G. Gibson, D. Nagle, K. Amiri, F. Chang, H. Gobioff, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for network-attached secure disks. Technical Report CMU--CS--97--112, Carnegie Mellon, March 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....by remote functions on Active Disks. Examples include a read modify write operation, or an atomic create that both creates a new file object and updates the corresponding directory object. Such advanced operations are already used for optimization of higher level file systems such as NFS on NASD [23]. The existence of strong shared memory objects does not ob Section 5.3 actually provides a stronger result, proving impossibility of constructing a different type of object than a consensus object. By the universality of the consensus object[27] this a fortiori implies impossibility of ....

....sophisticated special purpose functionality into the disk controllers. Ballots roughly correspond to rounds and to ranks in the round based and the ranked register respectively. The NASD project also addresses other aspects of the network attached disk technology such as file system support [23], security [24] and network protocols [22] Active Disks [42, 2] is a logical extension of the OSD concept which allows arbitrary application code to be downloaded and executed on disks. One of the applications of the active disks technology is enhancing disk functionality with specialized ....

G. A. Gibson, D. F. Nagle, K. Amiri, F. W. Chang, H. Gobioff, E. Riedel, D. Rochberg and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, July 1997.

.... and freeblock scheduling [26] More recent work in the storage community suggests that the next evolution in storage will place disks on a more general purpose network and not a standard SCSI bus [17] Some have suggested that these network disks export a higher level, object like interface [18], thus moving the responsibilities of low level storage management from the file system into the drives themselves. Although the specific challenges would likely be different in this context, the fixed object based interface between file systems and storage will likely provide an interesting ....

G. A. Gibson, D. F. Nagle, K. Amiri, F. W. Chang, H. Gobioff, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, Carnegie Mellon University, 1997.

....capability provided by SRPC can enhance the performance and functionality of ScFS. In our prototype, a network attached disk can be configured to export one of two interfaces to clients. The first and most commonly used is an object based interface similar to that proposed by Gibson et al. [19]. The exported object namespace is nonhierarchical and does not distinguish between directories and normal files. Read and write requests are arbitrarily fine grained, specifying an object identifier, offset, and length. The second is a blockbased interface, which allows us to explore the ....

G. A. Gibson, D. F. Nagle, K. Amiri, F. W. Chang, H. Gobioff, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS97 -118, Carnegie-Mellon University, 1997.

....the intelligent storage control modules to be embedded in a small chip at a cheap price and deal with direct access from clients, the location of disk drives need not be close to hosts any longer. This type of intelligent disk drives is 101 categorized into network attached secure disks (NASDs)[1 4]. In NASDs, upon receiving a data request from a client, the central file manager locates the data and provides the client with necessary information to access the data such as the address of a NASD, keys for access verification, etc. Then the client provides a NASD with proof of permission ....

G A Gibson et al., "File systems for Network-Attached Secure Disks," CMU technical report CMU-CS-97-118, 1997.

....functionality. Finally, the current status of OSD and related roadmaps are presented as a frame of reference. Brief History of OSD The most active work on OSD has been done at the Parallel Data Lab at Carnegie Mellon University (www.pdl.cmu.edu) originally under the direction of Garth Gibson [1,4,5,6,8]. This work focused on developing the underlying concepts of OSD and two closely related areas called Network Attached Secure Disks (NASD) and Active Disks. Other work has been done at the University of California at Berkeley [Keeton] the Universities of California Santa Barbara and Maryland [3] ....

G. Gibson et al, "Filesystems for Network-Attached Secure Disks", White Paper CMU-CS-97-118, July 1997

....for a rogue client to compromise the security of the NSL storage environment with this mechanism. A similar level of security must be developed for use in a current SAN environment before the true power of SAN file systems can be safely realized. Object based Network Attached Secure Disks [7] could solve this problem if they are accepted within the storage marketplace. 5. The Development of HPSS The HPSS collaboration [4, 5] took up the work of the National Storage Laboratory collaboration in 1992 under a Cooperative Research and Development Agreement (CRADA) between IBM and several ....

