Database Reference
In-Depth Information
Data Transfer Bottlenecks
: Cloud users and cloud providers have to think about
the implications of placement and traffic at every level of the system if they want
to minimize costs.
Application Parallelization
: Computing power is elastic but only if workload is
parallelizable. Getting additional computational resources is not as simple as
just upgrading to a bigger and more powerful machine on the fly. However,
the additional resources are typically obtained by allocating additional server
instances to a task.
Performance Unpredictability
: Many HPC applications need to ensure that all
the threads of a program are running simultaneously. However, today's virtual
machines and operating systems do not provide this service.
Throughout the topic, we will dive into detail with respect to the requirements,
characteristics an challenges of deploying data-intensive applications in cloud com-
puting platforms. Chapter
2
provides an overview of cloud computing technology
and also discusses the state-of-the-art of a few public cloud platforms. Chapter
3
provides an overview of cloud-hosted data storage systems. It starts with concepts,
challenges, and trade-offs of cloud databases in general, and ends with a broad
survey of the state-of-the-art of public cloud databases in three categorizations. Part
two also pays extra attentions on the NoSQL movement and the stat-of-the-art of
NoSQL database systems.
Chapter
4
addresses the performance evaluation problem on cloud platforms.
There have been a number of research efforts that specifically evaluated the
Amazon cloud platform. However, there has been little in-depth evaluation research
conducted on other cloud platforms, such as Google App Engine and Microsoft
Windows Azure. But more importantly, these work lack a more generic evaluation
method that enables a fair comparison between the various cloud platforms.
Motivated by this, in this topic, a novel approach called CARE, Cloud Architecture
Runtime Evaluation, is developed to perform four test set methods with different
load stresses against cloud hosting servers or cloud databases from the perspective
of the end-user or the cloud host. The framework is capable to address performance,
availability, and reliability characteristics of various cloud platforms. The overall
data analysis of faults and errors based on intensive collected data, for deducing
architecture internal insights, is also another contribution.
Chapter
5
investigates the replication evaluation on NoSQL database as a service.
NoSQL database as a service is part of the database as a service offering to comple-
ment traditional database systems by rejecting of general ACID transactions as one
common feature. NoSQL database as a service has been supported by many service
providers that offer various consistency options, from eventual consistency to single-
entity ACID. With different consistency options, the correlated performance gains
are unclear to many customers. Therefore, in this topic, a simple benchmark is
proposed for evaluating replication delay of NoSQL database as a service from the
customers' perspective. The detailed measurements over several NoSQL database
as a services offerings show how frequently, and in what circumstances, different
inconsistency situations are observed, and to what impact the customers sees
