System and method for identifying failing drives or media in media library

Abstract

Embodiments of methods and systems comprise identifying failing media and/or drives for a media library. Error data can be collected from media libraries. For each tape exhibiting an error rate of interest, a determination can be made whether the tape would still have been of interest had it not been loaded in certain drives. This information can be analyzed to identify failing drives or tapes.

Description

TECHNICAL FIELD

This disclosure describes various embodiments of methods and systems for identifying failing drives or media in a media library.

BACKGROUND

Data represents a significant asset for many entities. Consequently, data loss, whether accidental or caused by malicious activity, can be costly in terms of wasted manpower, loss of goodwill from customers, loss of time and potential legal liability. To ensure proper protection of data for business and legal purposes, many entities back up data to a physical storage media such as magnetic tapes or optical disks. Traditionally, backup would occur at each machine controlled by an entity. As the sophistication of network technology increased, many entities turned to enterprise level backup in which data from multiple machines on a network is backed up to a remote library. Typically, a library includes a variety of components which include a plurality of media for data storage, such as, for example, multiple magnetic tapes. Centralized data backup has the advantage of increased volume, efficiency and redundancy.

In many systems, the data to be backed up and backup commands are sent over a network from multiple machines on the network to a library. In many instances, the data to be backed up and the backup commands are routed to the library through a switch.

One example of a library commonly used in enterprise backup systems is a magnetic tape library. A magnetic tape library can comprise components such as tape cartridges (containing magnetic tape), robots, tape slots and tape drives. A typical magnetic tape library contains multiple cartridge slots in which tape cartridges can be stored. Tape cartridges, commonly referred to as tapes, are physically moved between cartridge slots and tape drives by a robot. The robot is controlled by commands received from the host devices on the network. When specific data is required, a host device determines which cartridge slot contains the tape cartridge that holds the desired data. The host device then transmits a move-element command to the robot and the robot moves the tape cartridge to a tape drive which reads the desired data from the tape cartridge.

In a SCSI tape library, devices that are part of the library are typically addressed by target number. Thus, each drive and robot of a tape library typically has a target number. Cartridge slots, on the other hand, are addressed by element numbers that are used by the robot to locate the slots. Because the robot also places tape cartridges in the drives, each drive is also associated with an element number.

Components of a library are subject to wear and other forms of degradation. The degradation of library components can deleteriously affect the reliability of the library. It can be difficult to identify the source of particular errors, however. Previous solutions to reducing errors include replacing drives and media without in-depth of analysis of why errors occurred. Consequently, time and money are spent replacing tapes and drives that are good. Likewise, drives and media that are actually failing may be retained, leading to continuing errors.

SUMMARY

Embodiments described herein provide systems and methods for identifying failing drives and tapes. According to one embodiment, tapes that experienced error rates of interest can be identified. A set of drive impact data can be recorded that indicates whether each tape in the identified set of tapes would still have experienced an error rate of interest had that tape not been loaded in various drives. This information can used to determine whether for each tape/drive combination, the tape or drive or both are suspected of causing the errors.

According to one embodiment, tapes having an error rate of interest can be identified by comparing the error rates for the tapes to a threshold. One embodiment can determine, based on the threshold, whether each tape in the identified set of tapes would still have experienced an error rate of interest had that tape not been loaded in various drives. The results can be recorded in drive impact data and used to identify whether tapes and/or drives are failing.

BRIEF DESCRIPTION OF THE FIGURES

A more complete understanding of embodiments of methods and systems and the advantages thereof may be acquired by referring to the following description, taken in conjunction with the accompanying drawings in which like reference numbers indicate like features and wherein:

FIG. 1 is a diagrammatic representation of one embodiment of a library;

FIGS. 2A-2C are diagrammatic representations of embodiments of example network topologies comprising a library and a monitoring appliance;

FIG. 3 is a flowchart of one embodiment of a method for collecting or storing data;

FIG. 4 is a flowchart of a method for tracking tape movements;

FIG. 5 depicts an XML representation of one embodiment of a data structure;

FIGS. 6A and 6B (FIG. 6) depicts an XML representation of one embodiment of a data structure;

FIG. 7 depicts XML representations of embodiments of data structures;

FIG. 8 is one example of a block diagram of a system for collecting or storing data;

FIG. 9 is a block diagram of one example of a controller operable to collect or store data;

FIG. 10 is one example of a graphical user interface for displaying data;

FIG. 11 is one example of a graphical user interface for displaying data;

FIG. 12 is one example of a graphical user interface for displaying data;

FIG. 13 is one example of a graphical user interlace for displaying data;

FIG. 14 is one example of a graphical user interface for displaying data; and

FIG. 15 is one example of a graphical user interface for displaying data;

FIG. 16 is a diagrammatic representation of error data;

FIG. 17 is a diagrammatic representation of a report;

FIG. 18 is a flow chart illustrating one embodiment of method for analyzing error data to allow failing drives and/or tapes to be identified; and

FIG. 19 is a diagrammatic representation of one embodiment of a management station.

DETAILED DESCRIPTION

Embodiments of systems and methods for library monitoring and identifying sources of errors are illustrated in the FIGURES, like numerals being used to refer to like and corresponding parts of the various drawings.

As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, product, article, or apparatus that comprises a list of elements is not necessarily limited only those elements but may include other elements not expressly listed or inherent to such process, product, article, or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).

Additionally, any examples or illustrations given herein are not to be regarded in any way as restrictions on, limits to, or express definitions of, any term or terms with which they are utilized. Instead these examples or illustrations are to be regarded as being described with respect to one particular embodiment and as illustrative only. Those of ordinary skill in the art will appreciate that any term or terms with which these examples or illustrations are utilized will encompass other embodiments which may or may not be given therewith or elsewhere in the specification and all such embodiments are intended to be included within the scope of that term or terms. Language designating such nonlimiting examples and illustrations includes, but is not limited to: “for example,” “for instance,” “e.g.,” “in one embodiment.”

