FR2867870A1 - Computer device, has monitoring module to monitor quality of operation of mass storage peripherals, and processing module to calculate and record history of quality parameters from collected and recorded activity data - Google Patents

Abstract

The device has mass storage peripherals (2) with a read and write device for receiving a removable storage device. The storage peripherals generate a called activity data, which represents its use and operation. A monitoring module monitors quality of operation of the peripherals. A processing module (7) calculates and records a history of quality parameters from the collected and recorded activity data.

Description

COMPUTER DEVICE WITH PERIPHERAL (S) OF MEMORY OF MASS

  The invention relates to a computing device comprising at least one mass storage device comprising at least one reader and / or recorder device adapted to receive at least one removable memory medium. For example, a mass storage device is formed of a reader device and / or tape recorder, or optical disk, or diskette ...

  These mass storage devices are used in particular for performing backups, performing data migrations between machines, or archiving sensitive data. It is therefore important that these mass storage devices have a perfectly reliable operation. It would be advantageous in this sense to be able to prevent a possible failure of such a device in order to avoid any loss of data. In addition, such a failure must be detected, diagnosed and reported to the user when it occurs during operation in its normal environment.

  However, to date, no known device can solve these problems simultaneously.

  For example, devices for diagnosing mass storage devices are known. To use these diagnostic devices, it is necessary to disconnect the mass memory device from its normal operating environment and associate it with the diagnostic device that performs a complete diagnostic of the various internal mechanisms and elements of the device. Nevertheless, such a diagnostic device can not monitor the device in operation and requires the use of a reader device and / or reference recorder or a reference removable memory medium to discriminate the origin of a failure.

  Moreover, in modern computing devices, it is sometimes provided backup and archive management software, which emit alarm messages only when a crippling failure occurs in reading and / or recording and / or in when the number of cycles of use of a removable memory medium is exceeded.

  In the same way, it is often provided on modern computing devices, general operation control modules of the computing device, which are responsible for centralizing the various status information of the constituent elements of the device and any error information or alarm messages provided by the various elements to the central unit. However, these modules are not suitable for specifically managing mass storage devices or controlling the quality of data that can be recorded on these devices. Nor do they make it possible to discriminate the origin of a possible failure, depending on whether it comes from the reader and / or recorder device or the removable memory medium, or even to prevent a failure on a mass storage device.

  The invention therefore aims to solve this general problem. It aims to propose a computer device in which it is carried out a preventive monitoring of at least one mass storage device in service.

  The object of the invention is more particularly to propose a computing device making it possible to prevent and / or detect any malfunction of a mass memory device and to determine if the origin of such a risk of dysfunction or of such a dysfunction comes from the removable memory medium and / or the reader and / or recorder device.

  More generally, the invention aims to provide a computing device that allows the user to have a thorough analysis of the performance of the various mass storage devices, without the need for disconnection or disabling. these mass storage devices, nor interfere with the normal operation of this computer system, and especially with the various software operating mass storage devices.

  The invention also aims at providing such a computing device that is simple, ergonomic, inexpensive to installation and use.

  To do this, the invention relates to a computing device 5 comprising: at least one mass storage device comprising at least one reader and / or recorder device, adapted to receive at least one removable memory medium, this memory peripheral mass being adapted to generate data, called activity data, representative of its use and / or operation, - at least one software application capable of performing read operations and / or recording with the device (s) (s) ) of mass memory in service, characterized in that it comprises: - at least one module for monitoring the performance of at least one mass storage device in service, said monitored device, this monitoring module being adapted to: detect for each monitored device whether at least one monitored device and / or device recorder receives a memo medium This monitored device is called an active monitored device, or on the contrary if it receives no removable memory medium, the monitored device is said to be a monitored device inactive, collect and record over time, for each monitored device active, activity data, called monitoring data, comprising data, called read quality and / or recording data, adapted to allow quality parameters to be calculated, whose evolution over time is representative of a drift of the quality of reading and / or recording operations by each monitored device, at least one processing module adapted to calculate and record, from the tracking data, a history of the quality parameters, said activity history, each monitored reader device and / or device recorder, this activity history being adapted to allow e to prevent and / or detect a malfunction of each monitored device.

  A computing device according to the invention may consist of one or more machine (s). It may be for example a single computer with its central unit comprising at least one processor and its peripherals connected to this central unit via a peripheral bus. It may also be just as well a plurality of computers connected to a network, or any other computer architecture provided with remote communication means or not between several machines and / or parts of machines.

  Advantageously and according to the invention, the device comprises a plurality of mass storage devices, and the monitoring module is adapted to be able to collect and record tracking data of several monitored devices.

  Advantageously and according to the invention, the activity history comprises at least one quality parameter chosen from the reading error rate and / or the recording error rate.

  Advantageously and according to the invention, the monitoring module is adapted to read in at least one memory of the computing device at least one identification data for each removable storage medium, said removable medium identified, received in each monitored device. In addition, the processing module is adapted to calculate and record further a history of quality parameters, said activity history, of each identified removable memory medium, with the activity history of each reader device and / or or recorder, to determine if the origin of this malfunction comes from the support (4) of removable memory and / or the device reader and / or recorder.

  Furthermore, advantageously and according to the invention, the activity history of each identified removable memory medium comprises at least one quality parameter chosen from the reading error rate and / or the recording error rate and / or the number of loading and / or unloading operations and / or the duration of use in a reader and / or recorder device.

  Furthermore, advantageously and according to the invention, the processing module is adapted to update a single centralized database comprising quality parameters of each device reader and / or monitored device recorder and the quality parameters of each support identified removable memory, this database forming the activity history of each reader device and / or recorder and each removable memory medium identified. Thus, the activity histories recorded in the form of a database make it possible to perform sorts, selections, various analyzes, making it possible to carry out a highly reliable preventive diagnosis of each reader and / or recorder device and of each support. removable memory used.

