DE29701140U1 - Data archive system - Google Patents

Description

EUROPEAN PATENT ATTORNEYS

MANITZ, .RATEMT- UIfB ₀ LAWYERS

"^iRARTNER

MANiTZ, FINSTERWALD & PARTNER ■ P.O. BOX 22 16 11 · 80B06 MUNICH

SAXON

Archiv-Systeme GmbH Hutbergstrasse 25

90475 Nuremberg GERMAN PATENT ATTORNEYS

DR. GERHART MANlTZ · dipl.-phys.

MANFRED FINSTERWALD · dipl.-ing., dipl.-wirtschafts.-ing.,

DR. HELIANE HEYN · dipl-chem.

DR. MARTIN FINSTERWALD ■ dipl.-ing.

STEPHAN THUL · dipl.-phys.

DR. DIETER PELLKOFER · DIPL.-ING.

CHRISTIAN SCHMIDT · dipl.-phys.

GÜNTHER KURZ · dipl.-ing.

WERNER GRÄMKOW ■ dipl.-ing. (-19831

BRITISH CHARTERED PATENT ATTORNEY

JAMES G. MORGAN ■ B. se. iphys.), d.m.s.

LAWYER

MARION CHRISTIANE SCHMIDT

REPRESENTATIVES BEFORE THE EUROPEAN PATENT OFFICE

80506 MUNICH

PO BOX 22 16 11 · ROBERT-KOCH-STRASSE

TELEPHONE (089) 21 99 430 ■ FAX (089) 29 75

DATE

Munich, 23.01.1997 S 6380-Dt/Ru

Data archiving system

MANlTZ ■ FINSTERWALD · HEYN · MORGAN · FINSTERWALD · SCHMIDT · PELLKOFER · SCHMIDT · KURZ ■ 80538 MUNICH ■ ROBERT-KOCH-STRASSE 1 ■ TAX No. DE 1 30 001 124

STEPHAN THUL ■ 70372 STUTTGART (BAD CANNSTATT) · SEELBERGSTRASSE 23/25 · TELEPHONE (0711)56 72 61 HYPOBANK MUNICH · 6 880 119 980 · Bank code 700 200 01 · POSTBANK MUNICH · 770 62-805 ■ Bank code 700 100 80 ■ VEREINSBANK MUNICH · 578 351 · Bank code 700 202 70

S 6380-Dt/Ru Description

The present invention relates to a data archiving system for reading and researching archived data.

In both business transactions and private use, information of all kinds, such as letters, invoices, delivery notes, etc., is increasingly no longer stored on paper but on magnetic or optical data storage devices. In this way, the space required to store the stored information can be significantly reduced compared to storing it on paper.

In order to further increase the acceptance of paperless information and to be able to make sensible use of the inherent advantages of such storage, such as searching for specific keywords, the maximum access time for a search for a desired document must be kept as short as possible. For such online research, all available data must be stored on a storage medium with direct, fast access. This means that the existing storage capacity, such as the storage capacity of a hard disk, is used up within a short time or that, when using CD-ROMs, for example, a large number of CD-ROM drives must be available, whereby the number of CD-ROM drives is limited for both technical and administrative reasons.

The object of the present invention is to provide a data archiving system with which any number of archived data can be managed, whereby for all data

sufficiently fast access is guaranteed.

This object is achieved according to the invention by a data archive system with at least two archive data management units connected to one another by a network, with at least two reading units for external data carriers containing archive data, and with at least one interface for transmitting external research data via the network to the archive data management units, wherein the archive data management units are each designed to control at least one of the reading units, the data carriers that contain archive data of the same priority are each divided into the same priority classes, the reading units divided into different priority classes have different maximum access times, in particular reading units divided into higher priority classes have shorter maximum access times than reading units divided into lower priority classes, and the priority classes of the reading units each correspond to the priority classes of the data carriers assigned to them.

In a method for using this data archive system, the data carriers are divided into different priority classes before starting and/or during the search, according to different priorities of the archive data stored on the respective data carriers, whereby the data carriers are assigned to different reading units according to their respective priority class, which have different maximum access times corresponding to the respective priority classes.

The concept of classifying data storage devices into corresponding priority classes depending on the priority of the data stored on them and reading units with different maximum access times means that

achieved that a data archive system designed according to the invention can be individually adapted to the needs of the customer both in terms of its scope and the desired access times. The archive data can have different priorities from the outset, according to which the shortest possible access time must always be guaranteed for certain stored data, while for data that is rarely used, for example, longer access times are acceptable.

However, the priorities of certain data storage devices can also shift over time, so that an automatic or manual reclassification of certain data storage devices into the specified priority classes takes place. For example, invoices that have already been paid are usually no longer needed, so that the priority of data storage devices that only contain fully paid invoices can be moved from a high to a low priority class. This reclassification can also be done based on the date, for example, since it can usually be assumed that the majority of invoices are paid 6 months after the invoice date, and thus data storage devices that only contain invoices that are older than 6 months can automatically be assigned to a lower priority class.

