Tenant Oriented Lock Granularity Adjustment Strategy in the Shared Storage Multi-tenant Database
نویسندگان
چکیده
SaaS is a perfect technology to achieve scale economy. It uses multi-tenant to allow database to share storage but with irregular load, untraceable transaction characteristics and so on. TOL mechanism provides one granularity between the upper coarse-grained granularity and the lower finegrained granularity for each tenant. But the granularity may not be the most suitable. We define the cluster method and the diffusion method to support the granularity adjustment for a single tenant. Cluster is to make the granularity adjust towards coarser while diffusion gains opposite effect. Because TOL takes advantage of data distribution in multi-tenant database, we combine several aspects when the adjustment strategy is established, including basic table plus extension table and transactions feature. It focuses on locating a more suitable granularity size for one tenant between coarse-grained and fine-grained. In addition, we give experiments indicating that the lock granularity adjustment method is feasible and the adjustment strategy is effective.
منابع مشابه
A Mixed Partitioning Approach for Multi - tenant Data Schema ⋆
The traditional Database partitioning has failed in multi-tenant shared scheme database; this paper proposed a mixed Partitioning Approach. By integrating vertical and horizontal partitioning, we can adaptive expand the multi-tenant database according to the number of server nodes or we can make an optimal partitioning. Experimental results show that our approach can improve the performance of ...
متن کاملProxy Service for Multi-tenant Database Access
The database of multi-tenant Software as a Service (SaaS) applications has challenges in designing and developing a relational database for multi-tenant applications. In addition, combining relational tables and virtual relational tables to make them work together and act as one database for each single tenant is a hard and complex problem to solve. Based on our multitenant Elastic Extension Ta...
متن کاملA new meta-data driven data-sharing storage model for SaaS
A multi-tenant database is the primary characteristic of SaaS, it allows SaaS vendors to run a single instance application which supports multiple tenants on the same hardware and software infrastructure. This application should be highly customizable to meet tenants’ expectations and business requirements. This paper examined current solutions on multi-tenancy, and proposed a new meta-data dri...
متن کاملResearch on Multi-Tenant Distributed Indexing for SaaS Application
Multi-tenant is the key feature for SaaS application, however, the traditional indexing mechanism has failed in multi-tenant shared scheme database. This paper proposed a multi-tenant distributed indexing mechanism. We create a global index first and then create the local index by MapReduce framework based on Hadoop. We also proposed the process of index update and index merging. Experimental r...
متن کاملMulti-tenant Main Memory Index Tree with Shared Structure
Multi-tenant main memory index is an important tool to accelerate data access to software as a service. Establishing main memory indexes for each tenant occupies lots of memory space and results in performance bottleneck. The data schemas and access patterns of different tenants are similar, which provides the conditions for tenants storing their index entries with shared structure in main memo...
متن کامل