Media library monitoring systems can collect data associated with a media library and components of the media library. Data may be collected over time and stored in the repository such that the repository contains data associated with individual library components such as, for example, drives, media changers or other library components. Because the repository contains data associated with different library components and because the data in the repository extends over time, the data in the repository may be organized such that the data is organized with respect to time and with respect to the individual library components or according to other parameters. Additionally, the data can be analyzed to identify failing drives and/or tapes.

A user interface may be used to display the collected and organized data to a user and the user may select data to view utilizing the user interface. Thus, collecting data in a repository over time and organizing the data allows a library and individual library components to be monitored and problems with the library or library components to be identified and rectified, for example, by allowing unreliable or unacceptably degraded components to be identified and bypassed or replaced, thereby enhancing the reliability of the library and proactively preventing data loss. According to one embodiment, reports can be presented to a user that allow the user to quickly identify failing drives and/or tapes.

FIG. 1 is a diagrammatic representation of one embodiment of a media library, in this example, a tape library. Library 100 can comprise drives 140a-140e, media changer 125 and associated robot 130, import/export element(s) 145 and slots 135a-135j. Drives 140a-140e can read/write data from/to magnetic tape (contained within cartridges), eject tape cartridges and perform other operations. Slots 135a-135j store the magnetic tape cartridges when they are not in a drive and robot 130 moves the magnetic tape cartridges between drives 140a-140e and slots 135a-135j. For example, robot 130 may move a tape cartridge stored at slot 135a to drive 140b so that data can be written to the tape cartridge. It should be noted that libraries may employ a single robot or multiple robots in an expandable or modular configuration.

To collect data associated with a library or library components, a monitoring appliance can query a library or library components over a network utilizing commands. In response to received commands, the library or library components may return data associated with a particular command to the monitoring appliance. In one embodiment, a monitoring appliance can query a library over a network utilizing SCSI commands such as the Read Element Status command, Log Sense Command, Inquiry Command and other commands.

A Read Element Status (RES) command is a command which is used to query the state of a library. A RES command is sent to a media changer and in response, the media changer returns data associated with the library or the media changer. Such data may include the locations of individual tape cartridges. Thus a RES command provides a snapshot of a library at any one time. Examples of a RES command can be found in “SCSI Media Changer Commands-2 (SMC-2)”, (INCITS T10 Project 1383D), Revision 7, Nov. 18, 2003, propagated by the T10 Technical Committee of the InterNational Committee on Information Technology Standards (INCITS), which is hereby incorporated by reference.

A Log Sense (LS) command is a command which is used to obtain data associated with a particular drive. A LS command is sent to a particular drive of a library and in response, the drive returns data associated with the drive and/or the media contained in the drive depending on the actual LS command. For example, such data might include: read errors, write errors, utilization and performance data, data regarding the data written and read to a media, media loaded, detail tables or other data. In one embodiment, the amount of data written over a period of time may be derived from data returned in response to one or more LS commands. More specifically, data returned in response to an LS command may include Log Sense page 0xC, 0x32 or 0x39 which may include data from which the amount of data written by a particular drive over a period of time may be derived. In one embodiment, deriving the amount of data written by a particular drive over a period of time may comprise one or more calculations. Examples of a LS command can be found in “SCSI Primary Commands-3 (SPC-3)”, (Project T10/1416-D), Revision 22a, Mar. 25, 2005, propagated by the T10 Technical Committee of the InterNational Committee on Information Technology Standards (INCITS), which is hereby incorporated by reference.

An Inquiry command is a command that is used to query relatively static information (which may include, for example, serial number, manufacturer or other relatively static information) from components of a library such as a drive or a media changer. According to one embodiment, Inquiry commands are used to query individual library components. That is, an individual Inquiry command may query a particular library component. Examples of an Inquiry command can be found in “SCSI Primary Commands-3 (SPC-3)”, (Project T10/1416-D), Revision 22a, Mar. 25, 2005, propagated by the T10 Technical Committee of the InterNational Committee on Information Technology Standards (INCITS), referenced above.

Methods and systems for collecting data from a library or library components can utilize a monitoring appliance which can be a Read Verify Appliance (RVA). The monitoring appliance queries a library or library components over time by sending RES, LS, Inquiry commands and/or other commands to the library or library components at intervals. Data returned in response to the commands is collected in a repository such that the repository may contain data associated with a plurality of library components of a library. For example, the repository may contain data regarding all the drives in a library. Data within the repository may be organized to allow a user to monitor various library components. For example, data may be organized in a chronological order so that a user can monitor the incidence of errors (for example, soft and hard read or write errors) over time. The data may further be organized such that superfluous data is removed. For example, redundant data might not be displayed to a user or might be purged from the repository. Likewise, unnecessary data or data that the system is configured not to monitor might be purged from the repository or not saved to the repository in the first place.

FIG. 2A is a diagrammatic representation of a system 200a in which a plurality of hosts 202a-202d have access to library 100 over network 205. Network 205 can comprise the Internet, a SAN, a LAN, a WAN, a wireless network or any other communications network known in the art. Hosts 202a-202d are coupled to library 100 via network 205 and switch 207. Similarly, library 100 can communicate with switch 207 over any suitable communications link or network known in the art and can use additional protocols such as iSCSI, TCP/IP, or other protocol known in the art. Monitoring appliance 220 is coupled to switch 207 such that it can send commands to library 100 or library components of library 100.

Switch 207 is connected to library 100. Thus switch 207 can forward commands (such as, for example, RES, LS or Inquiry commands) from monitoring appliance 220 to library 100. Switch 207 receives data generated in response to the commands from library 100 and forwards the data to monitoring appliance 220 which collects the data in a repository. Thus monitoring appliance 220 can continually query library 100 and components of library 100 for data which can be stored in a repository, allowing a user to monitor the components of library 100. Because monitoring appliance 220 collects data by sending commands to a library, monitoring appliance 220 does not need to intercept commands or responses from network 205 to obtain data associated with a library or library components. Thus, in one embodiment, monitoring appliance 220 can be an out-of-band appliance. This allows monitoring appliance 220 to be a non-intrusive device which does not monitor or interfere with commands from and responses to hosts 202a-202d. Consequently, monitoring appliance 220 can be a compartmentalized device which can be connected to a switch and which does not have to be integrated into network 205. An advantage of this out-of-band methodology is that a monitoring appliance can be used to monitor library components without cumbersome configuring.

