DocumentCode
627435
Title
Morpheus: Learning configurations by example
Author
Jeswani, Deepak ; Balani, Rahul ; Verma, A. ; Bhattacharya, Kankar
Author_Institution
IBM Res., New Delhi, India
fYear
2013
fDate
27-31 May 2013
Firstpage
76
Lastpage
84
Abstract
Cloud computing presents a model to automatically deploy workloads using standard templates. However, reconfiguring the application to work in the new environment is manual and time-consuming. In this work, we present Morpheus to automatically configure new instances of a workload by leveraging a pre-configured instance. Morpheus uses a configured instance of a workload, annotated with the configuration information, and automatically configures new instances of the workload. It is important to note that it can configure application instances that may be installed differently or may be running a different version of the application, middleware or operating system. Morpheus first maps configuration files from the template instance to configuration files in each new instance. In the second phase, configuration nodes in source file are mapped to the target file and configuration information is automatically added in the target instance. The mapping is performed using text and structure analysis in the configuration files. We used Morpheus to automatically deploy new workloads for a wide variety of multi-tier enterprise/web applications. It was able to deploy them without any manual intervention in 13-68 minutes for 100% of the workloads.
Keywords
cloud computing; file organisation; learning (artificial intelligence); middleware; cloud computing; configuration information; file configuration; learning configurations; middleware; morpheus; multitier enterprise-Web applications; operating system; preconfigured instance; source file; standard templates; template instance; Manuals; Middleware; Operating systems; Standards; Vegetation; Virtual machining; XML;
fLanguage
English
Publisher
ieee
Conference_Titel
Integrated Network Management (IM 2013), 2013 IFIP/IEEE International Symposium on
Conference_Location
Ghent
Print_ISBN
978-1-4673-5229-1
Type
conf
Filename
6572972
Link To Document