....(Zoning is a SAN capability that allows users to create multiple logical subsets of devices within a physical SAN as mentioned earlier. Access to devices within that zone is restricted to the members of the zone. For this reason and until improved technology such as secure object based devices [7] are available, server facilitated access is currently the safest course for a file or storage system shared across computers. SAN Core Both Servers have SAN Enabled Clients Direct Client Disk Transfer Direct Client Tape Transfer San Enabled Mover San Enabled Mover ....

Garth A. Gibson, David F. Nagle, Khalil Amiri, Fay W. Chang, Howard Gobioff, Erik Riedel, David Rochberg, and Jim Zelenka. "Filesystems for Network-Attached Secure Disks" CMU-CS-97-118 July 1997

....efforts that propose or would benefit from more expressive storage interfaces. For example, some have proposed to dynamically place data near the disk head [12, 54] or to piggyback write backs on rotational delays [4] Perhaps the clearest examples are recent proposals for object based storage [19, 18, 35] and those for Active disks [1, 41, 25] Object based storage defines a new storage interface that is much like a file system with a flat namespace; this does make the interface somewhat more expressive, but not nearly enough in our view. Active disks would allow application code to be downloaded ....

Garth A. Gibson, David F. Nagle, Khalil Amiri, Fay A. Chang, Howard Gobioff, Erik Riedel, David Rochberg,and Jim Zelenka. Filesystems for network-attached secure disks. CMU-CS-97-118. Computer Science Department, Carnegie-Mellon University, Pittsburgh, PA, July 1997.

....from more than one host, allows multiple access paths to data, thus increasing data availability. Simplicity: File systems built on network attached storage can leave low level layout decisions and performance optimizations to the drives, simplifying software and increasing maintainability [14]. Incremental Growth: As compared to a typical RAID array, network attached storage allows for essentially unlimited growth in the number of disks, removing the need for a fork lift replacement of an entire RAID array [10] Specialization: Network attached storage mixed within a cluster ....

G. A. Gibson, D. F. Nagle, K. Amiri, F. W. Chang, H. Gobioff, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, Carnegie-Mellon University, 1997.

....approach is the lack of scalability. As the number of clients grow, the server becomes a performance bottleneck. In a distributed server environment (e.g. Global File System (GFS) 20] Frangipani [22] CentraVision File System (CVFS) 2] File systems for network attached secure disks (NASD) [10]) a set of designated nodes share the load and functions of the server. Thus the load is distributed over the servers enabling scalability of the system. More nodes can be added as the number of clients increases. A serverless system (e.g. xFS [5] is employed in a trusted environment where ....

....in the public domain. In this section we present a comparison study of DFS with two other existing file systems. We use the popular network file system (NFS) 18] that is widely used in a distributed file sharing environment. We also use CMUNASD and the file system associated with it [16] [10], henceforth referred to as NASD FS. Also note that it is difficult, if not impossible, to make a fair comparison among these file systems as will be apparent below. Network File System: The Network File System (NFS) was designed by Sun Microsystems in 1985 to allow transparent file sharing among ....

G. A. Gibson, D. F. Nagle, K. Amiri, J. Butler, F. W. Chang, H. Gobioff, C. Hardin, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, Carnegie Mellon University, July 1997.

....but the failure of the server will affect all the clients at the same time without any graceful degradation. In a distributed server environment (e.g. Global File System (GFS) 16] Frangipani [18] CentraVision File System (CVFS) 2] File systems for network attached secure disks (NASD) [8,9]) a set of designated nodes share the load and functions of the server. Thus the load is distributed over the servers enabling scalability of the system. More servers can be added as the number of clients increases. There is no single point of failure and in case of failure of a server the system ....

G. A. Gibson, et al., "Filesystems for Network-Attached Secure Disks," Tech. Report CMU-CS-97118, Carnegie Mellon University, July 1997. Available at http://www.cs.cmu.edu/Web/Groups/NASD.

....maintained, the failure of the server will a ect all the clients at the same time without any graceful degradation. In a distributed server environment (e.g. Global File System (GFS) 25] Frangipani [28] CentraVision File System (CVFS) 2] File systems for network attached secure disks (NASD) [12]) a set of designated nodes share the load and functions of the server. Thus the load is distributed over the servers enabling scalability of the system. More nodes can be added as the number of clients increases. There is no single point of failure. In case of failure of a node the system ....