  Advantageously and according to the invention, the device further comprises at least one diagnostic module adapted for, from each activity history that it receives, to trigger an alarm event when at least one quality parameter takes a value corresponding to a risk of possible malfunction of the removable memory medium and / or a reader device and / or recorder.

  Advantageously and according to the invention, the processing module is adapted to calculate a value of at least one evolution parameter representative of the variation over time of a quality parameter. In addition, the diagnostic module is adapted to trigger an alarm event when at least one evolution parameter takes a value corresponding to a risk of possible malfunction of the identified removable memory medium and / or a reader device and / or recorder. A device according to the invention thus makes it possible, from the activity histories, to anticipate the possible failures of any monitored mass memory device, and to determine the possible origin of such a failure or a risk of failure (reader device and / or 2867870 6 recorder or removable memory medium).

  The device according to the invention thus makes it possible to monitor each active mass memory device in service, and to trigger an alarm even before a failure occurs. This avoids any loss of data.

  Advantageously and according to the invention, the diagnostic module is adapted to compare each evolution parameter with a predetermined threshold value and to trigger an alarm event when this threshold value is exceeded.

  In addition, advantageously and according to the invention, the diagnostic module is adapted to compare each quality parameter with a predetermined threshold value and to trigger an alarm event when this threshold value is exceeded.

  The diagnostic module advantageously provides the user with information corresponding to the alarm events, in particular in the form of an alarm message or an action message to be carried out.

  Advantageously and according to the invention, an alarm event comprises a message indicating at least one loading and / or unloading operation to be performed. Indeed, after detection of a probable malfunction, the diagnostic module may indicate to the operator is to place a removable memory medium in the reader device and / or recorder, or to place the removable memory medium previously used in a other device reader and / or recorder known operation. This simple operation and subsequent analysis with the help of the monitoring module, the processing module and the diagnostic module will make it possible to discriminate with certainty the origin of the probable malfunction previously detected. In addition, it should be noted that the loading and / or unloading operation can make it possible to avoid the occurrence of a fatal operating failure of the mass memory device concerned. This avoids any loss of data.

  The monitoring module of a device according to the invention is advantageously adapted to collect the monitoring data periodically, according to a predetermined period.

  Advantageously and according to the invention, this period is between 1s and 10min, in particular of the order of 1 min.

  As a variant, there is nothing to prevent the monitoring module from being itself controlled by another application, and for example an application for managing a pool of mass storage devices. Moreover, there is nothing to prevent the value of the monitoring period being adjustable either manually or automatically according to other parameters, for example the operating or charging rate of the mass storage device park.

  Furthermore, advantageously and according to the invention, the monitoring module is adapted to be able to transmit the monitoring data to the processing module, the processing module being adapted to calculate and record each activity history immediately after receiving these data from the processing module. monitoring.

  In addition, advantageously and according to the invention, the diagnostic module is adapted to run immediately after each record of an activity history by the processing module.

  Advantageously and according to the invention, each mass storage device comprises a local memory, said activity register, and at least one adapted controller to be able to record the activity data of these activity data in zones of the memory register. predetermined activity according to the nature of the activity data, and the monitoring module is adapted to read areas of the activity register corresponding to tracking data.

  The invention thus makes it possible to detect and permanently record the operating state of the various mass memory devices monitored and, based on the activity history, to prevent malfunctions, to detect them, and, most importantly, to know the origin of the malfunctions in a precise way.

  The invention also relates to a method of monitoring mass storage peripherals implemented in a computing device according to the invention.

  In a monitoring method according to the invention, the tracking data are collected and recorded and from this tracking data is constituted a history of activity of each identified removable memory medium and of each reader device and / or or recorder of each monitored device.

  The invention also relates to a device and a method characterized in combination by all or some of the characteristics mentioned above or below.

  Other features, objects and advantages of the invention will appear on reading the following description which refers to the appended figures in which: FIG. 1 is a diagram illustrating a first variant embodiment of a computing device according to FIG. FIG. 2 is a diagram illustrating a second variant embodiment of a computing device according to the invention, FIG. 3 represents an exemplary algorithm of a monitoring module according to the invention, FIG. example of an algorithm of a processing module according to the invention, - Figure 5 shows an example of an algorithm of a diagnostic module according to the invention, - Figure 6 shows a graph illustrating the evolution during the time of a quality parameter according to the invention, ie a rate of errors in writing ie attributable to a removable memory medium, - Figure 7 represents a graph illustrating the evolution over a period of time. the time of an evolution parameter according to the invention, ie a variation Ve over the time of the error rate ie of FIG. 6.

  The computing device 1 (shown in FIG. 1) comprises a plurality of mass memory peripherals 2, each of these mass memory peripherals 2 comprises at least one reader and / or recorder device 3 adapted to receive a removable memory support 4. at a time.

  The removable memory media 4 may for example be magnetic tape cassettes 4a, 4b or coils 4c, or 4d optical disc type (CD-Rom, DVD ...), or floppy 4th or card type electronic memory (not shown) or other. Each of the removable memory media 4 comprises an identifier, this identifier being either recorded during formatting and / or first writing on the removable memory medium 4 so that it can be read by the reader and / or recorder device 3 or written in the form of a code arranged on the support 4 and which can be grasped and / or read during the loading, so that it can be automatically communicated to the computing device 1, is again allocated by the computing device 1 according to the location in which the support 4 is stored.

  Each mass storage device 2 is connected to a peripheral bus 14 of the computing device 1. The computing device may comprise a plurality of peripheral buses 14, each of which may be connected to a plurality of mass memory peripherals 2. .