While a particular network topology is shown in FIG. 2A, a monitoring appliance may be used in other network topologies to collect data associated with a library or library components. FIGS. 2B and 2C are diagrammatic representations of example network topologies in which a monitoring appliance can be used to collect data associated with a library or components of the library.

For example, FIG. 2B is a diagrammatic representation of a system 200b having a network topology in which monitoring appliance 220 is coupled to media changer 125 and coupled to drives 140a-140e through switch 207. Thus, monitoring appliance 220 can communicate with (by, for example, sending commands to or receiving data from) media changer 125 and drives 140a-140e. By way of further example, FIG. 2C is a diagrammatic representation of a system 200c having a network topology in which monitoring appliance 220 is coupled to drives 140a-140e through switch 207 and is coupled to media changer 125 through network 209. Thus, monitoring appliance 220 can communicate with media changer 125 and drives 140a-140e. In system 200c, networks 205 and 209 can be different types of networks. For example, network 205 might be a fibre channel network whereas network 209 might be an IP network. It should be noted, however, that the topologies of systems 200a, 200b and 200c are provided by way of example and not limitation.

While shown as a physical media library in FIG. 2A-2C, library 100 can be a virtual media library that is a virtual representation of one or more physical media libraries as presented by switch 207, a library controller or other component. Examples of library virtualization are described in U.S. patent application Ser. No. 10/704,265, entitled SYSTEM AND METHOD FOR CONTROLLING ACCESS TO MULTIPLE PHYSICAL MEDIA LIBRARIES, and U.S. patent application Ser. No. 10/703,965, entitled SYSTEM AND METHOD FOR CONTROLLING ACCESS TO MEDIA LIBRARIES, both of which are hereby incorporated by reference herein.

FIG. 3 is a flow chart illustrating one embodiment of a method for collecting data and organizing the collected data so that it can be accessed by or displayed to a user. According to one embodiment, the method of FIG. 3 can be implemented as a set of computer executable instructions stored on a computer readable medium at, for example, monitoring appliance 220. The set of computer executable instructions can, when executed, collect data associated with library components in a non-intrusive manner. At collect data step 310, data is collected by a monitoring appliance by querying library components of a library for data by sending commands corresponding to the library components to the library components. In response, the library components return data associated with the library components. More specifically, in one embodiment, RES, LS and Inquiry commands are sent to library components and, in response, the library components return corresponding data. Data may be collected over time by collecting data from library components at intervals. For example, various commands can be sent every 10 seconds, every 20 seconds or with any other frequency. Thus, over time, library components may be queried a plurality of times, and as a result, the repository may contain data regarding library components at different points in time.

The monitoring appliance, at compile data step 320, can compile the returned data. For example, the monitoring appliance can compile the data in defined structures which, in one embodiment, may include XML structures or other structures or equivalents. A structure may contain data associated with a library component returned in response to one or more commands (such as, for example, RES, LS or Inquiry commands). For example, a XML structure can include data from RES commands and serial numbers of library components determined from Inquiry commands issued to corresponding library components. At store data step 330, the monitoring appliance stores the returned data in a repository. Storing data in a repository may comprise storing structures in the repository or may comprise unpacking returned data compiled at compile step 320 and storing the data in the repository. In one embodiment, the data may be stored in records in a database at the repository.

At organize data step 340, the data in the repository is organized. As part of step 340, data in the repository can be arranged in chronological order and/or superfluous or redundant data can be purged. Corresponding data can be grouped together. For example, data structures corresponding to a particular drive may be arranged in chronological order within the repository.

At process data step 350, data in the repository may be processed to display desired data to a user. For example, a user may only want to see data corresponding to a particular library component. At process data step 350, the data in the repository is processed such that the desired data is selected to be displayed to a user. Similarly, data may be processed into graphs or charts, or in accordance with any number of purposes or user desires.

In one embodiment, processing data in the repository can comprise comparing returned library states or data regarding the locations of individual tape cartridges stored in the repository to track the movement of one or more tape cartridges. For example, data in the repository corresponding to different times can be compared and the movement of tape cartridges in a library tracked by differencing the locations of tape cartridges in a library at different times. In embodiments in which data is stored in records in a database, records associated with different times may be compared to track the movement of tape cartridges. Processing data can further comprise correlating errors with a particular library component based on the movement of one or more tape cartridges within the library.

FIG. 4 is a flow chart illustrating one embodiment of a method for tracking the movement of one or more tape cartridges which can be implemented as part of process data step 350 of FIG. 3. According to one embodiment, the method of FIG. 4 can be implemented as a set of computer executable instructions stored on a computer readable medium at, for example, monitoring appliance 220. At step 460, data which may be, for example, data contained in structures or database records associated with different times may be compared. Based on comparisons between data, tape movements within the library can be tracked (step 465). For example, if in one structure or record associated with a time, a tape was at a specified drive in the library and in another structure or record corresponding to a subsequent time, the same tape is in a slot, it can be determined that the tape has been moved from the drive to the slot. Based on such comparisons between data in the repository, the movement of tapes within a library can be tracked over time. At step 470, errors may be correlated with library components based on tape movements. This allows the monitoring appliance to monitor the degeneration of library components. For example, if a drive registers an increasing number of errors, regardless of the robot or tape used with the drive, then the drive is likely degenerating and becoming increasingly unreliable. Accordingly, this allows the failing drive to be identified and bypassed or replaced before it causes data loss or library failure. While the method shown in FIG. 4 is described in the context of process data step 350 of FIG. 3, the method of FIG. 4 may be implemented as part of other steps, for example, as part of organize data step 340.

