System and method to maintain performance among N single raid systems during non-fault conditions while sharing multiple storage devices during conditions of a faulty host computer or faulty storage array controller

• US 6,330,687 B1
• Filed: 11/13/1998
• Issued: 12/11/2001
• Est. Priority Date: 11/13/1998
• Status: Expired due to Fees

• Current Assignee: Digi-Data Corporation (Synchronoss Technologies Incorporated)
• Original Assignee: Digi-Data Corporation (Synchronoss Technologies Incorporated)
• Inventors: Griffith, Geoffrey J
• Primary Examiner(s): Le, Dieu-Minh
• Assistant Examiner(s): VO, TIM T
• Application Number:
  US09/192,016
• Time in Patent Office: 1,124 Days
• Field of Search: 714/6, 714/2, 714/4, 714/5, 714/8, 714/11, 714/12
• US Class Current: 714/6.21
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 11/2089 Redundant storage control functionality

Abstract

In order to provide redundancy of host computers and storage array controllers, and thereby protect operation of a RAID system against host computer or storage array controller failure, two single RAID subsystems are conventionally provided in a conventional active-active configuration. This means each storage array controller has electrical access to each disk in its array as well as each disk in the other controller'"'"'s array. The resultant interference between the controllers in accessing the disk channels causes the active-active system to normally function at approximately 130% of the speed of a single RAID system, rather than the optimum 200% of the speed of a single RAID system. The system of this invention, FULL-SPEED ACTIVE-ACTIVE redundant RAID system, contains a normally open switch or repeater which allows access by each storage array controller only to that controller'"'"'s array of disks when both host computers and storage array controllers are operating normally. This provides a speed of 200% of the speed of a single RAID system. When one host computer or storage array controller fails, the switch or repeater is automatically closed, allowing the storage array controller of the functioning subsystem to control all of the disks of both single RAID subsystems.

Claims

1. A fault-tolerant RAID system having N host computers and N storage array controllers with redundancy in the host computers and storage array controllers comprising:

N single RAID subsystems, each single RAID subsystem comprising;

one host computer,

one storage array controller,

each host computer electrically connected to a storage array controller by connector means, and

one array of direct access storage units comprising a multiplicity of channels, each channel comprising a multiplicity of direct access storage units electrically connected with each other in series by connector means,

connector means connecting each storage array controller with one array, each storage array controller with the capacity to control two arrays, and

N cores, each core having switching means with an open position and a closed position, and connection means whereby each switching means may be connected to two channel connection means, and whereby each switching means controls the electrical connection between the two channel connection means,

each core connected by channel connection means to each channel of the array of a first single RAID subsystem and to each channel of the array of an adjacent second simple RAID subsystem, thereby dividing the N single RAID subsystems into N pairs, each pair comprising a first and a second adjacent single RAID subsystem and one core, and connecting means, the first adjacent single RAID subsystem of one pair being the second adjacent single RAID subsystem of an adjacent pair,

the switching means within each core connected to the channels so when the switching means is in the closed position each channel of the array of the first single RAID subsystem of a pair is electrically connected bidirectionally to a single channel of the second single RAID subsystem of the same pair, thereby allowing the storage array controller of the first single RAID subsystem to control both arrays of the pair,

indicator connecting means connecting the storage array controller of the first single RAID subsystem of a pair with the core of the pair,

the storage array controller of the second single RAID subsystem of the pair providing a normal operating signal to the host computer of the first single RAID subsystem of the pair via connector means when the storage array controller and the host computer of the second single RAID subsystem are operating normally,

the storage array controller of the second single RAID subsystem of the pair not providing a normal operating signal to the host computer of the first single RAID sub system of the pair via connector means when the storage array controller or the host computer of the second single RAID subsystem has failed,

the host computer of the first single RAID subsystem of the pair providing a closure signal via indicator connector means to the core when no normal operating signal is received from the storage array controller of the second single RAID subsystem of the pair,

the switching means being normally open so there is no electrical connection between the channels of one single RAID subsystem and the channels of the other single RAID subsystem of a pair when no closure signal is received by the core,

the switching means being changed to the closed position when the core receives a closure signal from the host computer of the first single RAID subsystem of the pair, thereby establishing an electrical connection between the channels of both of the single RAID subsystems of a pair and allowing the functioning controller and host computer of the first single RAID subsystem to control the channels of both the first and second single RAID subsystems of the pair.