  Each mass storage device 2 is adapted to generate data, called activity data, representative of its use and / or operation. For each type of mass memory device 2, at least some of these activity data are defined and normalized. These include the activity data taken up by the associated device bus definition standard 14. For example, the SCSI standard defines a number of activity data.

  Each memory device 2 preferably comprises a local memory, called activity register 8, for each device 3 reader and / or recorder of this mass memory device 2. This activity data is generated and recorded in this activity register 8, in particular data representative of the operation and / or use of this reader and / or recorder device with at least one removable memory support 4. . The mass memory device 2 further comprises at least one controller 13 associated with the activity register (s) 8 in order to enable the generation and the recording of the activity data in this (these) activity register (s) 8.

  The activity data is recorded using the controller (s) 13 in different memory areas of each activity register 8 according to the nature of the activity data. Thus, reading each memory area, specific to each category of activity data, makes it possible to retrieve the activity data by knowing their exact nature.

  Nothing prevents a mass memory device 2 from including only one activity register 8 for a plurality of reader and / or recorder devices 3 when data recovery means are provided. activities to identify the device 3 reader and / or recorder of which these activity data are each issues.

  Preferably, according to the first embodiment of the invention, each mass storage device 2 comprises only one reader and / or recorder device 3, a single activity register 8 and a single controller 13 associated with it. activity register 8.

  The controller 13 of each memory device 2 is adapted to be able to receive read commands from these memory areas via the peripheral bus 14 to which it is connected. The controller 13 of the mass memory device 2 is further adapted to provide on this device bus 14, the requested activity data. Each mass storage device 2 may be, for example, a SCSI-compliant mass memory device 2. This SCSI standard provides an architecture and a set of commands allowing access to activity data of the mass storage peripherals 3 via a bus 14 of SCSI devices to which they are connected.

  The invention is also applicable to other types of bus 14, for example according to the ATA or SATA or FC (Fiber Channel) or ESCON or FICON or other standards. The invention is applicable to memory peripherals 2 connected to peripheral buses 14, when reading quality and / or recording quality data, as described below, are generated by these peripherals 2 and that these data are accessible via this bus 14.

  The computer device 1 according to the invention comprises at least one monitoring module 6 adapted to be able to launch read commands for activity data on at least one peripheral bus 14 to at least one mass storage device 2 in use, said device 2 monitored, and read this data via this device bus 14 in order to save them in a memory.

  According to the first embodiment, the computing device 1 (represented in FIG. 1) furthermore comprises, in the traditional way, at least one central unit 5 equipped with at least one processor, at least one random access memory and at least one an input-output controller (not shown). The central unit 5 further comprises an internal bus (not shown) connecting the processor with the RAM and the input-output controller. The peripheral bus 14 is for example a SCSI type bus and is connected to a SCSI controller 12. The SCSI controller 12 is itself connected to the input-output controller via an output bus 23 so that it can be controlled by the central unit 5 via a driver software of the SCSI bus loaded in the RAM of the unit. central 5. The computer device 1 is furthermore equipped in the traditional way with an operating system such as WINDOWS, UNIX, LINUX ... The driver software of the SCSI bus is linked to the operating system and is simultaneously launched at this time. last when starting the computing device 1.

  In the first embodiment, the computing device 1 comprises a single computer 24 (symbolized in dotted lines). Each mass memory device 2 is supplied with electricity and connected to the bus 14 so as to be seen by the computer 24. The computer 24 comprises the central unit 5 in particular, as well as the traditional components of such a computer, such as a non-removable mass memory (not shown), for example a hard disk, a human / machine interface (keyboard, pointer, screen, etc.) and the associated peripheral cards (not shown) connected to the input bus 23 via bus and traditional controllers used. The device bus 14 being connected to this computer 24 via the SCSI controller 12, this single computer 24 controls all the mass storage peripherals 2 associated with this bus 14. In addition, this single computer comprises a software application (not shown). loaded in RAM and carrying out operations of reading and / or traditional recording with these devices 2 of mass memory in service.

  The computing device 1 comprises at least one monitoring module 6, at least one processing module 7 and at least one diagnostic module. According to the first embodiment, the computer device 1 comprises a single monitoring module 6. This monitoring module 6 can be a computer program loaded into the RAM of the central unit 5. This monitoring module 6 can thus launch, via the driver software of the SCSI bus, commands to the monitored devices 2 to determine if the It is loaded with a support 4 of removable memory, this memory device 2 then being said active. This can be done for example using the TEST UNIT READY command. If not, this device 2 is said to be inactive. Similarly, the monitoring module can initiate read commands to the active devices 2 in order to collect activity data from these active devices 2. This can be done for example using the LOG SENSE command.

  These collected and recorded activity data, called tracking data, comprise read and / or write quality data from which can be calculated and / or obtained quality parameters as described below. This reading and / or writing quality data comprises, for example: the total duration of use (actuation) of an internal mechanism of the device 3 reader and / or recorder of the monitored device, allowing the drive of the removable memory medium for reading and / or recording, - the total power-on time of monitored device 2, - the number of uncorrected read and / or write errors since the last activity data collection, - the number of corrected read and / or write errors since the last activity data collection, - the number of bytes written and / or read since the last activity data collection.

  In addition, the tracking data preferably includes identification data of the removable memory medium 4 loaded on an active device 2 and identification data of the reader and / or recorder device 3 on which this medium 4 is loaded. removable memory. In the case where the device 2 active comprises only one device 3 reader and / or recorder, the identifier of this device 3 reader and / or recorder corresponds to the identifier of said device 2 active.

  In practice, the tracking data generally comprise data enabling the identification of the monitored devices 2 as well as the devices 3 reader and / or recorder of these monitored peripherals 2. However, monitored device 2 often does not produce activity data for identifying memory medium 4 loaded on memory device 2. This information can then be obtained other than by interrogating the register (s) 8 of the memory device. This information can be, for example, provided by the user using the man / machine interface 25.