As described above, tracking the movement of tape cartridges or other media allows sources of errors to be identified. For example, a tape cartridge may encounter errors when used with a tape drive A but may encounter no errors when used with other tape drives. If other tape cartridges encounter errors when used with tape drive A, but encounter no errors when used with other tape drives, this implies that tape drive A is unreliable. Accordingly, tape drive A may be bypassed or replaced. Thus, by tracking the movement of one or more tape cartridges, errors can be correlated to a particular tape or drive. Tracking cartridges in a library and correlating data (such as, for example, errors or other data) with cartridges or drives is described in U.S. patent application Ser. No. 11/801,809, entitled METHOD AND SYSTEM FOR NON-INTRUSIVE MONITORING OF LIBRARY COMPONENTS, which is hereby incorporated by reference.

Returning to FIG. 3, the above method or steps of the above method set forth in FIG. 3 may be repeated at intervals over time such that the repository contains data associated with the library and components of the library over a period of time. For example, data associated with a library may be continually collected and compiled over time (steps 310 and 320). The period of time may be of any length, for example, days, weeks, months, years or any other length of time.

FIG. 5 is an example XML representation of a data structure containing data returned in response to an Inquiry command and a LS command. The data in the data structure may be stored in a repository along with data from multiple similar data structures containing data collected at different points in time, and the data may be organized in a chronological or other order in the repository. More specifically, in one embodiment, data may be unpacked from the data structure and stored in the repository. In a further embodiment, data may be stored in records in a database contained in the repository.

FIG. 6 is an example XML representation of a data structure containing data returned in response to an Inquiry command and a RES command. The data in the data structure may be stored in a repository along with data from multiple similar data structures containing data collected at different points in time, and the data may be organized in a chronological or other order in the repository. More specifically, in one embodiment, data may be unpacked from the data structure and stored in the repository. In a further embodiment, data may be stored in records in a database contained in the repository.

FIG. 7 is an example comparison of XML representations of structures, for example data structures, containing data returned in response to RES commands issued at different times. Comparing data may allow for the movement of tape cartridges within a library to be tracked. Because a RES command queries the state of a library, each structure in FIG. 7 may contain data regarding the state of the library, i.e. the location of tapes within the library at different times. Structure 510 contains data associated with a time. Structure 520 contains data associated with a subsequent time. By comparing the differences in the data, it is possible to determine that the tape with volume identity 000105 has moved from the slot at element identity 0x1007 to the drive at element identity 0x101. This movement is shown by directional arrow 530.

FIG. 8 is a diagrammatic representation of a system operable to collect data from a library or library components and store the collected data in a repository where it can be accessed by a user (virtual or otherwise). In FIG. 8, library 610 (which in one embodiment may be, for example, a library such as library 100 of FIG. 1 or any other library) is coupled to network 620 via switch 630 such that library 610 can receive commands sent over network 620. Data Engine 640 is also coupled to network 620 such that it can send commands over network 620 and receive data over network 620. Data engine 640 is coupled to storage engine 650 such that it can forward data to storage engine 650. In turn, storage engine 650 is coupled to repository 660 such that it can store data in repository 660. Interface engine 670 allows user 680 to access data stored in repository 660 utilizing user interface 690. According to one embodiment, data engine 640, storage engine 650, repository 660, interface engine 670 or user interface 690 can be implemented as a set of computer executable instructions stored on a computer readable medium at, for example, monitoring appliance 220. In other embodiments, repository 660 can be accessed by a management station or other computer.

Data Engine 640 of FIG. 8 collects data from library 610 by sending commands (which in on embodiment may include, for example, RES, LS or Inquiry commands) over network 620 to library 610 or library components of library 610. In response, library 610 or corresponding library components return data corresponding to the received commands over network 620 to data engine 640. Data engine 640 receives the data and forwards the collected data to storage engine 650. Storage engine 650 stores the data in repository 660. For example, in one embodiment, data returned in response to an Inquiry command and a LS command may be formatted into a single XML structure at data engine 640 and forwarded to storage engine 650. Storage engine 650 may receive XML structures or other structures containing data, unpack the data and store the data in repository 660. In a further embodiment, storage engine 650 stores the data in records contained in a database at repository 660. Through the repetition of the above described process at intervals over time, repository comes to contain data associated with the library and components of the library over a period of time. Data contained in repository 660 may be organized. Organizing data may include, for example, organizing data in a chronological or other order or purging redundant data.

Collecting data associated with a library or library components in repository 660 and organizing the data allows a user to view the data to monitor library components. In FIG. 8, user 680 may access data in repository 660 via interface engine 670 utilizing user interface 690. In one embodiment, user interface 690 is a graphical user interface (GUI), which allows for a user-friendly display of data contained in repository 660. User interface 690 may be capable of displaying desired data to user 680 or displaying data which may be useful in monitoring library components. For example, user 680 may utilize user interface 690 to select desired data to display in a graphical form. The desired data may be data regarding a particular library component of library 610 or data regarding a particular type of error. Desired data may span a user specified period. More specifically, user 680 may use user interface 690 to display write errors of a particular drive for a day, month, year or any other period. If the write errors increase over time, the drive may be degrading and may need to be replaced.

While in the above example, data engine 640, storage engine 650, repository 660, interface engine 670 or user interface 690 are described as being part of a monitoring device, data engine 640, storage engine 650, repository 660, interface engine 670 or user interface 690 may be contained in a component that is not physically part of the monitoring appliance. For example, data could be collected and compiled at the monitoring device but stored in memory coupled to the monitoring appliance. In one embodiment, interface engine 670 or user interface 690 run on a host computer which is coupled to repository 660 over a network. User 680 can access data in repository 660 utilizing interface engine 670 or user interface 690 which may be running on the host computer. Thus, a user may access the data in repository 660 over a network. Collected and compiled data may be partially stored in the monitoring appliance and partially stored in separate memory. In one example, a portion of collected data may be stored in a memory which is a component of data engine 640 or storage engine 650, while a portion of collected data may be stored in a memory which is a component of repository 660 or interface engine 670. Repository 660 of FIG. 8 may include a database, spreadsheet or other computer program in which data can be stored. In a network comprising multiple libraries, each library may have a corresponding data engine or storage engine which may forward collected data to a common repository or monitoring appliance.

