Currently, cloud computing is considered an important paradigm in Information Technology (IT). The key features for cloud computing capabilities can be summarized as: on-demand self-service, broad network access, resource pooling, measured service and rapid elasticity. The cloud service model uses virtual servers for managing, scheduling, communicating, networking and auto-scaling as the common layer to fulfill a cloud computing scheme [1
]. Among the proven benefits of cloud computing and applications, elasticity is the primary feature for both service providers and users. Elasticity, which refers to the extent to which resources provisioned by service providers change in relation to the changing user demand, allows service providers to maintain a high level of performance quality for application services.
Cloud-based application cases have been reviewed and we considered unresolved issues related to cloud platforms [2
]. With regard to auto-scaling schemes, related concepts and taxonomy were surveyed [3
]. A technical review of auto-scaling for elastic cloud-based applications was provided, and the Gartner group describes auto-scaling as an automatic expansion or contraction of system capacity, and indicated that such a capacity is a commonly desired feature in cloud infrastructure as a service and platform as a service offering [4
]. In other words, auto-scaling refers to the significant capability of a cloud computing environment to utilize virtualized computing resources automatically. In this scheme, virtualized resources can be increased or decreased dynamically by adapting resource utilization to satisfy the given requirements. Auto-scaling contributes to cost control. The key features of auto-scaling are the ability to scale-out, i.e., automatic addition of resources during increased demand, and scale-in, i.e., automatic termination of unused resources when demand decreases. Scale-out and scale-in schemes are referred to as horizontal scaling. Unlike horizontal scaling, vertical scaling increases computation resources in existing nodes. Auto-scaling at the service level is important because services run on a set of connected virtual machines. The optimal model-driven configuration of cloud auto-scaling infrastructure was studied [5
]. It was implemented an open-source cloud environment with auto-scaling to access resources for a flexible period with varying requirements in bioinformatics and biomedical workflows [6
], and was implemented an auto-scaling model in simulation experiments using the Amazon Elastic Compute Cloud (EC2) to reduce resource costs and test the quality of service in terms of response time and availability [7
]. Three features of auto-scaling from a technical perspective were described [8
]: covering variable demands, web application architecture, and distributing instances across availability zones. Cloud-based data-processing services in a geospatial web were pointed out as a very important issue in the area of remote sensing [9
]. Cloud computing contributes to remote sensing applications for maximizing imagery resources, disseminating on-demand insight to end users globally, and improving efficiency while reducing cost and risk [10
]. Advantages of remote sensing image processing within a cloud environment were discussed [11
]. The auto-scaling performance test using a concurrent user number and an average response time of a spatial web portal based on a cloud-enabled framework implementation by Amazon EC2, Microsoft Azure, and NASA Nebula was carried out with respect to the dynamic workload [12
]. Auto-scaling is a feature of OpenStack [13
], where Heat is a template-based service that manages the lifecycle of OpenStack applications and defines the relationships among resources. The purpose of this study is to investigate the applicability of auto-scaling schemes to geo-based image processing services through performance tests with respect to some practical remote sensing algorithms, in an OpenStack cloud computing environment.