Virtual I/O scheduler: a scheduler of schedulers for performance virtualization
Proceedings of the 3rd international conference on Virtual execution environments
pClock: an arrival curve based approach for QoS guarantees in shared storage systems
Proceedings of the 2007 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Towards fairness and efficiency in storage systems
ACM SIGMETRICS Performance Evaluation Review
Towards distributed storage resource management using flow control
ACM SIGOPS Operating Systems Review
Resource overbooking and application profiling in a shared Internet hosting platform
ACM Transactions on Internet Technology (TOIT)
PARDA: proportional allocation of resources for distributed storage access
FAST '09 Proccedings of the 7th conference on File and storage technologies
Differential virtual time (DVT): rethinking I/O service differentiation for virtual machines
Proceedings of the 1st ACM symposium on Cloud computing
mClock: handling throughput variability for hypervisor IO scheduling
OSDI'10 Proceedings of the 9th USENIX conference on Operating systems design and implementation
Maestro: quality-of-service in large disk arrays
Proceedings of the 8th ACM international conference on Autonomic computing
vIC: interrupt coalescing for virtual machine storage device IO
USENIXATC'11 Proceedings of the 2011 USENIX conference on USENIX annual technical conference
Demand based hierarchical QoS using storage resource pools
USENIX ATC'12 Proceedings of the 2012 USENIX conference on Annual Technical Conference
FAIRIO: A Throughput-oriented Algorithm for Differentiated I/O Performance
International Journal of Parallel Programming
Balancing fairness and efficiency in tiered storage systems with bottleneck-aware allocation
FAST'14 Proceedings of the 12th USENIX conference on File and Storage Technologies
Hi-index | 0.00 |
I/O consolidation is a growing trend in production environments due to the increasing complexity in tuning and managing storage systems. A consequence of this trend is the need to serve multiple users/workloads simultaneously. It is imperative to make sure that these users are insulated from each other by virtualization in order to meet any service level objective (SLO). This paper presents a 2-level scheduling framework that can be built on top of an existing storage utility. This framework uses a low-level feedback-driven request scheduler, called AVATAR, that is intended to meet the latency bounds determined by the SLO. The load imposed on AVATAR is regulated by a high-level rate controller, called SARC, to insulate the users from each other. In addition, SARC is workconserving and tries to fairly distribute any spare bandwidth in the storage system to the different users. This framework naturally decouples rate and latency allocation. Using extensive I/O traces and a detailed storage simulator, we demonstrate that this 2-level framework can simultaneously meet the latency and throughput requirements imposed by an SLO, without requiring extensive knowledge of the underlying storage system.