It should be noted that the intervals between querying a library or library components can be fixed periods of time or of variable duration or a combination of the two. Individual library components may also be queried at different intervals: for example, a drive that is being heavily used may be queried more frequently than a drive which is not being used. In one embodiment, the data engine queries a library or library component every 30 seconds. In another embodiment, the time duration between querying is less than the backup time or the tape movement time. In further embodiments, the intervals can be determined by computer algorithm or user input.

It should further be noted that data may be derived from collected data and stored in the repository or other data storage. More specifically, collected data returned in response to one or more commands may be used to derive derived data which may be stored in the repository. Deriving derived data may comprise one or more calculations. The derived data may be organized. For example, derived data may be stored in particular records in a database. Examples of derived data may include, for example, the bits read by a drive over time (which may be, for example, in bits/second), the data transfer rate of a drive or other data. Statistical analysis may be performed on data stored in a repository. For example, statistical analysis may be performed on collected data or derived data and may be used to predict device or cartridge failure based on read errors or write errors or other data.

FIG. 9 is a diagrammatic representation of a monitoring appliance controller 700 (“controller 700”). Controller can include a processor 702, such as an Intel Pentium 4 based processor (Intel and Pentium are trademarks of Intel Corporation of Santa Clara, Calif.), a primary memory 703 (which may include, for example, RAM, ROM, Flash Memory, EEPROM or other computer readable medium known in the art) and a secondary memory 704 (which may include, for example, a hard drive, disk drive, optical drive or other computer readable medium known in the art). A memory controller 707 can control access to secondary memory 704. Controller 700 can comprise a communications interface 706 (which may comprise, for example, fibre channel interface, Ethernet port or other communications interface known in the art) to connect controller 700 to, for example, a switch of a network. An I/O controller 712 can control interactions with the switch. Similarly, an I/O controller 714 can control interactions over I/O interfaces 708 and 710. Controller 700 can include a variety of input devices. Various components of controller 700 can be connected by a bus 726.

Secondary memory 704 can store a variety of computer instructions that include, for example, an operating system such as a Windows operating system (Windows is a trademark of Redmond, Wash. based Microsoft Corporation) and applications that run on the operating system, along with a variety of data. More particularly, secondary memory 704 can store a software program 730 that collects, compiles, stores, organizes or processes data associated with a library or library components. During execution by processor 702, portions of program 730 can be stored in secondary memory 704 and/or primary memory 703.

Because a repository may contain data associated with a library and library components spanning a period, a user may access the data to monitor the library or library components over a period of time. For example, a user may use a user interface to display data associated with a library component for a desired period of time. The data may be displayed in a graphical format.

FIGS. 10-15 are examples of representations of data associated with a library or library component(s) and contained in a repository. The representations are part of a GUI which may be utilized by a user to access and display data stored in a repository. FIG. 10 is one example of a graphical representation 800 of data associated with a drive. More specifically, graphical representation 800 displays data associated with a particular drive in drive utilization graph 810 and drive performance graph 820. Drive utilization graph 810 displays the drive utilization of a drive over a 30 day period. Drive performance graph 820 displays the drive performance over a 30 day period. Selector 830 can be used to select the period of time that the data displayed in graphs 810 and 820 spans.

FIG. 11 is one example of a graphical representation 900 of data associated with a drive. More specifically, graphical representation 900 displays data associated with a particular drive in data written graph 910 and data read graph 920. Data written graph 910 displays the amount of data written per day over a 30 day period. Data displayed in data written graph 910 may be derived from, for example, Log Sense pages 0xC, 0x32 or 0x39 returned in response to one or more LS commands. Data read graph 920 displays the amount of data read per day over a 30 day period. Selector 930 can be used to select the period of time that the data displayed in graphs 910 and 920 spans.

FIG. 12 is one example of a graphical representation 1000 of data associated with a drive. More specifically, graphical representation 1000 displays data associated with a particular drive in hard write errors graph 1010 and soft write errors graph 1020. Hard write errors graph 1010 displays the occurrence of hard write errors associated with the drive over a 30 day period. Soft write errors graph 1020 displays the occurrence of soft write errors associated with the drive over a 30 day period. Selector 1030 can be used to select the period of time that the data displayed in graphs 1010 and 1020 spans.

FIG. 13 is one example of a graphical representation 1100 of data associated with a drive. More specifically, graphical representation 1100 displays data associated with a particular drive in hard read errors graph 1110 and soft read errors graph 1120. Hard read errors graph 1110 displays the occurrence of hard read errors encountered by the drive over a 30 day period. Soft read errors graph 1120 displays the occurrence of soft read errors encountered by the drive over a 30 day period. Selector 1130 can be used to select the period of time that the data displayed in graphs 1110 and 1120 spans.

FIG. 14 is one example of a graphical representation 1200 of data associated with a drive. More specifically, graphical representation 1200 displays data associated with a particular drive in tape load count graph 1210. Tape load count graph 1210 displays the number of tape loads of the drive on a daily basis over a 30 day period. Selector 1220 can be used to select the period of time that the data displayed in graph 1210 spans.

FIG. 15 is one example of a graphical representation 1300 of data associated with a drive. More specifically, graphical representation 1300 displays data associated with a particular drive in drive summary 1310 and tapes used list 1320. Drive summary 1310 displays drive data such as serial number, firmware version, whether the drive needs cleaning, or other drive data. Tapes used list 1320 lists the tapes loaded into the drive over a 30 day period. Selector 1130 can be used to select the period of time that the data displayed in list 1320 spans. FIGS. 10-15 are provided by way of example, not limitation, and data associated with a library or library components may be presented in any number of ways. Furthermore, data contained in a repository may be accessed over a network using a host computer such that a user interface containing data from the repository may be displayed on a display associated with the host computer.

