JP2000047950A - External storage device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the burden of switching to a host computer while keeping high reliability to the fault of plural paths with which storage devices are connected to each other. SOLUTION: An input/output request and transfer data from a host computer 100 are transmitted/received through an SCSI port 200 and SCSI interface board 300. Storage devices 400-410 are connected to each other in an array shape so as to be able to be shared from plural SCSI interface boards 300-320. An interface monitoring device 500 monitors conditions of interface boards 300-320 at all times, and when any abnormality is found out, the interface board of light statistic load is selected out of normally operating interface boards so that the ID switching of interface boards 300-320 is performed between the both. Thus, the user of an external storage device 150 can continuously perform processing even in the case of operation defect on the interface board and the reliability of the device is improved as a whole.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an external storage device connected to a computer device or a control device, and more particularly to an external storage device having high reliability.

[0002]

2. Description of the Related Art In the event of an interface failure due to a demand for reliability of non-stop operation and non-stop maintenance of an external storage device, processing is continued by switching the interface to be processed on the host computer side and accessing again. It is common to do. In this case, in the case of an open platform, path switching cannot be performed automatically, so that much time must be spent on failure detection, switching, and data recovery. JP-A-8-335
No. 155 describes non-stop operation.
In this method, it is possible to cope with a single identifier (ID) by using a redundant configuration. However, this method requires a redundant configuration for each ID in order to support a plurality of IDs, and cannot cope with a case where all the redundant configurations for a single ID stop operating.

[0003]

As a recent market trend, a demand for a multipurpose file server to which a highly reliable external storage device is connected has been increasing. Since a plurality of disk images can be shown to a client, it is required to deal with transaction processing for important data with high reliability. For that purpose, it is necessary to realize high reliability for the storage device inside the external storage device. However, in the conventional technology, no consideration has been given to the complexity of path switching for an open platform device failure. SUMMARY OF THE INVENTION It is an object of the present invention to provide an external storage device that requires high reliability, while maintaining high reliability against failures in a plurality of paths connecting the storage devices and not placing a burden on the host computer for switching.

[0004]

According to one aspect of the present invention, there is provided at least one storage device for storing data transmitted to and received from an upper-level device, and the upper-level device and the storage device interposed between the storage device and the upper-level device. And a plurality of storage controllers for controlling the transmission and reception of the data between the storage controller and an identifier used by the higher-level device to specify the storage controller is set in each of the storage controllers. Interface means for connecting the storage control device to the host device so that the plurality of storage devices look equivalent when viewed from the host device,
Connected to all of the storage control devices, and monitors the presence or absence of a failure in the storage control device. When a failure occurs, an identifier provided between the storage control device having the failure and the one storage control device having no failure. Monitoring means for mutually setting exchanges.

[0005] Further, the interface means is provided for each of the storage control devices, and includes means for setting a different identifier for each of the storage control devices. Further, the plurality of interface means to which the storage control devices are respectively connected are daisy-chain connected.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the external storage device according to the present invention will be described below. FIG. 1 is a diagram showing one configuration of an entire system including an external storage device according to the present embodiment. The external storage device 150 corresponds to a plurality of storage control devices S
SCSI port (200-22) corresponding to the interface means directly connected to the CSI interface board (300-320)
0), an interface monitoring unit 500 for monitoring the SCSI interface board, and a plurality of storage devices (400-410).

A plurality of storage devices (400-410), a plurality of SCSI interface boards (300-320), and a SCSI port (200-220) can be freely connected and set. Although not shown, this setting can be performed by controlling a connection switch arranged in the external storage device by a microprocessor or the like provided in the external storage device. It can also be performed by manually controlling the switch board provided in the above. This allows multiple interface boards for a single storage device (400-410)
(300-320) can be shared. SCSI ID (350
-370) has a different value for each interface board (300-320), and the SCSI ID (350-370) is set by a microprocessor or the like provided in an external storage device, not shown. SCSI with interface board (300-320) connected
The ports (200-220) are connected in a daisy chain, and when the individual interface boards (300-320) are operable, the host computers (100-110) execute different processes in parallel.

The interface monitoring unit 500 is connected to and communicates with all the interface boards (300-320).
The internal status of the interface board (300-320) is always received at regular intervals and analyzed to check the operation confirmation.If any malfunction of the interface board (300-320) is confirmed, contact the outside. , SCSI ID (350-370)
Can be exchanged dynamically. During the operation of the interface board, the specified interface board receives an I / O request for each SCSI ID, and performs processing corresponding to the request to the storage device connected by setting.

In the case of FIG. 1, a request for SCSI ID 1 by the host computer 100 is received by the interface board 300 and the storage device 400 processes the request. However, the interface board 30
Consider when 0 stops working. At this time, the interface monitoring unit 500 detects an operation failure of the interface board 300 because the transmission of the internal situation from the interface board 300 is not performed or an abnormality is detected in the received internal situation.

[0010] The interface monitoring section 500 is an interface board (here, the load means a ratio operated during a certain period of time) of the interface boards (310, 320) that are operating normally. ) And confirm that the interface board 320 is not malfunctioning.
And 370 are exchanged. At the same time, with this exchange, the interface boards 300 and 320 and the storage device (400-41
The connection with 0) is also exchanged. This exchange allows the host computer 100 to continue processing without being aware of the failure of the interface board 300.