  In an alternative embodiment, the monitoring module 6 identifies a memory support 4 loaded into an active device 2 by interrogating a memory of a robotic arm (not shown) carrying out operations for loading and unloading the supports 4 of removable memory in the mass storage devices 2 which he has memorized a physical position and an identifier. Such a robotic arm generally comprises an optoelectronic reading code reader, for example barcodes for entering codes arranged on each support 4 of removable memory. The codes entered make it possible to identify in the memory of the robotic arm each memory support 4 loaded in a reader and / or recorder device 3. The robotic arm preferably comprises an interface making it possible to connect it to a peripheral bus 14 of the computing device 1. The monitoring module 6 can thus interrogate the memory of the robotic arm in order to obtain the identifier of a removable memory support 4. loaded into a device 2 identified by the monitoring module 6.

  The tracking data may further comprise data, called alert data signaling a malfunction of the active device 2. This data is for example the TAPE ALERT data provided by the SCSI standard for certain types of peripherals. In the case where the memory device 2 is a SCSI-type device, the warning data signal in particular: - an abnormal internal temperature, - an abnormal internal moisture content, - a rupture of the strip (in the case where the storage medium 4 is of the magnetic strip type). The monitoring module 6 can carry out these collection and recording operations for each device 3 and / or device recorder 2 monitored according to the algorithm represented in FIG. 3. In this example, the computer device 1 comprises a robotic arm connected to a peripheral bus 14 so as to be able to perform the loading and unloading of the removable memory media 4 and to be able to identify each support 4 of removable memory loaded in each peripheral 2 active as described above. This robotic arm can be, for example, of the type included in a library of magnetic tapes (in English Autoloader) such as those sold by STORAGETEK (LOUISVILLE, COLORADO, USA), QUANTUM ATL (SAN JOSE, CALIFORNIA, USA), OVERLAND (SAN DIEGO, CALIFORNIA, USA) ...

  Step 100 firstly initiates playback commands to the monitored device 2 to obtain identification data of this mass memory device 2: - the serial number of the mass storage device 2, - the mass memory device 2 manufacturer's identity, says VENDORID, - the mass memory device 2 model, called PRODUCTID.

  The subsequent step 101 consists in making a comparison of the model identifier (using the PRODUCTID and VENDORID data) obtained with a list, of which the monitoring module 6 is equipped, of mass memory device models 2. supported. If the model of mass storage device 2 is supported, step 103 is executed, otherwise end step 102 is executed.

  Step 103 consists in waiting for a given period, referred to as the loading period of support 4 of removable memory. This period is preferably set by the user.

  The next step 104 is to detect whether the monitored device 2 is active, i.e. if the reader and / or recorder device is loaded with a removable memory medium 4.

  If the monitored device 2 is inactive, that is to say if the device reader and / or recorder is not loaded with a support 4 of removable memory, it goes back to step 103 prior to waiting. If a memory medium 4 is detected in the monitored device, step 105 is performed. The subsequent step 105 is to create a file, called LOGFILE, for tracking data record of the monitored monitored device 2. Step 105 further comprises recording, in the LOGFILE file, the identification data of step 100 and initiating a robotic arm command to identify the removable memory medium 4 loaded in the active device 2. The identifier of this memory medium 4 is then saved in the LOGFILE file.

  The subsequent step 106 is to execute read commands of the active activity device 2 activity register 8 memory areas in order to collect the following tracking data: - the sum of the write-corrected errors since the last data collection follow-up, says COR WRITE, - the sum of uncorrected errors in writing since the last collection of monitoring data, says UNCOR WRIT, - the sum of bytes written since the last collection of monitoring data, says BYTES WRITTEN, - the sum of the read-corrected errors since the last tracking data collection, says CORRE READ, - the sum of the uncorrected read errors since the last tracking data collection, says UNCOR READ, - the sum of bytes read since the last collection of tracking data, says BYTES READ, - the TAPE ALERT (SCSI standard for tape drive) at the time of this data collection of the follow-up, says TAPE ALERT FLAGS.

  If no error is detected by the subsequent test 107 when ordering reading of these memory areas, step 108 records the tracking data read in step 106 in the LOGFILE file. This tracking data is associated when saved in the LOGFILE file to a collection number, called LINE, and the date and time of that collection, says DATE. These values LINE and DATE are generated by the monitoring module 6.

  The subsequent step 109 consists in waiting for a given duration, called the collection period, at the end of which the reading step 106 and the subsequent steps for a new collection of the monitoring data are repeated. The collection period is preferably set by the user.

  If an error is detected by the test 107 during the reading of the memory areas of the activity register 8, the test 110 is executed which makes it possible to determine whether this error is an error of absence of support 4 of removable memory in the memory. 2 mass memory device, such an error indicating that an unloading of the removable media 4 memory mass has occurred. If this is the case, the LOGFILE file is closed in step 112.

  Step 112 also consists in collecting and recording in the 5 LOGFILE files, before closing this file, the following reading and / or writing quality data: the total duration of powering up of the mass memory device 2 , said POWER_ON_HOURS, - the total duration of use (actuation) of the internal mechanism of the device 3 reader and / or recorder of the mass memory device 2, called TAPE MOTION HOURS.

  The LOGFILE file is then communicated to the processing module 7 at the subsequent step 113. The monitoring module 6 then repeats step 103 and the subsequent steps for the creation of a new LOGFILE representing a cycle of use of a memory medium 4 on the monitored device 2.