Tapes that experienced error rates of interest can be identified from the data in repository 660 (FIG. 8). An error rate of interest can be any error rate or range of error rates. In one particular embodiment, the error rate of interest can be an error rate that indicates that a tape experienced too many errors. According to one embodiment, tapes that experienced an error rate of interest can be identified by comparing the error rates for each tape to a threshold. The threshold can be a user assigned number, a statistical number derived from collected error rates (such as a linear regression of error rates across a large set of error rate data or other statistical measure), or other number. A single threshold can be used for a particular media library or group of media libraries or different thresholds can be used for different drives, based on drive technology, data types or other factors. Tape tracking information can be used to correlate error rates to a tape/drive combination. This can be done in repository 660 as part of storing data in repository 660 or can be done using data retrieved from repository 660 and tape tracking information. Thus, the error rates for each tape/drive combination for tapes of interest can be identified.

FIG. 16 is a diagrammatic representation of one embodiment of an example set of data 1400 for tapes that exhibited an error rate of interest. In this example, a threshold of 0.02 errors per gigabyte is used so that tapes that experienced an error rate of greater than this threshold are identified. Column 1405 lists the various tapes in a media library or multiple media libraries that experienced error rates of interest. Columns 1410 list the soft error rate (e.g., errors per gigabyte) for each tape in a drive. Cell 1415, for example, shows the error rate of Tape 1 in Drive 1 as being 5.240 errors per gigabyte. If a tape was loaded in a drive for multiple sessions, this error rate can be the highest error rate experienced by the tape in the drive, an average of the error rates experienced by the tape in the drive, the sum of error rates experienced by the tape in the drive or other measure of error rates experienced by the tape in the drive over multiple sessions. A “y” in a cell indicates that the tape was loaded in the drive, but there were no errors and an “n” indicates that the tape was not loaded in that drive. Column 1420, in this example, represents the highest error rate for each tape and row 1425 represents the highest error rate for each drive.

A set of drive impact data can be determined and recorded that indicates whether a tape would still have experienced an error rate of interest had the tape not been loaded in particular drives. For example, if Drive 4 is removed from consideration for Tape 1, the highest error rate for Tape 1 will drop to 5.204, an error rate that is still of interest. In the case of Tape 3, as another example, removing Drive 1 causes the error rate to drop to 0.012. This indicates that Tape 3 would not have been of interest had it not been loaded in Drive 1. While, in these examples, the highest error rate experienced by a tape is used to determine if the tape experienced an error rate of interest in the absence of a drive, in other embodiments the average error rate experienced by the tape across drives, or other measure of error rate can be used.

The impact of a drive on the error rate of a tape can be recorded. For example, in the case of Tape 1, the impact of not considering Drive 1 can be recorded to indicate that Tape 1 would still have been of interest even if it had not been loaded in Drive 1. Similarly, in the case of Tape 3, the impact of not considering Drive 1 can be recorded to indicate that Tape 3 would not have been of interest had Tape 3 not been loaded in Drive 1.

Other impacts can also be recorded. As an example, it is possible that a tape cartridge may only have been loaded in one drive (e.g., Tape 2 was only loaded in Drive 4). Therefore, removing the drive from the consideration of the error rate of will naturally drop the error rate to 0. It is unclear, in this case, whether Tape 2 should still be considered of interest. The uncertain impact of not considering Drive 4 can be recorded.

In other cases, a tape may still have experienced an error rate of interest when a particular drive is not considered, but the error rate of interest may be lower than when that drive was considered. Additional thresholds can be used to categorize the drop in error rate when not considering a particular drive. For example, if the highest error rate for a tape drops enough to bring it halfway between the threshold from where it had been, the tape may still be of interest, but less so. Using the example of FIG. 16, removing Drive 1 from consideration with respect to Tape 8, drops the highest error rate from 1 to 0.36. This error rate is still above the threshold, but by less than half of when Drive 1 is considered. The impact of Drive 1 on the error rate of Tape 8 can be recorded to indicate that removing Drive 1 from consideration causes a decrease in the error rate for Tape 8, but not enough to bring the error rate below the 0.02 errors per gigabyte threshold.

Any number of different defined impacts can be recorded in the drive impact data. The drive impact data can be stored in memory, in a database or otherwise stored for analysis. In one embodiment, the drive impact data can be presented to a user in a format that allows the user to easily determine if a drive or tape should be considered suspect. FIG. 17 is a diagrammatic representation of one embodiment of a report 1430 with visual indicators corresponding to the drive impact data. A trellis background indicates that the tape would still have been of interest even if it had not been loaded in the drive. A user, seeing the multiple trellised backgrounds in the row for Tape 1 (Cells 1435, 1440, 1445, 1450), can then determine that Tape 1 may be bad and should be checked or removed.

The background of Cell 1450 represents that the error rate for a tape dropped below the threshold when that drive was not considered. That is, the tape would not have been of interest had it not been loaded in that drive. For example, Cell 1450 indicates that Tape 3 would not be of interest had it not been loaded in Drive 1. This can indicate that Drive 1 is suspected of being responsible for the error rate in cell 1450.

The background of Cell 1455 indicates that the impact of the drive on the error rate is inconclusive because the tape was not used in any other drives. For example, Cell 1455 indicates that while an error rate of 1.473 errors per gigabyte occurred with the Tape 4/Drive 1 combination, but Tape 4 was not used in any other drives. It is therefore unclear what impact Drive 1 had on the error rate of Tape 4.

The background of Cell 1460 indicates that the tape would still have been of interest in the absence of a drive, but significantly less so. For example, Cell 1460 indicates that the error rate of Tape 8 is still of interest in the absence of Drive 1, but by less than half of when Drive 1 is considered.

Looking at report 1430, a user can quickly identify that Drives 1 and 2 may be failing and that Drive 3 is likely good. The state of Drive 4 is more uncertain because a number of tapes experienced above the threshold error rates in Drive 4, but most of those tapes would still have been of interest in the absence of Drive 4. Similarly, the user can easily identify potentially failing Tapes. For example, the user can see that Tape 1 experienced errors in a number of drives, including at least one drive that is otherwise considered good. Therefore, the user can easily identify Tape 1 as failing. The user can do a similar review of the other tapes to determine whether to check or remove tapes.

The visual indicators provided above are provided by way of example and not limitation. The visual indictors can include icons, color coding, text and/or visual indicators that correspond to the drive impact data for the tape/drive combination. The visual indicators may also convey other information, including, but not limited to whether all the operations performed on a tape for a particular Tape/Drive combination were READ errors or WRITE errors, or other information. While, in the above example, drive impact data is presented to a user for review, in other embodiments the drive impact data can be analyzed by a program according to a set of criteria to identify failing tapes and or drives.

