Submitted manuscript, 555 KB, PDF document
Available under license: None
Final published version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
}
TY - GEN
T1 - MANTICORE
T2 - a framework for partitioning software services for hybrid cloud
AU - Kaviani, Nima
AU - Wohlstadter, Eric
AU - Lea, Rodger
PY - 2012/12/6
Y1 - 2012/12/6
N2 - Hybrid cloud deployment can be an attractive option for companies wanting to deploy software services on scalable public clouds, while still assuming local control over sensitive data resources. A hybrid deployment, despite providing better control, is difficult to design since code must be partitioned and distributed efficiently between public and private premises. This paper describes our research into automated partitioning of software services for hybrid clouds. We have identified two specific shortfalls of existing partitioning research which are important to a hybrid cloud setting: (i) inflexibility in placement of software function execution between public/private hosts and (ii) no support for making explicit tradeoffs between monetary cost and performance. We propose a new software profiling and partitioning framework (called MANTICORE) which addresses these problems. Experiments on an open-source Web application show that the new approach ensures better performance without increasing costs.
AB - Hybrid cloud deployment can be an attractive option for companies wanting to deploy software services on scalable public clouds, while still assuming local control over sensitive data resources. A hybrid deployment, despite providing better control, is difficult to design since code must be partitioned and distributed efficiently between public and private premises. This paper describes our research into automated partitioning of software services for hybrid clouds. We have identified two specific shortfalls of existing partitioning research which are important to a hybrid cloud setting: (i) inflexibility in placement of software function execution between public/private hosts and (ii) no support for making explicit tradeoffs between monetary cost and performance. We propose a new software profiling and partitioning framework (called MANTICORE) which addresses these problems. Experiments on an open-source Web application show that the new approach ensures better performance without increasing costs.
U2 - 10.1109/CloudCom.2012.6427541
DO - 10.1109/CloudCom.2012.6427541
M3 - Conference contribution/Paper
SN - 9781467345118
SP - 333
EP - 340
BT - Cloud Computing Technology and Science (CloudCom),
PB - IEEE
ER -