Managing energy and server resources in hosting centers
SOSP '01 Proceedings of the eighteenth ACM symposium on Operating systems principles
Hippodrome: Running Circles Around Storage Administration
FAST '02 Proceedings of the Conference on File and Storage Technologies
Traveling to Rome: QoS Specifications for Automated Storage System Management
IWQoS '01 Proceedings of the 9th International Workshop on Quality of Service
Design and evaluation of a continuous consistency model for replicated services
OSDI'00 Proceedings of the 4th conference on Symposium on Operating System Design & Implementation - Volume 4
| Hi-index | 0.00 |
Clients are coming to rely more and more on external services to meet the needs of their users, and the clients are increasingly simple caches of soft state -- "truth" is maintained elsewhere. As a result, the user experience of dependability is better served by making those services ultra-dependable than by increasing the reliability of an individual client. We explore here some of the consequences of this statement, and conclude that developing scalable, dependable services may be a more fruitful approach than an extreme emphasis on "dependable OSes". Along the way we look at quantifying "dependability" in this new world; some of what it takes to provide dependable, large-scale services; and some approaches that we are exploring to do so.