FIG. 18 is a flow chart illustrating one embodiment of a method for identifying drive failure and media failure in the context of a tape media library. At step 1500 a threshold for an acceptable error rate can be established. The threshold applied for a tape can depend on both the drive technology of the drive in which the tape was loaded, the device manufacturer or other factors. The error rates can be correlated to tape/drive combinations through tape tracking to give an error rate for each tape in a particular tape/drive combination (step 1505). Tape/drive combinations that had error rates greater than the threshold can be identified (step 1510) and the highest error rate for each tape in the identified combinations can be determined (step 1515). For each tape/drive combination identified as having an error rate greater than the threshold, the question remains as to whether the tape or drive should be suspected of causing the error rate.

According to one embodiment, to determine if a drive is suspect of causing the errors for a tape/drive combination, the drive of that combination is removed from determination of the highest error rate for the tape (step 1520). If the highest error rate for the tape drops below the threshold, as determined at step 1525, the drive can be considered as being suspect (step 1530) because the tape would no longer have been of interest. Otherwise the drive can be identified as not having a sufficient impact on the error rate of the tape (step 1535). Steps 1520-1535 can be repeated for each drive in which a tape was loaded and for each tape having experienced a greater than the threshold error rate.

It should be noted that the impact of a drive on the error rate of a tape can be otherwise categorized to identify, for example, whether a drive is suspected as being the cause of errors and the level of certainty. For example, if it is determined that the highest error rate for a tape drops below the threshold in step 1525, but the tape was only loaded in one drive, the impact of the drive on the error rate is inconclusive. As another example, if the highest error rate for a tape is still greater than the threshold, as determined at step 1525, but there is a sufficient drop in the highest error rate of the tape, the drive can be considered suspect with a reduced level of certainty. In some embodiments, the drive can be considered more suspect the greater the drop or greater the percentage drop in the error rate for the tape.

A report can be output that visually identifies the impact of drives on the error rates of tapes (step 1540). The visual indicators can depict the level of certainty that a drive is suspected of causing errors or other information. According to one embodiment, the report can be formatted as a chart with cells for each tape/drive combination being considered. Each cell can include a visual indication of whether the tape would have been of interest if it had not been loaded in that drive. The chart can be formatted in manner that allows a user to easily identify potentially failing tapes and drives.

FIG. 19 is a diagrammatic representation of a management station 1600 that can include a processor 1602, such as an Intel Pentium 4 based processor (Intel and Pentium are trademarks of Intel Corporation of Santa Clara, Calif.), a primary memory 1603 (which may include, for example, RAM, ROM, Flash Memory, EEPROM or other computer readable medium known in the art) and a secondary memory 1604 (which may include, for example, a hard drive, disk drive, optical drive or other computer readable medium known in the art). A memory controller 1607 can control access to secondary memory 1604. Management station 1600 can comprise a communications interface 1606 (which may comprise, for example, fibre channel interface, Ethernet port or other communications interface known in the art) to connect management station to network 1650. An I/O controller 1612 can control interactions with network 1650. Similarly, an I/O controller 1614 can control interactions over I/O interfaces 1608 and 1610. Management station 1600 can include a variety of input devices. Various components of management station 1600 can be connected by a bus.

Management station 1600 can include a set of instructions 1655 that are executable to retrieve data 1660 from a data storage medium 1665, such as a repository of media library data, error data, local storage or other data storage. Instructions 1655 can be executed to analyze data 1660 to determine a set of drive impact data and generate reports to users. In some embodiments, instructions 1655 can also process drive impact data to identify failing media and/or drives. Instructions 1655 can also be implemented on a monitoring appliance or be distributed. Management station 1600 can be any suitable computer system that can execute instructions 1655 as software or firmware or can include an ASIC or other hardware to perform operations.

Embodiments can also be implemented with respect to libraries of media other than magnetic tapes. For example, the library can comprise a plurality of optical disks (i.e., an optical jukebox) or removable hard drives. Other libraries can utilize a combination of different storage media such as hard drives, magnetic media or optical media. Furthermore, the error rates analyzed can include, but are not limited to, hard errors, soft errors, a combination thereof or other error rates.

While systems and methods been described with reference to particular embodiments, it should be understood that the embodiments are illustrative and that the scope of the invention is not limited to these embodiments. For example, while embodiments described above have been described with regard to RES, LS and Inquiry commands, this is by way of illustration and not limitation. In some embodiments, other commands may be used to collect data associated with a library or library components. Many variations, modifications, additions and improvements to the embodiments described above are possible. It is contemplated that these variations, modifications, additions and improvements fall within the scope of the invention as detailed in the following claims.

Claims

1. A method of identifying failing media library components, comprising: querying one or more library components of a media library at intervals, wherein querying one or more library components comprises: sending first commands to a drive of the media library over a network at intervals,sending one or more second commands to the drive or a media changer of the media library over the network,sending third commands to the media changer over the network at intervals, andcollecting data returned in response to the first commands, the second commands or the third commands in a repository; andidentifying a potentially failing tape or a drive based on the collected data by determining for each drive of a set of tape/drive combinations experiencing error rates of interest whether a drive is considered as being suspect of having an impact on an error rate of a tape of the set of tape/drive combinations based on a change in a highest error rate for the tape in the identified set of the tape/drive combinations upon removing the drive of the set of tape/drive combinations.

2. The method of claim 1, further comprising: identifying the set of tape/drive combinations as experiencing the error rates of interest by identifying each tape/drive combination of the set of tape/drive combinations as having an error rate greater than a threshold error rate for determining whether a tape or a drive is causing the error rate.

