Electrical computers and digital processing systems: memory – Storage accessing and control – Specific memory composition
Reexamination Certificate
1999-06-30
2003-03-04
Kim, Matthew (Department: 2186)
Electrical computers and digital processing systems: memory
Storage accessing and control
Specific memory composition
C711S209000, C711S202000, C711S170000
Reexamination Certificate
active
06529996
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to tape data storage systems and, in particular, to a plurality of tape devices which are connected to a plurality of data processors via a network and which collectively implement a virtual, distributed tape data storage subsystem. The virtual, distributed tape data storage subsystem realizes multiple virtual devices, which are available to any of the data processors via the network which has a scalable bandwidth and can be changed on demand.
Problem
It is a problem in the field of data storage subsystems to provide adequate data storage service to the data processors that are connected to the data storage devices. There are numerous data storage media in use as well as corresponding data storage subsystem configurations which attempt to improve the data storage capabilities of the data storage media that is used to implement the data storage devices. For example, increases in the area density of data storage products translate into lower data storage costs per bit, but do not always yield higher data transfer rates. To achieve increased data transfer rates requires architectural approaches to data storage rather than data storage device improvements. One other aspect of this data storage problem is that the allocation of customer data to a single type of data storage media represents a limitation when faced with. widely varying data storage needs. This limitation can be partly obviated by balancing I/O activity across an array of data storage devices of a data storage subsystem. However, a fixed array configuration of data storage devices also limits the scalability in performance and provides no facility for applications to request changes in performance. An architecture where the data storage devices are located behind a server further limits the delivered performance since the bandwidth is limited by the server itself. Therefore, architecting a data storage subsystem that can efficiently serve the needs of the applications extant on the data processors is a daunting problem. There are numerous factors which effect performance and this problem is particularly pertinent to tape devices, since the tape media is experiencing significant enhancements to its data storage capacity.
The traditional tape device is directly connected to a single data processor in a dedicated tape device configuration. The data processor has exclusive use of the tape device and typically communicates with the tape device via a SCSI interface. However, the use of dedicated tape devices is an expensive proposition where there are a plurality of data processors to be served, especially if the data access loads generated by the plurality of data processors are erratic. In this data storage subsystem architecture, the utilization of the tape devices and the efficiency of the data storage function are less than optimal, since each data processor is limited to its dedicated tape device and its physical constraints.
An alternative data storage subsystem architecture is to connect a plurality of tape devices along with a plurality of data processors to a common data communication network. In this architecture, the data processors all have access to all of the tape devices. The data processors run tape server software to manage the access protocol for the plurality of tape devices. Among the problems with the network interconnected tape devices is that it is difficult to share a tape device among a plurality of data processors. To provide enhanced response time, the tape devices can be served by an automated tape cartridge library system which mounts/dismounts the tape cartridges for the plurality of tape devices served by the automated tape cartridge library system. However, the tape cartridge library systems typically have a SCSI interface in the data path, and the SCSI interface introduces a number of physical limitations to the operation of the automated tape cartridge library system. The first limitation is that only a small number of tape devices can be attached to a SCSI bus compared to other bus architectures. The second limitation is the limited bandwidth of the SCSI bus that is shared by these tape devices. The length of the SCSI cable may also represent an additional limitation, since the length of the SCSI bus is typically limited to 25 feet.
A variation of this network data storage architecture is the use of a plurality of tape devices configured into a tape array. The tape devices are configured in a redundant array of data storage devices in a manner analogous to the Redundant Array of Inexpensive Disks (RAID) which is a well known architecture in the disk device technology. The tape array is typically located behind a server, which is directly connected to the network, and the bandwidth for data transfers between the data processors and the tape array is not scalable and is also limited by the characteristics of the server. The tape array itself is also not scalable or easily changed due to the server limitation.
These various tape device based data storage subsystem architectures are all limited in their data exchange bandwidth and are restricted by the use of a single data storage media. The alternative to a single media data storage subsystem is the integration of a plurality of data storage media types and data storage subsystem architectures into a single data storage subsystem, typically termed a mass storage subsystem. One such data storage subsystem designed to address this problem is the mass storage system described in the paper by Sam Coleman and Steve Miller titled “Mass Storage System Reference Model: Version 4”, published May 1990 by the IEEE Technical Committee on Mass Storage Systems and Technology. This mass storage system interconnects a plurality of data processors with a diversity of data storage subsystems via a high bandwidth switched network for the transmission of data therebetween at high data transfer rates. A separate network is used to interconnect the data processors with the mass storage system controller, which manages the processing of data transfer requests received from the data processors over the control network. The mass storage system controller is directly connected to the controllers of the various data storage subsystems and transmits data file retrieval requests to the selected controller in response to the received data file request received from the data processors. The file staging process used therein copies a data file in its entirety from the mass storage system to the client data processor via the high bandwidth network before the user accesses any of the requested data. Usually the data file is staged from an archival storage device, but staging from a direct access storage device is possible as well. File staging is initiated by a client data processor transmitting a request to the mass storage system identifying the data file by name. The mass storage system maintains mapping information indicative of the physical memory storage location occupied by the requested data file. The retrieved mapping information is used by the mass storage system to transmit a file retrieval request to the archival storage device on which the requested data file is stored. Upon receipt of this request, the designated storage device copies the requested data file over the high speed network to a local, direct access data storage device that serves the requesting client data processor.
A significant limitation of this architecture is that the data is managed on a data file basis. Each client data processor request for a data file causes the mass storage system to access the mapping tables to locate the requested data file. When a client data processor sequentially accesses a plurality of data files, the mass storage system must successively access the mapping tables to identify each requested data file. As the extent of the mass storage system data storage capacity increases, the size and extent of the mapping tables proportionately increases and the time required to retrieve mapping information for each data file becom
Leonhardt Michael Lawrence
Nguyen Thai
Raymond Robert Michael
Bailey Wayne P.
Bataille Pierre-Michel
Duft, Graziano and Forest Brooks & Kushman P.C.
Kim Matthew
Storage Technology Corporation
LandOfFree
Network attached virtual tape data storage subsystem does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Network attached virtual tape data storage subsystem, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Network attached virtual tape data storage subsystem will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3035843