Abstract
Multi-cloud computing is no different than other Cloud computing (CC) models when it comes to providing users with self-services IT resources. For instance, a company can use services of one specific cloud Service Provider (CSP) for its business, as it can use more than one CSP either to get the best of each without any vendor lock-in. However, the situation is different regarding monitoring a multi-cloud environment. In fact, CSPs provide in-house monitoring tools that are natively compatible with their environment but lack support for other CSP's environments, which is problematic for any company that wants to use more than a CSP. In addition, third party cloud monitoring tools often use agents installed on each monitored virtual machine (VM) to collect monitoring data and send them to a central monitoring server that is hosted on premise or on a Cloud, which increases bottlenecks and latency while transmitting data or processing it. Therefore, this paper presents a monitoring framework for multi-cloud environments that implements edge computing and RESTFul microservices for a high efficiency monitoring and scalability. In fact, the monitoring framework “EMMCS” uses SNMP agents to collect metrics, and performs all monitoring tasks at the edge of each cloud to enhance network transmission and data processing at the central monitoring server level. The implementation of the framework is tested on different public cloud environments, namely Amazon AWS and Microsoft Azure to show the efficiency of the proposed approach.
Highlights
Now-a-days, the IT world witnesses an exponential evolution since the emergence of the Cloud computing (CC) [1] [13] paradigm where hardware and software are provided as reasonably priced and payed-per-use services compared to acquisition, on premise hosting, self-deployment and maintenance by the client
This paper presented a novel approach of a monitoring framework for Multi-cloud environment where all heavy processing are done at the periphery of the cloud rather than sending raw data to be processed on the main datacenter
The framework, called EMMCS, is scalable and modular using a microservices-oriented architecture where each service is provided with its own RESTful API
Summary
Now-a-days, the IT world witnesses an exponential evolution since the emergence of the CC [1] [13] paradigm where hardware (such as CPUs, memory, storage, unlimited bandwidth, virtual network equipment etc.) and software (such as WEB and application servers, databases, frameworks etc.) are provided as reasonably priced and payed-per-use services compared to acquisition, on premise hosting, self-deployment and maintenance by the client. *-as-a-Service (i.e. Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS) and Software-as-a-Service (SaaS)) along with many features such as global high availability, resource polling, scalability, elasticity, and on-demand services delivered in an automated manner With these features, CC can offer unlimited computing resources (virtually), reasonably priced and tailored services, accessible from anywhere through a WEB browser. Clients are more frustrated by the vendor lock in practice that most of CSPs do in a way or another This has encouraged clients to spread their workload across different types of cloud providers, i.e. the multi-cloud strategy [2], rather than relying on one CSP, which gives clients more choices and benefits while making their businesses highly available and fault tolerant. The www.ijacsa.thesai.org (IJACSA) International Journal of Advanced Computer Science and Applications, Vol 10, No 1, 2019 processed data are sent back to the MMS that can run on premise or hosted on a cloud provider
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