  If an error is detected during the reading of the activity register memory areas in step 107, and this error is not an error of absence of mass memory support 4, as determined by the In test 110, a test 111 is performed to determine whether a maximum threshold value of the number of permissible read errors of the memory areas of the register 8 is exceeded or not.

  If not, proceed to step 112. If yes, execute step 109.

  The processing module 7 is, according to the first embodiment of the invention, a program loaded into the random access memory of the single computer 24. The processing module 7 is adapted to keep updated, from the LOGFILE files created by the monitoring module 6, an activity history for each identified memory medium 4 and an activity history for each device 3 reader and / or device recorder 2 monitored. Each of these activity histories can be realized by a table of a database of the single computer 24. This database 15 can be recorded on the hard disk of the single computer 24. In addition, this database 15 can be managed using a database management software such as MYSQL, ORACLE or other.

  Preferably, the processing module 7 performs an update of the activity history of a removable memory medium 4 and the activity history of a reader and / or recorder device 3 following a cycle. of using the loading or unloading of this removable memory medium 4 on this device 3 reader and / or recorder, that is to say when the monitoring module 6 provides the processing module 7 LOGFILE file from this cycle use.

  The processing module 7 additionally makes it possible to calculate, from each LOGFILE file produced by the monitoring module 6, quality parameters based on read quality data and / or record the LOGFILE file. These quality parameters are calculated for each LOGFILE file and recorded in the activity history of the memory medium 4 identified by the LOGFILE file and in the activity history of the device 3 reader and / or recorder in order to detect a possible drift of these quality parameters over time.

  The quality parameters are, for example: the total duration of the power up of the monitored device 2 associated with a device 3 reader and / or recorder, the total duration of use (actuation) of a support 4 of removable memory by the internal mechanisms of the supervised peripherals 2, the total number of cycles of use of removable memory medium 4 made on a reader and / or recorder device, the total number of cycles of use of a support 4 of removable memory, - the average write error rate Te obtained during a cycle of use of a removable memory medium on a device 3 reader and / or recorder, - the average error rate in reading Te obtained during a cycle of use of a support 4 of removable memory on a device 3 reader and / or recorder, - a quality index Qe representative of the write performance of a device 3 reader and / or recording by compared with other devices 3 monitored reader and / or device recorder 2, - a quality index Qe representative of the quality of a device 3 reader and / or recorder compared to other devices 3 scanned readers and / or device recorders 2 the error rate ie attributable to the removable memory medium 4 during a duty cycle, the imputable read-over error rate of the removable memory medium during a duty cycle.

  The processing module 7 is further adapted to calculate evolution parameters representative of the variation over time of a quality parameter from the activity histories.

  These quality parameters are, for example: the variation Ve of the write error rate ie between two cycles of use of a removable memory medium 4, the variation of the error rate between reading two cycles of using a removable memory medium 4.

  The values of the error rates Te, Te, ie and i.e, variations Ve and V. Q as well as quality indices Qe and Q.e are preferably calculated by the processing module 7 for each LOGFILE file. These data are, after their calculation, recorded in the activity history of the removable memory medium 4 and the corresponding reader and / or recorder device 3.

  For example, the transfer module 7 can calculate the average write error rate Te using reading and / or recording quality data:

NB LINE

E COR_WRITE (LINE) LINE = 1 Te =

NB LINE

  E BYTES_WRITTEN (LINE) LINE = 1 where NB_LINE is the total number of collections performed for the usage cycle.

  In addition, the processing module 7 can be adapted to calculate the read error rate Tt using reading quality data and / or recording:

NB ONE

E CORRE READ (UNE) LINE = 1

NB ONE

  E BYTES_READ (LINE) LINE = 1 The activity history of each removable memory medium 4 preferably comprises a recording for each cycle of use of this medium 4 by a reader and / or recorder device 3. Also, each record of an activity history of a removable memory medium 4 preferably comprises: the date corresponding to this cycle of use of the removable memory medium 4, the total number of cycles of use of this memory medium 4; removable memory support 4 made on the date corresponding to a duty cycle, the identifier of the device 3 reader and / or recorder corresponding to this duty cycle, - the total number of bytes read during this cycle of use the total duration of actuation of the strip 4 of the removable memory medium 4 by the internal mechanism of the reader and / or recorder device 3 at the date corresponding to this cycle of use (for the support 4 of removable memory at magnetic tape), - the average read error rate Te for this cycle - the average write error rate Te for this cycle - the error rate tie for this duty cycle, - the error rate for this cycle of use, - the variatio n W for this duty cycle, - the variation Ve for this duty cycle, the total number of cycles of use of a support 4 of removable memory can be obtained by incrementing its value recorded in the history of activity for the previous usage cycle.

  In addition, the total duration of actuation of the strip of the support 4 of removable memory can be simply obtained from the data TAPE MOTION HOURS LOGFILE file, the history of activity of this support 4 removable memory and the activity history of the device 3 reader and / or recorder identified by the LOGFILE file.

  The activity history of each reader and / or recorder device preferably comprises a record for each cycle of use of a removable memory medium holder 4 on this reader and / or recorder device. Also, each record of an activity history of a reader and / or recorder device preferably comprises: the date corresponding to this cycle of use of a removable memory medium support on this reader and reader device; / or recorder, - the total number of cycles of use having been made on this device 3 reader and / or recorder on the date corresponding to this cycle of use, the identifier of the memory medium 4 used on this reader device and and / or recorder for this cycle of use, of use, of use, 2867870 22 - the total number of bytes read by the device 3 reader and / or recorder during the cycle of use, - the total number of bytes recorded by the device 3 reader and / or recorder during the cycle of use, - the total duration of the power on device 3 reader and / or recorder on the date corresponding to this cycle of use, - the duration of total use ( the internal mechanism of the reader and / or recorder device 3, the average read T error rate for this duty cycle, the average write error rate T 0 for this duty cycle, the Qe quality index for this cycle of use, - the Qe quality index for this cycle of use.

