Conclusions - Cloud Data Management - page 173

scalability are considered to be of the most important features which are provided
by NoSQL systems [ 218 ]. In practice, both of the commercial NoSQL offerings
(e.g. Amazon SimpleDB) and commercial DaaS offerings (e.g. Amazon RDS,
Microsoft SQL Azure) do not provide their users with any flexibility to dynamically
increase or decrease the allocated computing resources of their applications. While
NoSQL offerings claim to provide elastic services of their tenants, they do not
provide any guarantee that their provider-side elasticity management will provide
scalable performance with increasing workloads [ 75 ]. Moreover, commercial DaaS
pricing models require their users to pre-determine the computing capacity that
will be allocated to their database instance as they provide standard packages of
computing resources (e.g. Micro , Small , Large and Extra Large DB Instances). In
practice, predicting the workload behavior (e.g. arrival pattern, I/O behavior, service
time distribution) and consequently accurate planning of the computing resource
requirements with consideration of their monetary costs are very challenging tasks.
Therefore, the user might still tend to over-provision the allocated computing
resources for the database tier of their application in order to ensure satisfactory
performance for their workloads. As a result of this, the software application is
unable to fully utilize the elastic feature of the cloud environment.
Xiong et al. [ 234 ] have presented an provider-centric approach for intelligently
managing the computing resources in a shared multi-tenant database system at the
virtual machine level. The proposed approach consists of two main components:
1. The system modeling module that uses machine learning techniques to learn
a model that describes the potential profit margins for each tenant under
different resource allocations. The learned model considers many factors of the
environment such as SLA cost, client workload, infrastructure cost and action
cost.
2. The resource allocation decision module dynamically adjusts the resource allo-
cations, based on the information of the learned model, of the different tenants in
order to achieve the optimum profits.
Tatemura et al. [ 220 ] proposed a declarative approach for achieving elastic
OLTP workloads. The approach is based on defining the following two main
components:
1. The transaction classes required for the application.
2. The actual workload with references to the transaction classes.
Using this information, a formal model can be defined to analyze elasticity of the
workload with transaction classes specified. In general, we believe that there is a
lack of flexible and powerful consumer-centric elasticity mechanisms that enable
software application to have more control on allocating the computing resources for
the database tier of their applications over the application running time and make
the best use of the elasticity feature of the cloud computing environments. More
attention from the research community is required to address these issues in the
future work.