinations uncover that the proposed approach brings down working costs for the database application by a 29% of total operational costs; which estimated in Microsoft Azure hourly rates. We consider the issue of tenant placement for profit maximization Service level agreements (SLA). We proposed Multi-tenant placement algorithm (MTP), in which the proposed algorithm saturate only one server with tenants having SLA penalty. This may lead to operate other server normally. MTP serves proper utilization of the server resources while best-fit and next-fit approaches will not respond if numbers of tenant are increase. In MTP, we include harmonize replica indexes to maintain replica on specific sever. Scalability and efficiency are measurable criteria like CPU usage, running time etc. to quantify the tenant placement; these criteria are satisfied in our proposed algorithm.

Published
2020-06-06
How to Cite
Arpita Shah, Amrinbanu Shaikh. (2020). Improvement of Tenant Placement through In-memory Database. International Journal of Advanced Science and Technology, 29(05), 13575 - 13586. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/26777