Dependence Management in Component-Based Distributed Systems
IEEE Concurrency
How do APIs evolve? A story of refactoring: Research Articles
Journal of Software Maintenance and Evolution: Research and Practice - IEEE International Conference on Software Maintenance (ICSM2005)
Reducing downtime due to system maintenance and upgrades
LISA '05 Proceedings of the 19th conference on Large Installation System Administration Conference - Volume 19
Collateral evolution of applications and databases
Proceedings of the joint international and annual ERCIM workshops on Principles of software evolution (IWPSE) and software evolution (Evol) workshops
Proceedings of the 10th ACM/IFIP/USENIX International Conference on Middleware
Middleware'09 Proceedings of the ACM/IFIP/USENIX 10th international conference on Middleware
Safe and automatic live update for operating systems
Proceedings of the eighteenth international conference on Architectural support for programming languages and operating systems
Disnix: A toolset for distributed deployment
Science of Computer Programming
Hi-index | 0.00 |
Traditional approaches for online upgrades of distributed systems rely on dependency tracking to preserve system integrity during and after the upgrade. Because dependency reification can become intractable, we aim to enforce the isolation of the old and new versions during the upgrade. We achieve this by installing the new version in a "parallel universe" - a separate physical or virtual infrastructure that does not communicate directly with the old version. This allows our upgrading middleware to treat the complex IT infrastructure as a black box with unknown hidden-dependencies, and to validate the upgrade by cross-checking the outputs of the two universes.