  The total number of cycles of use of media 4 of removable memory on a device 3 reader and / or recorder can be obtained by incrementing its value recorded in the activity history for the previous cycle of use.

  The quality index Qe of each device 3 reader and / or recorder can be obtained by comparing the average error rate values Te obtained during the cycles of use of a plurality of removable memory media 4 on this device 3 reader and / or recorder with the average error rates Te obtained with these same media 4 of removable memory on the other devices readers and / or recorders of the computing device. Several statistical operators such as the average, the standard deviation and others make it possible to perform such a comparative analysis of the write performance of this reader and / or recorder device compared with the other readers and / or recorders of monitored devices 2, using the average rates of errors Te mentioned above.

  For example, the processing module 7 is preferably adapted to be able to calculate, from the activity histories of each reader and / or recorder device 3, the mean e of the average write error rates Te obtained by the devices 3. readers and / or recorders of the computing device 1 during the use cycles of media 4 of removable memory on these devices 3 readers and / or recorders. A quality index Qe was assigned to each of these devices 3 reader and / or recorder by comparing the mean average error rates Te obtained by this device 3 reader and / or recorder with the overall average i 'e. The value of the quality index Qe is, for example, the higher the average of the mean error rates Te of the reader and / or recorder device 3 is relative to the value T e.

  The quality index Q.e can be obtained, using the calculated average T2 error rates, similarly to the quality index Qe by the treatment module 7.

  Other monitoring data Te and T2 could be used to calculate quality indices.

  In addition, the error rate Te can be obtained according to the following equation: the following equation: tie = T2xQ Similarly, the error rate tie can be obtained according to ie = Te x Qe Moreover, the variation Ve can be obtained according to the equation Ve = _te te 'where i' corresponds to the calculated rate i for this support 4 of removable memory during the previous cycle of use of this support 4.

  The variation W can be obtained similarly to the variation Ve. The processing module 7 is adapted to perform the creation and updating of activity history according to the algorithm illustrated in FIG. 4 each time a LOGFILE file is received.

  Step 401 consists of reading the identification data of the LOGFILE file in order to identify the reader and / or recorder device 3 as well as the removable memory support 4 having been loaded into this reader and / or recorder device.

  If the database 15, interrogated by the processing module 7 during the test 402, does not include an activity history table corresponding to the device 3 reader and / or recorder, a table is created for this device 3 reader and / or recorder in step 403 in the database 15. The test step 404 is then executed.

  If the database 15, interrogated by the processing module 7 during the test 404, does not include an activity history table corresponding to the removable memory support 4, a table is created for the removable memory support 4 to step 405 in the database 15. Step 406 is then executed.

  Step 406 consists of calculating the quality parameters, in particular Te and Te, from the tracking data of the LOGFILE file.

  The subsequent step 407 consists in recording the tracking data of the LOGFILE file as well as the quality parameters, in particular Te and Te, calculated in step 406 in the history table of the reader and / or recorder device 3. the date indicated by the LOGFILE file.

  The subsequent step 408 consists in recording quality parameters, in particular Te and Te, calculated in step 406 in the history table of the removable memory support 4 according to the date printed by the LOGFILE file.

  The subsequent step 409 consists in calculating a new value for Qe, Qe, Ve, Ve, ie, and tie from the activity history tables of the reader and / or recorder device 3 and the removable memory medium 4.

  The subsequent step 410 consists of recording the values of Qe and Qe calculated in step 409 in the activity history table of the reader and / or recorder device 3 and recording the values Ve, Ve, ie and tie calculated. in step 409 in the activity history table of the removable memory medium 4.

  According to the first embodiment, the computer device 1 comprises a diagnostic module 9. This diagnostic module can be a program loaded into the RAM of the single computer 24 so that it can access the base. data 15 and the tables of this database 15 constituting the activity history of the removable memory media 3 and devices 3 readers and / or recorders of the computer device 1. The diagnostic module is adapted to detect a malfunction or a risk of malfunction of a support 4 of removable memory or a device 3 reader and / or recorder. In addition, the diagnostic module 9 generates alarm events (described below) for each malfunction or each risk of malfunction detected.

  The diagnostic module 9 preferably executes the algorithm illustrated in FIG. 5 for each LOGFILE file created by the monitoring module 6 and after this LOGFILE file has been processed by the processing module 7.

  This algorithm makes it possible to perform a diagnosis of the removable memory support 4 and the reader and / or recorder device 3 identified in the LOGFILE file from their respective activity histories recorded in the database 15 by the processing module 7.

  Step 501 consists in: determining, for each TAPE ALERT FLAG tracking data item, the activity history of the reader and / or recorder device 3, if one of these tracking data takes a predetermined value signaling a malfunction, - generating an alarm event corresponding to each reported malfunction.

  The subsequent step 502 consists in: determining, for each quality parameter, the activity histories of the reader and / or recorder device and the removable memory support device, if this quality parameter exceeds a warning threshold value set for this quality parameter, this average threshold value being representative of a risk of malfunction of the removable memory medium 4 or reader and / or recorder device associated with this quality parameter, - to generate an alarm event for each risk of malfunctioning detected.

  The subsequent step 503 consists in determining whether one of the quality indices Qe or Qe of the activity history of the reader and / or recorder device exceeds a warning threshold value set for these quality indices, this threshold value being representative of a risk of malfunction of the device 3 reader and / or recorder.

  If the warning threshold value of step 503, test step 504 is executed. This step 504 consists in determining whether one of the quality indices Q.e or Qe exceeds an alarm threshold value set for these quality indices, this threshold value being representative of a malfunction of the reader and / or recorder device 3.

