Coda: A Highly Available File System for a Distributed Workstation Environment
IEEE Transactions on Computers
Locality-aware request distribution in cluster-based network servers
Proceedings of the eighth international conference on Architectural support for programming languages and operating systems
Group communication specifications: a comprehensive study
ACM Computing Surveys (CSUR)
Performance study of dispatching algorithms in multi-tier web architectures
ACM SIGMETRICS Performance Evaluation Review
Recovery Guarantees for General Multi-Tier Applications
ICDE '02 Proceedings of the 18th International Conference on Data Engineering
Fault-tolerance for Stateful Application Servers in the Presence of Advanced Transactions Patterns
SRDS '05 Proceedings of the 24th IEEE Symposium on Reliable Distributed Systems
Chain replication for supporting high throughput and availability
OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
Proceedings of the ACM/IFIP/USENIX 2003 International Conference on Middleware
Integrating fault tolerance and load balancing in distributed systems based on CORBA
EDCC'05 Proceedings of the 5th European conference on Dependable Computing
The design of the TAO real-time object request broker
Computer Communications
Dynamically scaling applications in the cloud
ACM SIGCOMM Computer Communication Review
A study on scalability of services and privacy issues in cloud computing
ICDCIT'12 Proceedings of the 8th international conference on Distributed Computing and Internet Technology
Escape capsule: explicit state is robust and scalable
HotOS'13 Proceedings of the 14th USENIX conference on Hot Topics in Operating Systems
Hi-index | 0.00 |
This paper proposes a novel replication architecture for stateful application servers that offers an integrated solution for fault-tolerance and load-distribution. Each application server replica is able to execute client requests and at the same time serves as backup for other replicas. We propose an effective load balancing mechanism that is only load-aware if a server is close to become overloaded. Furthermore, we present transparent reconfiguration algorithms that guarantee that each replica has the same number of backups in a dynamic environment where replicas can join or leave at any time. Our evaluation shows that our approach scales and distributes load across all servers even in heterogeneous environments while keeping the overhead for fault-tolerance and load-balancing small.