Depending on the amount of data to be managed, more than two archive data management units and reading units can be provided, whereby the reading units and the data carriers can also be divided into more than two different priority classes.

In this way, a modular data archiving system can be developed that can be optimally configured to the existing data volumes and the corresponding priorities.

Thus, for example, when introducing a data archive system designed according to the invention, the costs can be kept relatively low, while if the data stock grows to a correspondingly large number, the system can be expanded without any problems. In the course of this expansion, the existing data storage devices can then also be reclassified into different priority classes.

According to an advantageous embodiment of the invention, the reading units of all archive data management units are suitable for reading all data carriers used. This ensures that a reclassification of the data carriers used into each of the existing priority classes is possible, so that no hardware-related restrictions occur when configuring a data archive system designed according to the invention.

According to a further preferred embodiment of the invention, the reading units are designed as single drives or as multiple drives or as jukeboxes, each in particular for optical data carriers, in particular for CD-ROMs. The single drives are preferably assigned to a higher priority class than the multiple drives and the multiple drives are assigned to a higher priority class than the jukeboxes.

By using CD-ROMs, a high writing density is achieved at low production costs, and data exchange is easy thanks to the existing standards for CD-ROMs (ISO 9660). However, any other data storage media, such as magnetic or magneto-optical data storage media, can also be used.

While single drives have the fastest access times,

» 4

, multiple drives or jukeboxes have the advantage that a larger number of data storage devices can be managed. While multiple drives can usually accommodate between 2 and 10 data storage devices, jukeboxes can manage up to 1000 data storage devices or more. Due to the mechanical insertion and removal of data storage devices in a jukebox, the maximum access times are significantly higher than with a multiple drive.

According to a further preferred embodiment of the invention, the archive data stored on the data carriers are managed via centrally stored index data. In particular, a central data management unit is connected to the network, in which the index data is centrally stored and with which the index data is centrally managed, in particular updated.

By centralizing the index data for all existing data carriers, particularly fast access to the desired archive data is possible, since when conducting a corresponding search, the archive data management unit with the corresponding reading unit that contains the desired data carrier can be determined directly from the index data.

Further preferred embodiments of the invention are specified in the subclaims.

The invention is explained in more detail below using an embodiment with reference to the single figure.

The figure shows a data archive system that includes a network 1, via which two archive data management units 2, 3 are connected to one another. The archive data management units

» »9

units are designed as personal computers and are referred to below as image servers. Instead of personal computers, any other computers can be used as long as the functionality described below can be achieved.

The image server 2 comprises three SCSI controllers, each of which is connected to and controls seven CD-ROM drives 5.

The image server 3 comprises three SCSI controllers 6, each of which is connected to and controls seven CD-ROM changers 7.

A CD-ROM jukebox 8 is also connected to the network 1, which includes four drives 9 and a large number of CD-ROMs 10 on which archive data is stored.

Furthermore, an offline station 11 is connected to the network 1, which contains a large number of additional CD-ROMs 10.

The CD-ROM drives 5, the CD-ROM changers 7 and the drives 9 of the CD-ROM jukebox 8 are designed in such a way that the CD-ROMs 10 can be exchanged between these reading devices at will.

A central data management unit 12 as well as a first research unit 13 and a second research unit 15 connected via an interface 14, for example in the form of a modem, are also connected to the network 1. Both the first research unit 13 and the second research unit 15 are designed as personal computer systems and include input units 16, 17 as well as screens 18, 19 and printers 20, 21 for displaying and printing data.

The functionality of a data archiving system designed according to the invention is described in more detail below:

The embodiment of a data archive system designed according to the invention shown in the figure supports four different priority classes. The highest priority class is formed by the image server 2 with the CD-ROM drives 5, through which access to CD-ROMs 10 arranged in the CD-ROM drives 5 is possible within a maximum of 3 seconds. Access to archive data located on a CD-ROM 10 within a CD-ROM drive 5 is therefore referred to as immediate online operation.

Since the image server 2 is equipped with three SCSI controllers 4, each of which can control seven CD-ROM drives, access to 21 CD-ROMs is possible in immediate online operation. Since approximately 15,000 scanned DIN A4 pages can be stored per CD-ROM, access to approximately 315,000 scanned DIN A4 pages is possible via the image server 2 within a maximum of 3 seconds.

The second priority class is formed by the image server 3. The image server 3 comprises three SCSI controllers 4, each of which controls 7 quadruple CD-ROM drives 7. The maximum access time for such quadruple CD-ROM changers is around 6 seconds, so that the image server 3 can access 84 CD-ROMs and thus 1,260,000 scanned DIN A4 pages within a maximum of 6 seconds. This operation is referred to as near-online operation.

