Migrating Weather Production to External IaaS
Turkki, Aapeli (2017)
Metropolia Ammattikorkeakoulu
2017
All rights reserved
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-201705117618
https://urn.fi/URN:NBN:fi:amk-201705117618
Tiivistelmä
Demands as to the accuracy of weather prediction services are growing due to that the climate chancing rapidly. With the help of various weather services it is possible to help for example governments and businesses to prepare for incoming storms, flood and other weather phenomena. In order to be able to provide such services, there is a growing need to store, transfer and process huge data sets.
This thesis investigates the best practices to deploy, configure and build a scalable post- processing environment to an external Infrastructure as a Service (Iaas). The case company, Finnish Meteorological Institute, is looking at options to use external IaaS to expand the current weather prediction environment, since the current data transfer speeds are not enough to transfer needed data sets to the case company’s premises on time. The case company is also looking at IaaS as a cost effective option to get on demand processing capacity when needed.
The study goes through well-known open source technologies used to build a post-processing environment. The theoretical part also explains the IaaS architecture and how it differs from on-premise infrastructure
The practical part explains the steps needed to build a post-processing environment to an external IaaS, as well as creating a test plan and testing phases. The study also presents what was needed to integrate a post-processing environment to part of a current weather production environment.
The outcome of this thesis is a scalable post-processing environment on an external IaaS. The study provides necessary information for what is needed to build a post-processing environment.
This thesis investigates the best practices to deploy, configure and build a scalable post- processing environment to an external Infrastructure as a Service (Iaas). The case company, Finnish Meteorological Institute, is looking at options to use external IaaS to expand the current weather prediction environment, since the current data transfer speeds are not enough to transfer needed data sets to the case company’s premises on time. The case company is also looking at IaaS as a cost effective option to get on demand processing capacity when needed.
The study goes through well-known open source technologies used to build a post-processing environment. The theoretical part also explains the IaaS architecture and how it differs from on-premise infrastructure
The practical part explains the steps needed to build a post-processing environment to an external IaaS, as well as creating a test plan and testing phases. The study also presents what was needed to integrate a post-processing environment to part of a current weather production environment.
The outcome of this thesis is a scalable post-processing environment on an external IaaS. The study provides necessary information for what is needed to build a post-processing environment.