The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The disk is always the bottleneck of storage system for decades. The emergency of non-volatile memory, such as Phase Change Memory, Spin-Transfer Torque and Memristors provides an opportunity to improve the performance of storage system due to its comparable access speed like DRAM. However, existing storage systems are designed and optimized for disk, directly replacing disk with the non-volatile...
To meet the increasingly demand of massive storage system, we have designed and implemented a parallel file system called Cappella. It consists of Client, Object Storage Server (OSS) and Metadata Server (MDS). Its I/O performance and data reliability mostly rely on the OSS. In this paper, we present the design and implementation of the OSS. By comparing with existing OSS implementation, our OSS can...
Both physical disk failure and logical errors such as software error, user abuse and virus attacks may cause data lose. The risk of logical errors is far greater than physical disk failure. Moreover, existing RAID solution cannot satisfy the reliability requirement in face of the logical errors in data centers. It is therefore becoming increasingly important for RAID-based storage systems to be able...
This paper presents a scalable and adaptive decentralized metadata lookup scheme for ultralarge-scale file systems (more than Petabytes or even Exabytes). Our scheme logically organizes metadata servers (MDSs) into a multilayered query hierarchy and exploits grouped Bloom filters to efficiently route metadata requests to desired MDSs through the hierarchy. This metadata lookup scheme can be executed...
With data amount growing at an ever increasing rate, file system, as the basis of data management, its functionality and content are becoming more and more complex. Thus the file system is forced to be upgraded constantly, in order to satisfy the growing demands. However, the reinvention of new file systems is always a high-cost task. Although there are frameworks such as User-space File System that...
Fast non-volatile storage, such as non-volatile RAM and battery-backed volatile RAM, is widely deployed as the write cache in the storage controller to hide the latency of writes and prevent data loss. Due to the high cost of nonvolatile RAM and the short life of battery, the write cache is always much smaller than the read cache. To extend the size of write cache, we propose a Compressed Write Cache...
This paper presents a scalable and adaptive decentralized metadata lookup scheme for ultra large-scale file systems (ges Petabytes or even Exabytes). Our scheme logically organizes metadata servers (MDS) into a multi-layered query hierarchy and exploits grouped bloom filters to efficiently route metadata requests to desired MDS through the hierarchy. This metadata lookup scheme can be executed at...
In this paper, we implement the incorporation of a popularity-based multi-threaded reconstruction optimization algorithm, PRO, into the recovery mechanism of the Linux software RAID (MD), which is a well-known and widely-used availability-oriented disk array scheme. To evaluate the impact of PRO on RAID- structured storage systems such as MD, we conduct extensive trace-driven experiments. Our results...
Object-based storage controller (OSC) plays a decisive role in the performance of the whole object-based storage systems (OBSS). A new OSC based on switch fabric proposed in this paper. The new architecture aims to improve the OSC's disk I/O performance. Through parallel data transfer using two independent high-speed PCI-X buses attached to the switch fabric, the possibility that a single PCI bus...
Ethernet-based network storage protocols such as iSCSI have become increasingly common in today's storage network. In this paper, we propose and implement another new protocol vSCSI (Vl-attached SCSI) for competing with iSCSI in LAN environment, and experimentally compare performance ofvSCSI and iSCSI for environments with no data sharing across machines. Using VI replaces TCP/IP, vSCSI achieves higher...
Trace-based analysis has always been used to guide and motivate more effective hardware and software design. Existing traces available, however, are unsuitable for studying object-based storage system, a very active area of current storage research, because they are collected from the traditional file systems and don't contain information about object. Thus, a kind of distributed object tracing system...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.