....In this section we present a comparison study of DFS with two other existing le systems. We use the popular network le system (NFS) 23] that is widely used in a distributed le sharing environment. The other le system we use for the comparison is the NASD and the le system associated with it [21, 12], henceforth referred to as NASD FS. Also note that it is dicult, if not impossible, to make a fair comparison among these le systems as will be apparent below. Network File System: The Network File System (NFS) was designed by Sun Microsystems in 1985 to allow transparent le sharing among ....

[Article contains additional citation context not shown here]

G. A. Gibson, D. F. Nagle, K. Amiri, J. Butler, F. W. Chang, H. Gobio, C. Hardin, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, Carnegie Mellon University, July 1997.

....maintained, the failure of the server will a ect all the clients at the same time without any graceful degradation. In a distributed server environment (e.g. Global File System (GFS) 18] Frangipani [20] CentraVision File System (CVFS) 2] File systems for network attached secure disks (NASD) [9]) a set of designated nodes share the load and functions of the server. Thus the load is distributed over the servers enabling scalability of the system. More nodes can be added as the number of clients increases. There is no single point of failure. In case of failure of a node the system ....

....In this section we present a comparison study of DFS with two other existing le systems. We use the popular network le system (NFS) 16] that is widely used in a distributed le sharing environment. The other le system we use for the comparison is the NASD and the le system associated with it [14, 9], henceforth referred to as NASD FS. Also note that it is dicult, if not impossible, to make a fair comparison among these le systems as will be apparent below. 19 Network File System: The Network File System (NFS) was designed by Sun Microsystems in 1985 to allow transparent le sharing among ....

G. A. Gibson, D. F. Nagle, K. Amiri, J. Butler, F. W. Chang, H. Gobio, C. Hardin, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, Carnegie Mellon University, July 1997. 23

.... 3 NASD Implementation To experiment with the performance and scalability of NASD, we designed and implemented a prototype NASD storage interface, ported two popular distributed file systems (AFS and NFS) to use this interface, and implemented a striped version of NFS on top of this interface [Gibson97b]. The NASD interface offers logical partitions containing a flat name space of variable length objects with size, time, security, clustering, cloning, and uninterpreted attributes. Access control is enforced by cryptographic capabilities authenticating the arguments of each request to a file ....

Gibson, G.A., et al., "Filesystems for NetworkAttached Secure Disks," in preparation, CMUCS -97-118.

....controls on a distributed file system that uses network disks. A network disk is basically a hard drive coupled with a board computer with a network interface. In a fully implemented system, there would be many of these disks, with the files spread across several in a RAID configuration [6]. The file system discussed in this study provides a raw block read write create service. Each of these schemes is analyzed to determine what cryptographic functions are performed by the host and network disk for a read and write operation. Combining this information with the performance analysis ....

GARTH A. GIBSON, DAVID F. NAGLE, KHALIL AMIRI, FAY W. CHANG, HOWARD GOBIOFF, ERIK REIDEL, DAVID ROCHBERG, AND JIM ZELENKA, "Filesystems for Network-Attached Secure Disks," http://www.pdl.cs.cmu.edu/PDLFTP /NASD/CMU-CS-97-118.pdf.

....as part of a distributed application (such as a distributed filesystem, database, or video server) and serves as a repository for data. A single type of storage device may be used in a wide variety of applications, thus only the most common set of semantics are fixed in the NASD interface [Gibson97b], a proposed network attached storage system. The limited interface of storage devices forces most of Clients Storage Network Server LAN Client NASD NASD Server LAN (a) Server Attached Disk (b) Network Attached Secure Disk Figure 1 1 SAD versus NASD On the left, the traditional approach, which ....

....management and application specific semantics from generic data movement operations. A sophisticated server machine handles the former while a low level storage device focuses on the latter. Our group has studied and proposed a design for a next generation interface for commodity storage devices [Gibson97b]. This interface is currently the basis for prestandards discussions among storage vendors as a potential follow on to SCSI as part of the National Storage Industry Consortium s Network Attached Storage Working Group [NSIC96] and formed the starting point for Seagate s Object Oriented Disk (OOD) ....