  If so, an alarm event corresponding to the malfunction thus detected is generated in step 505. If not, an alarm event corresponding to the risk of malfunction detected in step 503 is generated in step 506.

  Test step 507 is executed after execution of step 505 or 20 after execution of step 506.

  If it is determined in step 503 that none of the quality indices Q.e or Qe exceeds the warning threshold value, the test step 507 is executed.

  The test step 507 consists of determining whether one of the error rates tie or ie of the activity history of the removable memory medium 4 exceeds a warning threshold value set for these error rates. reading, this threshold value being representative of a risk of malfunction of this support 4 of removable memory.

  If the warning threshold value of step 507 is exceeded, test step 508 is executed. This step 508 consists of determining whether one of the error rates' U or exceeds an alarm threshold value 31 set for these error rates, this threshold value being representative of a malfunction of the memory medium 4. removable. If so, an alarm event corresponding to the malfunction thus detected is generated in step 509. If not, an alarm event corresponding to the risk of malfunction detected in step 507 is generated in step 510 .

  The test step 511 is executed after the execution of the step 509 or after the execution of the step 510.

  If it is determined in step 507 that none of the error rates ze or te exceeds the warning threshold of step 507, step 511 is executed.

  The test step 511 consists in determining whether one of the variations Ve or Ve of the activity history of the removable memory medium 4 exceeds a warning threshold value 32 fixed for these variations, this threshold value being representative of a risk of malfunction of this support 4 removable memory.

  If the warning threshold value 32 of step 511 is exceeded, the test step 512 is executed.

  This step 512 consists in determining whether one of the variations Ve or W exceeds an alarm threshold value 33 set for these variations, this threshold value being representative of a malfunction of the removable memory support 4. If so, an alarm event corresponding to the malfunction thus detected is generated in step 513. If not, an alarm event corresponding to the risk of malfunction detected in step 511 is generated at step 514 .

  The diagnostic module 9 makes it possible to prevent the malfunction 3 reader and / or recorder or a support 4 of removable memory by generating an alarm event in the steps 502, 506, 510 and 514. This alarm event can take the form of a message to the user, message that can be communicated via the man / machine interface 25. This message can inform the user of the risk of malfunction detected. The message may further suggest to the user the maintenance operations to be performed on the use of this removable support 4 or this device 3 reader and / or recorder in order to prevent a malfunction of this support 4 of removable memory or this device 3 player and / or recorder.

  Similarly, the diagnostic module 9 can detect a malfunction of a reader device and / or recorder and generate alarm events in the steps 501, 505, 509 and 513. These alarm events can take the form a message to the user via the home / machine interface 25, informing the user of the malfunction and proposing a maintenance operation to be performed.

  Alternately or in combination, diagnostic reports may be sent to a software application (not shown) for managing the archiving and backups on the devices 2 of the computing device 1. These diagnostic reports may take the form of a file communicated via means of transmission to a computer (not shown) comprising this software application.

  These transmission means may be, for example, the communication network 16. The software application would then be kept informed of a possible risk of malfunction of each support 4 of removable memory and each device 3 reader and / or recorder. The software application could then perform data migration operations recorded on a removable memory media 4 failing or at risk of failure to another support 4 of removable memory, the software application could further limit the use of certain devices 3 reader and / or recorder and some removable media 4 or perform maintenance operations on certain devices 3 reader and / or recorder, depending on the received report files, in order to limit the loss or risk of loss of data from backup and archiving.

  It should be noted that other means for communicating alarm events to said software application such as sending commands conforming to an application programming interface (API) of this application. software application.

  In a second embodiment (represented in FIG. 2), the computer device 1 according to the invention comprises several machines (computers).

  It comprises a machine, said monitoring machine 21, similar to the computer 24 of FIG. 1 comprising a processing module 7 and a diagnostic module 9. The computer device 1 comprises other machines, called storage servers 20, 20a, 20b, each comprising a central unit 5, similar to that of the computer 24, connected to peripherals 2 of mass memory. The central units of the storage servers 20, 20a, 20b each comprise a monitoring module 6 loaded in RAM.

  In the example shown, the peripherals 2 associated with a storage server 20a may be as described for the single computer 24 directly connected to the central unit 5 of this storage server 20a via a bus 14. of SCSI-type peripherals and a controller 12 of this peripheral bus 14.

  A fiber channel switch 19 can conveniently link the peripherals 2 to the storage server 20b. The central unit 5 is connected to the switch 19 via a Fiber Channel peripheral card. The peripherals 2 are connected to this Fiber Channel switch 19 and can furthermore be of SCSI type and be adapted to each comprise a Fiber Channel / SCSI bus converter (not shown) in order to be able to communicate with the switch 19 and the server 20b. storage.

  The devices 2 associated with another server 20b storage can further be distributed on a network type SAN (English storage area network). Such a SAN may be of the Fiber Channel type. Each peripheral 2 of this SAN can thus receive commands initiated by the monitoring module 6.

  The storage servers 20a, 20b can communicate with the monitoring machine 21 via a communication network 16 such as the Internet or a local area network (Ethernet for example). To do this, each storage server 20a, 20b and the monitoring machine 21 is provided with a network card 17 adapted for communication on the network 16. Thus, the LOGFILE files generated by the monitoring modules 6 of the servers 20 , 20a, 20b storage can be sent to the monitoring machine 21 and be processed by its processing module 7.

  The diagnostic module 9 of the monitoring machine 21 also carries out a diagnosis of the reader and / or recorder devices and the removable memory media 4 after each processing of a LOGFILE file by the processing module as described above.

  Other architectures of the computing device 1 than those given by way of non-limiting examples (FIGS. 1 and 2) can of course be envisaged.

