Subtleties in tolerating correlated failures in wide-area storage systems
NSDI'06 Proceedings of the 3rd conference on Networked Systems Design & Implementation - Volume 3
Understanding disk failure rates: What does an MTTF of 1,000,000 hours mean to you?
ACM Transactions on Storage (TOS)
Dynamo: amazon's highly available key-value store
Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
The cost of a cloud: research problems in data center networks
ACM SIGCOMM Computer Communication Review
DRAM errors in the wild: a large-scale field study
Proceedings of the eleventh international joint conference on Measurement and modeling of computer systems
Characterizing cloud computing hardware reliability
Proceedings of the 1st ACM symposium on Cloud computing
Understanding network failures in data centers: measurement, analysis, and implications
Proceedings of the ACM SIGCOMM 2011 conference
Towards realizing a low cost and highly available datacenter power infrastructure
HotPower '11 Proceedings of the 4th Workshop on Power-Aware Computing and Systems
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
Don't settle for eventual: scalable causal consistency for wide-area storage with COPS
SOSP '11 Proceedings of the Twenty-Third ACM Symposium on Operating Systems Principles
Spanner: Google's globally-distributed database
OSDI'12 Proceedings of the 10th USENIX conference on Operating Systems Design and Implementation
Underprovisioning backup power infrastructure for datacenters
Proceedings of the 19th international conference on Architectural support for programming languages and operating systems
Hi-index | 0.00 |
Cloud providers and other data center operators are using geo-distributed data centers. But these data centers largely continue to employ the same designs as were appropriate for single data centers. These designs are wasteful because they do not take full advantage of geo-distribution. Geo-redundancy can reduce other redundancy at multiple intermediate layers in individual data centers and decrease costs. We discuss options for changing infrastructure design to realize such savings. Our proposal opens up an exciting and novel area of investigation into the design of software that can effectively leverage such platforms.