3. The method of claim 2, further comprising: (a) removing the drive of a tape/drive combination of the set of tape/drive combinations from determination of a highest error rate for the tape of the tape/drive combination,(b) If the highest error rate for the tape drops below the threshold error rate, the drive of the tape/drive combination is considered as being suspect of having an impact on the error rate of the tape of the tape/drive combination, and(c) repeating the steps (a) and (b) for each drive of the set of tape/drive combinations in which a tape was loaded and for each tape of the set of tape/drive combinations having experienced the error rate greater than the threshold error rate.

4. The method of claim 1, wherein sending the first, second or third commands further comprises: issuing SCSI commands at the intervals and receiving the returned data.

5. The method of claim 4, further comprising: determining the set of tape/drive combinations, and error rates for the set of tape/drive combinations, from the returned data;identifying a tape in the set of tape/drive combinations that experienced an error rate greater than a first threshold error rate;determining drive impact data for the tape that indicates whether the tape would still have experienced greater than the first threshold error rate had that tape not been loaded in one or more drives in which the tape was loaded; andfor a drive in the set of tape/drive combinations in which the tape was loaded, identifying whether the drive is suspected of causing the tape to have experienced greater than the first threshold error rate based on the drive impact data.

6. The method of claim 1, further comprising: displaying a first visual indicator associated with the drive in response to determining that the tape would not have experienced greater than the first threshold error rate had the tape not been loaded in that drive and displaying a second visual indicator associated with the drive in response to determine that the tape would have experienced greater than the first threshold error rate had the tape not been loaded in that drive.

7. The method of claim 1, wherein sending the first commands further comprises: sending Log Sense (LS) commands to the drive of the media library over the network at the intervals; andand receiving the returned data.

8. The method of claim 1, wherein sending the one or more second commands further comprises: sending one or more Inquiry commands to the drive or the media changer of the media library over the network; andand receiving the returned data.

9. The method of claim 1, wherein sending the third commands further comprises: sending Read Element Status (RES) commands to the media changer over the network at the intervals; andand receiving the returned data.

10. A media library monitoring system comprising: a processor;a memory; andcomputer executable instructions stored on the memory and executable to perform the steps of:querying one or more library components of a media library at intervals, wherein querying one or more library components comprises: sending first commands to a drive of the media library over a network at intervals,sending one or more second commands to the drive or a media changer of the media library over the network,sending third commands to the media changer over the network at intervals, andcollecting data returned in response to the first commands, the second commands or the third commands in a repository; andidentifying a potentially failing tape or a drive based on the collected data by determining for each drive of a set of tape/drive combinations experiencing error rates of interest whether a drive is considered as being suspect of having an impact on an error rate of a tape of the set of tape/drive combinations based on a change in a highest error rate for the tape in the identified set of the tape/drive combinations upon removing the drive of the set of tape/drive combinations.

11. The system of claim 10, further comprising: a monitoring appliance configured to perform the sending first commands to a drive of the media library over a network at intervals, sending one or more second commands to the drive or a media changer of the media library over the network, sending third commands to the media changer over the network at intervals and receiving of the returned data, the monitoring appliance further configured to store the returned data in the repository.

12. The system of claim 11, further comprising: a management station configured to perform the determining for each drive of the set of tape/drive combinations experiencing error rates of interest whether the drive is considered as being suspect of having the impact on the error rate of the tape of the set of tape/drive combinations based on the change in the highest error rate for the tape in the identified set of the tape/drive combinations upon removing the drive of the set of tape/drive combinations.

13. The system of claim 10, wherein the computer instructions comprise a data engine and a storage engine, the data engine configured to provide the returned data to the storage engine in a structured format and the storage engine is configured to read the data in the structured format and store the returned data in the repository.

14. The system of claim 13, wherein the data engine is further configured to format SCSI data as an XML document and provide the XML document to the storage engine.

15. The system of claim 14, wherein the storage engine is configured to unpack the XML document and store the returned data as database records.

16. A non-transitory computer program product comprising a computer readable storage medium storing a set of computer executable instructions, the computer executable instructions comprising instructions executable to perform the steps of: querying one or more library components of a media library at intervals, wherein querying one or more library components comprises: sending first commands to a drive of the media library over a network at intervals,sending one or more second commands to the drive or a media changer of the media library over the network,sending third commands to the media changer over the network at intervals, andcollecting data returned in response to the first commands, the second commands or the third commands in a repository; andidentifying a potentially failing tape or a drive based on the collected data by determining for each drive of a set of tape/drive combinations experiencing error rates of interest whether a drive is considered as being suspect of having an impact on an error rate of a tape of the set of tape/drive combinations based on a change in a highest error rate for the tape in the identified set of the tape/drive combinations upon removing the drive of the set of tape/drive combinations.

17. The computer program product of claim 16, further comprising instructions executable to perform the steps of: identifying the set of tape/drive combinations as experiencing the error rates of interest by identifying each tape/drive combination of the set of tape/drive combinations as having an error rate greater than a threshold error rate for determining whether a tape or a drive is causing the error rate.

18. The computer program product of claim 17, further comprising instructions executable to perform the steps of: (a) removing the drive of a tape/drive combination of the set of tape/drive combinations from determination of a highest error rate for the tape of the tape/drive combination,(b) If the highest error rate for the tape drops below the threshold error rate, the drive of the tape/drive combination is considered as being suspect of having an impact on the error rate of the tape of the tape/drive combination, and(c) repeating the steps (a) and (b) for each drive of the set of tape/drive combinations in which a tape was loaded and for each tape of the set of tape/drive combinations having experienced the error rate greater than the threshold error rate.

19. The computer program product of claim 16, further comprising instructions executable to perform the steps of: issuing SCSI commands at the intervals and receiving the returned data.

20. The computer program product of claim 19, further comprising instructions executable to perform the steps of: determining the set of tape/drive combinations, and error rates for the set of tape/drive combinations, from the returned data;identifying a tape in the set of tape/drive combinations that experienced an error rate greater than a first threshold error rate;determining drive impact data for the tape that indicates whether the tape would still have experienced greater than the first threshold error rate had that tape not been loaded in one or more drives in which the tape was loaded; andfor a drive in the set of tape/drive combinations in which the tape was loaded, identifying whether the drive is suspected of causing the tape to have experienced greater than the first threshold error rate based on the drive impact data.