Venti: A New Approach to Archival Storage
FAST '02 Proceedings of the Conference on File and Storage Technologies
A Scalable Solution to the Multi-Resource QoS Problem
RTSS '99 Proceedings of the 20th IEEE Real-Time Systems Symposium
Memory resource management in VMware ESX server
OSDI '02 Proceedings of the 5th symposium on Operating systems design and implementationCopyright restrictions prevent ACM from being able to make the PDFs for this conference available for downloading
Parallax: managing storage for a million machines
HOTOS'05 Proceedings of the 10th conference on Hot Topics in Operating Systems - Volume 10
Argon: performance insulation for shared storage servers
FAST '07 Proceedings of the 5th USENIX conference on File and Storage Technologies
Adaptive control of virtualized resources in utility computing environments
Proceedings of the 2nd ACM SIGOPS/EuroSys European Conference on Computer Systems 2007
Parallax: virtual disks for virtual machines
Proceedings of the 3rd ACM SIGOPS/EuroSys European Conference on Computer Systems 2008
Hi-index | 0.00 |
Cloud computing has become one of the hottest topics in both academia and industry. In cloud computing, virtualization is the fundamental technology, and the resource management is the key issue. Storage, an important resource, traditionally is allocated to users according to users' predictions, so may cause bad space utilization because of overprovisioning or underprovisioning. Dynamic adjusting is difficult because of the fixed address mapping used in the traditional storage systems. In this paper, we present an Allocation-On-Demand Incremental (AoDI) volume. It uses an appending rather than overwriting strategy to deal with the write requests. So we can obtain accurate storage space usage easily. Accompany with an automatic volume extension technique, AoDI can allocate storage space always matching users' real-time requirement, so avoids space waste. Based on appending strategy, we also design a Non-COW (Non Copy On Write) snapshot that is dramatically faster than traditional COW snapshots. Since snapshot is a frequent operation in cloud or virtualization systems, Non-COW snapshot can improve overall performance effectively. Another advantage of AoDI is translating random requests into sequential requests that obviously can speed up random access. We implement AoDI based on LVM (Logical Volume Manager) in Linux platform. Our experimental results show the great performance advantage of AoDI in snapshot and random access.