The third priority class is formed by the CD-ROM jukebox 8, which comprises four drives 9. The maximum access time for a CD-ROM jukebox is approximately 10 to 15 seconds.

so that research on a CD-ROM 10 located inside the jukebox 8 is referred to as far-online operation. As a CD-ROM jukebox 8 can contain up to 1,000 or more CD-ROMs 10, access to around 15 million scanned DIN A4 pages is possible within a maximum of 15 seconds.

The fourth and lowest priority class is formed by the offline station 11, in which CD-ROMs 10 are stored which, for example, only have to be kept due to legal regulations and are not usually required for research. These CD-ROMs 10 can also be inserted, for example, into a CD-ROM drive provided in the offline station 11 or into the CD-ROM jukebox 8, so that the archive data stored on the CD-ROMs 10 stored offline can still be accessed.

Before starting the first search with the data archive system designed according to the invention or optionally during a search, the existing CD-ROMs 10 are divided into the four predefined priority classes according to the importance of the archive data stored on them. This division can be done manually or automatically, whereby, for example, CD-ROMs 10 which contain invoices that have already been paid can be automatically divided into the priority class of the offline station 11.

To search for a specific document, the user enters a corresponding search request, for example an SAP number, via the input units 16, 17 of the search units 13, 15 designed as personal computers. In the present example, the search unit 13 is directly connected to the network 1, while the search unit 15 is connected via a modem 14, for example via a

^m ·

normal telephone connection or an ISDN connection to the network, so that research on data media archived outside the company is possible.

The entered search request is converted into a predefined search format, transferred to the central server 12 and stored there in a search data directory.

After inserting a CD-ROM 10 into a CD-ROM drive 5 of the image server 2, into a CD-ROM changer 7 of the image server 3 or into the storage container of the CD-ROM jukebox 8, the inserted CD-ROMs 10 are scanned by the image server 2, the image server 3 or the jukebox 8, whereupon the information about the content of the respective inserted CD-ROM 10 is transferred to the central server 12 via the network 1. From this information, an index database is built up in the central server 12, which contains index data for all available CD-ROMs 10. Optionally, the image servers 2, 3 and the jukebox 8 can also check the inserted CD-ROMs 10 at fixed intervals and report changes in the content of the CD-ROMs 10 to the central server 12, whereupon the index data is updated accordingly.

When a CD-ROM 10 is removed from a CD-ROM drive 5, a CD-ROM changer 7 or the CD-ROM jukebox 8, corresponding offline information is transferred to the central server 12, so that the index data can be used to identify which of the CD-ROMs 10 in the database are available online and which, in contrast, are only available offline.

The index data stored on the central server 12 thus directly reflects which CD-ROM 10 is in which drive 5, 7, 9 of the image server 2, 3, or the CD-ROM jukebox.

box 8 is inserted or which CD-ROMs 10 are offline in the offline station 11.

The central server 12 checks the search data directory at fixed time intervals, for example at intervals of 0.25 seconds, for corresponding search data and determines on the basis of the centrally stored index data whether the desired document is available online and in which reading unit 5, 7, 9 it is located. If the desired document has been recognized as available online on the basis of the index data, a corresponding document address derived from the index data is stored in a document address directory on the central server 12. This directory is searched by the image servers 2, 3 and the CD-ROM jukebox 8 at fixed time intervals, for example every 0.25 seconds, for corresponding document addresses, whereby each image server 2, 3 or the jukebox 8 recognizes on the basis of a found document address whether it is a document that is available on a CD-ROM 10 inserted in its own drive 5, 7, or 9.

The desired document is transferred from the corresponding CD-ROM 10 via the network 1 to the research unit 13 or via the modem 14 to the research unit 15, where it can be displayed on a screen 18 or 19 and printed out on a printer 20 or 21.

If the desired document is recognized by the central server 12 as not being available online, a corresponding offline request is sent via the network 1 to the offline station 11, whereupon an operator working in the offline station 11 manually inserts the desired CD-ROM 10 into a separate CD-ROM drive or, for example, into the jukebox 8.

If a document is recognized by the central server 12 as being available online or not available online, a corresponding message is sent to the research unit 13 or 15, which is displayed on the screen 18, 19 of the research unit 13, 15. In this way, the person conducting the research is informed about the current status and any delays that may occur if the desired document is only accessible offline.

The mailbox procedure described as an example, in which communication between the individual devices connected to network 1 takes place by storing data in directories, has the advantage that all components except the central server 12 can fail and the other components are still fully functional. In principle, however, communication between the individual devices connected to network 1 can also function according to other principles.

