ACM SIGOPS Operating Systems Review
Unreliable failure detectors for reliable distributed systems
Journal of the ACM (JACM)
The Totem multiple-ring ordering and topology maintenance protocol
ACM Transactions on Computer Systems (TOCS)
Database Management Systems
Reliable Multicast between Micro-Kernels
Proceedings of the Workshop on Micro-kernels and Other Kernel Architectures
SRDS '96 Proceedings of the 15th Symposium on Reliable Distributed Systems
ACM Computing Surveys (CSUR)
Scalability of the microsoft cluster service
WINSYM'98 Proceedings of the 2nd conference on USENIX Windows NT Symposium - Volume 2
Bigtable: a distributed storage system for structured data
OSDI '06 Proceedings of the 7th USENIX Symposium on Operating Systems Design and Implementation - Volume 7
Paxos made live: an engineering perspective
Proceedings of the twenty-sixth annual ACM symposium on Principles of distributed computing
Dynamo: amazon's highly available key-value store
Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
Queue - Object-Relational Mapping
Communications of the ACM - Rural engineering development
PNUTS: Yahoo!'s hosted data serving platform
Proceedings of the VLDB Endowment
Queue - Scalable Web Services
Hadoop: The Definitive Guide
Cassandra: a decentralized structured storage system
ACM SIGOPS Operating Systems Review
Benchmarking cloud serving systems with YCSB
Proceedings of the 1st ACM symposium on Cloud computing
Scalable consistency in Scatter
SOSP '11 Proceedings of the Twenty-Third ACM Symposium on Operating Systems Principles
Windows Azure Storage: a highly available cloud storage service with strong consistency
SOSP '11 Proceedings of the Twenty-Third ACM Symposium on Operating Systems Principles
Zab: High-performance broadcast for primary-backup systems
DSN '11 Proceedings of the 2011 IEEE/IFIP 41st International Conference on Dependable Systems&Networks
Harmony: Towards Automated Self-Adaptive Consistency in Cloud Storage
CLUSTER '12 Proceedings of the 2012 IEEE International Conference on Cluster Computing
| Hi-index | 0.00 |
The dramatic increase of data stored, processed and reused in all sectors leads to the evolution of modern storage systems known as NoSQL databases such as Amazon Dynamo, Cassandra, Big-Table, PNUTS, HBase. In order to boost the availability and performance of the system, these storage systems follow eventual consistency and don't offer tight consistency by default. Paxos is commonly used in this context to ensure tight consistency on demand. But it adds extra costs on messages management, mostly complexity and size. In addition, Paxos shrinks the space for other research areas such as cache memory optimization and load-balancing. This paper gives an opportunity to propose and discuss the challenges of a new consistency protocol for modern storage systems entitled 'LibRe'. LibRe follows the Eventual Consistency model. In addition, it logs operations executed on each node in the distributed system. This additional information is used by the load balancer and ensures that requests are not forwarded to a node where the data needed to serve the request are stale. Since Eventual Consistency already offers better Availability and Partition tolerance, the aspiration of associating LibRe with eventual consistency is to work out a better consistency management service providing also availability and partition tolerance. The simulation results for consistency and latency in LibRe are compared among traditional Pessimistic Consistency, Eventual Consistency, and Paxos. The overall results are discussed and new opportunities for research works are provided.