[Article contains additional citation context not shown here]

Gibson, G., Nagle, D., Amiri, K., Chang, F., Gobioff, H., Riedel, E., Rochberg, D., and Zelenka, J., Filesystems for Network-Attached Secure Disks, School of Computer Science, Carnegie Mellon University, Technical Report CMU-CS-97-118, 1997.

....relatively low level, and has not been updated in many years. One of the contentions of previous work on Network Attached Secure Disks (NASD) is that the simple block level interface of SCSI should be replaced by a richer object interface that allows the drives to manage more of their own metadata [Gibson97a]. This allows both offloading of the host and provides additional opportunities for optimization at the devices. In this system, a particular filesystem built on top of network attached disks decides how to map user level objects , such as files, directories, or database tables onto the objects ....

....is up to the filesystems that are implemented on top of NASD. The group at Carnegie Mellon has explored the mapping of the Network File System (NFS) and the Andrew File System (AFS) onto the object interface, where each file and directory in these filesystems is mapped onto a separate NASD object [Gibson97a]. One could assume more complex mappings where larger groups of files are mapped onto objects, or where a single file is broken into multiple objects, but this work only considers the case where files map onto single NASD objects. This allows the entire file to be treated as a single unit for ....

Gibson, G., Nagle, D., Amiri, K., Chang, F.W., Gobioff, H., Riedel, E., Rochberg, D., and Zelenka, J. "Filesystems for Network-Attached Secure Disks" Technical Report CMU-CS-97112, Carnegie Mellon University, March 1997.

....connected to the network. The primary motivation for the NASD architecture is cost effective, scalable performance. Directly attaching disks to the network can decrease the file server load [6] and increase the aggregate bandwidth available beyond what is available from a single storage server [7, 10]. File management is different in NASD systems than in classical distributed file systems such as Andrew File System [12] Sun s Network File System [23] or Novell Netware. In classical distributed file systems, a file server manages file storage on devices that are physically part of the file ....

Garth A. Gibson, David F. Nagle, Khalil Amiri, Fay W. Chang, Howard Gobioff, Erik Riedel, David Rochberg, and Jim Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, Carnegie Mellon University, March 1997.

No context found.

G. Gibson, D. Nagle, K. Amiri, F. Chang, H. Gobioff, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for network-attached secure disks. Technical Report CMU--CS--97--112, Carnegie Mellon, March 1997.

No context found.

Garth Gibson, David Nagle, Khalil Amiri, Fay Chang, Howard Gobioff, Erik Riedel, David Rochberg, Jim Zelenka, "Filesystems for Network-Attached Secure Disks" CMU Computer Science Technical Report, CMU-CS-97-118. July 1997.

No context found.

G. A. Gibson, D. F. Nagle, K. Amiri, F. W. Chang, H. Gobio #, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS97 -118, Carnegie Mellon University, 1997.

No context found.

G. A. Gibson et al., "Filesystems for Network-Attached Secure Disks," 1997.

No context found.

G. A. Gibson, D. F. Nagle, K. Amiri, F. W. Chang, H. Gobioff, E. Riedel, D. Rochberg and J. Zelenka. Filesystems for network-attached secure disks. Technical Report CMU-CS97 -118, July 1997.

No context found.

G. Gibson, D. Nagle, K. Amiri, F. Chang, H. Gobioff, E. Riedel, D. Rochberg, and J. Zelenka. Filesystems for network-attached secure disks. Technical Report CMU--CS--97--112, Carnegie Mellon, March 1997.

No context found.

G. A. Gibson et al., "Filesystems for Network-Attached Secure Disks," 1997.

No context found.

Garth Gibson, David Nagle, Khalil Amiri, Fay W. Chang, Howard Gobioff, Erik Riedel, David Rochberg, and Jim Zelenka. Filesystems for Network-Attached Secure Disks. Technical Report CMU-CS-97-118, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, 1997.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC