Dataplow, Inc.

Other

Dataflow, Inc.

Steven R Soltis

Michael H Miller

Kenneth William Preslan

Raymond Wesley Gilson

Edward Alan Soltis

Matthew O'Keefe

Gerald A Houlder

James Allen Coomes

Thomas M Ruwart

A shared storage distributed file system is presented that provides clients with access to a storage area network (SAN) attached storage device. The access is provided either via block-level requests over the SAN or via file-level requests made to a server over a local area network (LAN). File requests are categorized between SAN requests and LAN requests. This categorization can be based on dividing the requests between write and read requests, or between requests that alter data stored on the storage device and requests that do not alter data on the storage device. In other embodiments, the categorization can be made on the size of the file or the data involved in the file request. In one embodiment, the client is able to access and interpret metadata stored on the storage device.

Storage area network file system

A shared storage distributed file system is presented that provides users and applications with transparent access to shared data stored on network attached storage devices by utilizing layering techniques to inherit file management functionality from existing file systems. The present invention stores meta-data for the shared data as real-data in a standard, non-modified, client-server distributed file system, such as NFS. In effect, the standard client-server file system acts as a meta-data server. The name space consisting of inode files stored as real-data on the meta-data server acts as the name space for the shared data. Similarly, file attributes of the inode files are utilized as the file attributes of the shared data. By utilizing an existing client-server system as the meta-data server, development time and complexity are greatly reduced, while speed advances in the underlying client-server system may be incorporated without alteration of the present invention. A method for communicating with network attached storage devices over layered file systems is also presented.

Shared file system

A shared storage distributed file system is presented that provides applications with transparent access to a storage area network (SAN) attached storage device. This is accomplished by providing clients read access to the devices over the SAN and by requiring most write activity to be serialized through a network attached storage (NAS) server. Both the clients and the NAS server are connected to the SAN-attached device over the SAN. Direct read access to the SAN attached device is provided through a local file system on the client. Write access is provided through a remote file system on the client that utilizes the NAS server. A supplemental read path is provided through the NAS server for those circumstances where the local file system is unable to provide valid data reads. Consistency is maintained by comparing modification times in the local and remote file systems. Since writes occur over the remote file systems, the consistency mechanism is capable of flushing data caches in the remote file system, and invalidating metadata and real-data caches in the local file system. It is possible to utilize unmodified local and remote file systems in the present invention, by layering over the local and remote file systems a new file system. This new file system need only be installed at each client, allowing the NAS server file systems to operate unmodified. Alternatively, the new file system can be combined with the local file system.

A system includes shared Small Computer System Interface (SCSI) storage devices for processing clients coupled by a fiber channel interface. The storage devices include storage blocks, and locks controlling their use by clients. Clients issue actions to the storage devices for performing operations on the locks. A client may exclude other clients from using storage blocks using a state element to acquire the lock for shared or exclusive use. If the client modified the data, a version counter in the lock is updated when the lock is released. If an activity bit is set, the version counter is updated upon both reading and writing. Other clients can forcibly release a lock owned by a failed client by monitoring its version counter. Expiration timers associated with the locks allow acquired locks to be released by timing out. A serverless global file system (GFS) manages use of the shared storage resources, and allows remapping of the locks to the storage blocks, for example, based on activity of the locks.

Growing community of inventors

Rochester, MN, United States of America

Steven R Soltis