The switching of the SCSI ID is performed by the S of the host computer.
Execute during retransmission of SCSI command due to CSI command execution failure. Since the interval between retransmission of SCSI commands to the external storage device is on the order of tens of seconds, if the period is sufficiently shorter, the interface board is disconnected and the SCSI ID of the command retransmission destination exists on the SCSI bus. If it doesn't, there is no effect. Therefore, in the event of a failure of the interface board 300, once the interface boards 300 and 320 are disconnected from the SCSI bus, exchange of ID settings is completed, and
By performing the work of connecting to the bus, the ID of the interface board is exchanged.

FIG. 2 shows the configuration of the entire system including the external storage device after the exchange of the IDs and the exchange of the connection between the interface board and the storage device accompanying the exchange.

When a request for SCSI ID 3 (the interface board with SCSI ID 3 is malfunctioning as described above) occurs in a state where the SCSI IDs 1 and 3 have already been exchanged. Similarly interface board 31
A board with a light load among 0 and 320 switches IDs and continues processing. Therefore, considering the same processing when multiple interface board operation failures occur,
If one interface board can operate normally in the external storage device, it can operate as an external storage device.

FIG. 3 is a block diagram showing the internal configuration of the interface monitoring unit 500. A microprocessor unit 510 (hereinafter abbreviated as MPU) sequentially reads out and decodes programs (including information) stored in a random access memory 520 (hereinafter abbreviated as RAM), executes the programs while decoding them, and executes an interface through a connection mechanism 530. Board 300-3
20 status monitoring and ID replacement controls are being performed.
The connection mechanism 530 connects to the interface boards 300-320 in response to a request from the MPU, and
Can receive the operation status.

An example of the above-described monitoring procedure will be described below with reference to the flowchart of FIG. Interface monitoring unit
500 is all SCSI ports 200-2 daisy chained
When an input from a higher-level device is made to 20, all interface boards that have detected the input are connected to the MP via the connection mechanism 530.
An interrupt is sent to the U510 (however, multiplexing of interrupts does not occur due to the interrupt mask), the MPU 510 is started (step 600), and the ID specified by the input from the higher-level device in step 601, ie, the ID specified by the external request, Is received from the connection mechanism 530. At step 602, it is determined whether or not there is any inconsistency in the contents of the state. If the interface board is operating, the load information received at step 607 is stored in the RAM.

If the interface board is not operating, at step 603, a lightest load interface board is selected by referring to the load information in the RAM, out of the interface boards that are currently operable and are not in use or in a locked state waiting for a request. . The ID of the board selected in Step 605 is exchanged with the ID of the board that is not operating, and Step 606 is performed.
Then, the fact that the board selected in step 605 is in the lock state waiting for a request is stored in the RAM. If there is no boat satisfying the above condition, the process is terminated, the re-transmission of a request from the outside is waited, and at this time, the switching is performed again. Here, when only one interface board is operable, a request specifying an inactive ID is alternately output to the external storage device 10.
If sent to 0, it is possible that ID switching between the aforementioned alternate IDs is performed endlessly at 605, and no request is executed at all. The wait-for-request lock state used is necessary to avoid this. This lock state is not changed until one external request is executed.

[0017]

According to the present invention, in the case of an external storage device having a configuration in which a plurality of storage devices are connected in an array, the user of the device can perform an interface abnormality without being aware of the mounting state of the storage devices constituting the device. The processing can be continued continuously, and the reliability of the entire apparatus is improved.

[Brief description of the drawings]

FIG. 1 is a diagram showing one configuration of an entire system including an external storage device.

FIG. 2 is a diagram showing a configuration of the entire system after ID switching.

FIG. 3 is a block diagram illustrating an internal configuration of an interface monitoring unit.

FIG. 4 is a diagram showing a flowchart of a monitoring procedure in an interface monitoring unit.

[Explanation of symbols]

100-110 Host computer 150 External storage device 200-220 SCSI port 300-320 Interface board 350-370 SCSI I of interface board
D 400 to 410 Storage device 500 Interface monitoring unit 510 Microprocessor unit (MPU) 520 Random access memory (RAM) 530 Connection mechanism

Continued on the front page F term (reference) 5B014 EA04 EB04 GD05 GD06 GD21 GD32 HA09 HB01 HB14 HB15 HC12 HC13 5B083 AA00 BB01 BB03 BB11 CD09 CD10 CD11 EE06 EE11 GG04

Claims (3)

[Claims] An information processing apparatus comprising: at least one storage device for storing data transmitted to and received from a higher-level device; and one or more storage devices interposed between the storage device and the higher-level device and located between the higher-level device and the storage device. An external storage device including a plurality of storage control devices that control the transmission and reception of the data, wherein each of the storage control devices is set with an identifier used by the higher-level device to specify the storage control device, Interface means for connecting the storage control device to the high-level device so that the plurality of storage devices look equivalent when viewed from the high-level device; presence / absence of a fault in the storage control device connected to all the storage control devices Monitoring means for monitoring and setting the identifiers of the storage controller having a fault and one storage controller having no fault when the fault is present. Storage device. 2. The external storage device according to claim 1, wherein said interface means is provided for each of said storage control devices, and comprises means for setting a different identifier for each of said storage control devices. External storage device. 3. The external storage device according to claim 2, wherein the plurality of interface units to which the storage control devices are respectively connected are daisy-chain connected.