  The mass storage peripheral 2 network monitoring architecture according to the invention is particularly useful for monitoring a large quantity of removable memory media 4, for example at least one hundred removable memory media 4 which can be loaded on at least ten devices 3 reader and / or recorder.

  Each monitoring module 6 may alternatively be a hardware module, such as a peripheral card, connected to a peripheral bus 14 to which is connected at least one mass storage device 2. This hardware monitoring module 6 would include means for sending, via a communication network, to a processing module 7 LOGFILE files collected for each monitored device 2.

  It should also be noted that several variants are possible for the realization of the monitoring module 6. In addition, the processing module 7 and / or the diagnostic module 9 can be made in the form of a hardware module, such as a card peripheral. In addition, a hybrid software / hardware solution can be selected for each of the modules according to the invention. In addition, the processing module 7 and / or the monitoring module 6 and / or the diagnostic module 9 can be combined in a single module performing the functions of these modules taken separately.

Claims (15)

  1 / device (1) comprising: - at least one mass storage device (2) comprising at least one device (3) reader and / or recorder, adapted to receive at least one support (4) of removable memory, this peripheral (2) of mass memory being adapted to generate data, called activity data, representative of its use and / or operation, - at least one software application capable of performing reading and / or recording operations with the device (s) (2) of mass memory 10 in use, characterized in that it comprises: - at least one monitoring module (6) of the performance of at least one peripheral device (2) ) of mass memory in service, said device (2) monitored, this monitoring module (6) being adapted to: 15. detect for each device (2) monitored if at least one device (3) reader and / or recorder receives a support (4) of removable memory This monitored device (2) being called monitored peripheral (2) active, or conversely if it receives no support (4) of removable memory, the monitored device is said monitored device inactive, 20. collect and record during time, for each monitored active device (2), activity data, called tracking data, comprising data, called read quality data and / or recording, adapted to allow the calculation of quality parameters, of which evolution over time is representative of a drift in the quality of reading and / or recording operations by each device (2) monitored, - at least one processing module (7) adapted to calculate and record, from the tracking data, a history of the quality parameters, said activity history, of each monitored device (3) and / or device recorder (2), this activity history being adapted to allow to prevent and / or detect a malfunction of each device (2) monitored.   2 / Apparatus according to claim 1, characterized in that it comprises a plurality of peripherals (2) of mass memory, and in that the monitoring module (6) is adapted to collect and record data monitoring of several devices (2) monitored.   3 / Apparatus according to one of claims 1 or 2, characterized in that the activity history of each device (3) reader and / or recorder comprises at least one quality parameter selected from the error rate in reading and / or the error rate in recording.   4 / Device according to one of claims 1 to 3, characterized in that the monitoring module (6) is adapted to be able to read in at least one memory of the computing device at least one identification data for each support (4) removable memory, said removable support (4) identified, received in each device (2) monitored, and in that the processing module (7) is adapted to calculate and record further a history of quality parameters, said history of activity, each identified removable memory medium (4) allowing, with the activity history of each device (3) reader and / or recorder, to determine if the origin of this malfunction comes from the support (4) of removable memory and / or device (3) reader and / or recorder.   5 / Apparatus according to claim 4, characterized in that the activity history of each support (4) of removable memory identified comprises at least one quality parameter selected from the read error rate and / or the rate of errors in recording and / or the number of loading and / or unloading operations and / or the duration of use in a device (3) reader and / or recorder.   6 / Apparatus according to one of claims 4 or 5, characterized in that the processing module (7) is adapted to update a single centralized database (15) comprising the quality parameters of each device (3) monitored reader and / or device recorder (2) and the quality parameters of each removable storage medium (4) identified, this database forming the activity logs of each device (3) reader and / or recorder and each support (4) of removable memory identified.   7 / Apparatus according to one of claims 4 to 6, characterized in that it comprises at least one diagnostic module (9) adapted for, from each history of activity it receives, trigger an event of alarm when at least one quality parameter takes a value corresponding to a risk of possible malfunction of the identified removable memory medium (4) and / or a reader and / or recorder device (3).   8 / Apparatus according to claim 7, characterized in that the processing module (7) is adapted to calculate a value of at least one evolution parameter representative of the variation in time of a quality parameter, and the diagnostic module is adapted to trigger an alarm event when at least one evolution parameter takes a value corresponding to a risk of possible malfunction of the identified removable memory medium (4) and / or a device (3) reader and / or recorder.   9 / Apparatus according to claim 8, characterized in that the diagnostic module (9) is adapted to compare each evolution parameter with a predetermined threshold value and trigger an alarm event when this threshold value is exceeded.   10 / Apparatus according to one of claims 7 to 9, characterized in that the diagnostic module (9) is adapted to compare each quality parameter with a predetermined threshold value and trigger an alarm event when the threshold value is exceeded .   11 / Apparatus according to one of claims 1 to 10, characterized in that the monitoring module (6) is adapted to collect the monitoring data periodically, according to a predetermined period.   12 / Apparatus according to claim 11, characterized in that the period is between ls and 10min -not of the order of 1 min-.   13 / Device according to one of claims 1 to 12, characterized in that the monitoring module (6) is adapted to be able to transmit the monitoring data to the processing module (7), the processing module (7) being adapted to calculate and record each activity history immediately after receiving this tracking data.   14 / Apparatus according to claim 7 and one of claims 1 to 13, characterized in that the diagnostic module (9) is adapted to run immediately after each record of an activity history by the processing module .   15 / Apparatus according to one of claims 1 to 14, characterized in that each device (2) of mass memory comprises a local memory, said register (8) of activity, and at least one controller (13) adapted for ability to record the activity data in areas of the predetermined activity register, depending on the nature of the activity data, and the monitoring module (6) is adapted to read areas of the activity register (8) corresponding to tracking data.