The advantage of the modular system is that, on the one hand, the number of online drives can be adapted to the customer's needs and, on the other hand, depending on the importance of the data, different access times can be achieved, which are associated with correspondingly different costs. In particular, when a data archive system designed according to the invention is newly installed, only an immediate online image server can be set up, which can be expanded to up to 21 CD-ROM drives if required. Only when the maximum capacity of the image server in immediate online mode has been reached must a decision be made as to which documents can be converted to near-online mode, so that the corresponding hardware (near-online image server with associated CD changers or far-online image server) must also be purchased at this point.

Claims (19)

S 6380-Dt/Ru Protection claims 1. Data archiving system with at least two archive data management units (2, 3, 8) connected to one another by a network (1), with at least two reading units (5, 7, 9) for external data carriers (10) containing archive data, and with at least one interface (14) for transmitting external research data via the network (1) to the archive data management units (2, 3, 8), wherein the archive data management units (2, 3, 8) are each designed to control at least one of the reading units (5, 7, 9), those data carriers (10) that contain archive data of the same priority are divided into the same priority classes, the reading units (5, 7, 9) divided into different priority classes have different maximum access times, in particular reading units (5, 7, 9) divided into higher priority classes have shorter maximum access times than reading units (5, 7, 9) divided into lower priority classes, and the priority classes of the reading units (5, 7, 9) each correspond to the priority classes of the data carriers (10) assigned to them. 2. Data archive system according to claim 1, characterized in that the reading units (5, 7, 9) of all archive data management units (2, 3, 8) are suitable for reading all data carriers (10) used. 3. Data archive system according to claim 1 or 2, characterized in that the reading units (5, 7, 9) are designed as single drives (5) or as multiple drives (7) or as jukeboxes (8, 9), each in particular for optical data carriers (10). 4. Data archive system according to claim 3, characterized in that the individual drives (5) are classified into a higher priority class than the multiple drives (7) and the multiple drives (7) are classified into a higher priority class than the jukeboxes (8, 9). 5. Data archiving system according to one of the preceding claims, characterized in that the archive data stored on the data carriers (10) are managed via centrally stored index data. 6. Data archive system according to claim 5, characterized in that a central data management unit (12) is connected to the network (1), in which the index data are stored centrally for the purpose of management and in particular updating. 7. Data archive system according to claim 6 , characterized in that the central data management unit (12) is designed to receive and evaluate incoming search data via the interface (14) and, via the stored index data, enables an assignment between the incoming search data and the archive data management unit (2, 3, 8) which has access to the data carrier (10) which contains the archive data corresponding to the search data. 8. Data archive system according to claim 6 or 7, characterized in that one of the archive data management units (2, 3, 8) is used as the central data management unit (12). 9. Data archiving system according to one of the preceding claims characterized in that the interface (14) is connected to a search unit (13, 15) for generating and transmitting the search data to the interface (14), wherein the search data can be generated in particular from search commands entered via an input unit (16, 17). 10. Data archive system according to claim 9, characterized in that the archive data determined can be transmitted to the research unit (13, 15) via the interface (14). ¹ 11. Data archive system according to claim 9 or 10, characterized in that the search unit (13, 15) comprises at least one output unit (18, 19, 20, 21) for displaying and/or printing the determined archive data. 12. Data archiving system according to one of the preceding claims, characterized in that the archive data management units (2, 3, 8) and/or the central data management unit (12) and/or the research unit (13, 15) are designed as computers, in particular as personal computers. 13. Data archive system according to one of claims 3 to 12, characterized in that the maximum access time to an archive data record contained on a data carrier (10) in a single drive (5) is less than approximately five seconds, in particular less than approximately four seconds, preferably less than approximately three seconds. 14. Data archive system according to claims 3 to 13, characterized in that the maximum access time to an archive data record contained on a data carrier (10) in a multiple drive (7) is between approximately three and approximately nine seconds, in particular between approximately four and approximately eight seconds, preferably approximately six seconds. 15. Data archive system according to claims 3 to 14, characterized in that the maximum access time to an archive data record contained on a data carrier (10) in a jukebox (8) is between approximately eight and approximately 25 seconds, in particular between approximately eleven and approximately 20 seconds, preferably approximately 15 seconds. 16. Data archiving system according to one of the preceding claims, characterized in that several reading units (5, 7) are connected to an archive data management unit (2, 3) in particular via a SCSI interface (4, 6) and can be controlled by the latter. 17. Data archiving system according to one of the preceding claims, characterized in that a maximum of seven reading units (5, 7) are connected to an archive data management unit (2, 3) via a SCSI interface (4, 6). 18. Data archiving system according to one of the preceding claims, characterized in that a maximum of three SCSI interfaces (4, 6) are provided on each archive data management unit (2, 3). 19. Data archiving system according to one of the preceding claims, characterized in that the data carriers (10) are designed as CD-RCMs.