2. The system of claim 1 wherein the switching means is a repeater.

3. The system of claim 1 wherein the direct access storage units are disks, tapes, or CDS.

4. The system of claim 1 wherein the direct access storage units are disks.

5. The system of claim 1 wherein the connecting means connecting each storage array controller with one array is an IDE, SCSI or Fibre Channel interface.

6. The system of claim 1 wherein the number of single RAID subsystems is 2, there is only one pair of single RAID subsystems, each single RAID subsystem functions as both the first and second RAID subsystem of the pair, and there is only one core which is connected by indicator means to each member of the pair of single RAID subsystems.

7. The system of claim 2 wherein an independent repeater circuit serves each channel.

8. The system of claim 5 wherein the connecting means connecting each storage array controller with one array is a SCSI interface.

9. A fault-tolerant RAID system having 1 host computer and 2 storage array controllers with redundancy in the host computers and storage array controllers comprising:

two single RAID subsystems, each single RAID subsystem comprising;

one host computer,

one storage array controller,

each host computer electrically connected to a storage array controller by connector means, and

one array of direct access storage units comprising a multiplicity of channels, each channel comprising a multiplicity of direct access storage units electrically connected with each other in series by connector means, the connector means having channel connection means,

connector means connecting each storage array controller with one array,

each storage array controller with the capacity to control two arrays, and

one core having switching means with an open position and a closed position, and connection means whereby each switching means may be connected to two channel connection means, and whereby each switching means controls the electrical connection between the two channel connection means,

each core connected by channel connection means to each channel of the array of a first single RAID subsystem and to each channel of the array of an adjacent second single RAID subsystem, forming a pair of single RAID subsystems,

the switching means within the core connected to the channels so that when the switching means is in the closed position each channel of the array of one single RAID subsystem of a pair is electrically connected bidirectionally to a corresponding channel of the other single RAID subsystem of the pair, thereby allowing the storage array controller of one single RAID subsystem to control both arrays of the pair,

indicator connecting means connecting the storage array controllers of each single RAID subsystem of a pair with the core of the pair,

the storage array controller of each single RAID subsystem of the pair providing a normal operating signal to the host computer of the other single RAID subsystem of the pair via connector means when the storage array controller and the host computer of that single RAID subsystem are operating normally,

the storage array controller of either single RAID subsystem of the pair not providing a normal operating signal to the host computer of the other single RAID system of the pair via connector means when the storage array controller or the host computer of that single RAID subsystem has failed,

the host computer of either single RAID subsystem of the pair providing a closure signal via indicator connector means to the core when no normal operating signal is received from the storage array controller of the other single RAID subsystem of the pair,

the switching means being normally open so there is no electrical connection between the channels of one single RAID subsystem and the channels of the other single RAID subsystem of a pair when no closure signal is received by the core,

the switching means being changed to the closed position when the core receives a closure signal from the host computer of either single RAID subsystem of the pair, thereby establishing an electrical connection between the channels of both of the single RAID subsystems of a pair and allowing the functioning controller and host computer of one single RAID subsystem to control the channels of both single RAID subsystems of the pair.

10. The system of claim 9 wherein the switching means is a repeater.

11. The system of claim 9 wherein the direct access storage units are disks, tapes, or CDS.

12. The system of claim 9 wherein the direct access storage units are disks.

13. The system of claim 9 wherein the connecting means connecting each controller with one array of direct access storage units is an IDE, SCSI or Fibre Channel interface.

14. The system of claim 9 wherein the connecting means connecting each controller with one array of direct access storage units is a SCSI interface.

15. The system of claim 10 wherein an independent repeater circuit serves each channel.

16. In a redundant RAID system having N single RAID subsystems and a core with repeater switching means connecting the channels of each single RAID subsystem with the channels of the adjacent single RAID subsystem, the process of controlling the arrays of direct access storage units of both the first and the second adjacent single RAID subunits by the first single RAID subsystem when the second adjacent single RAID subunit has a faulty storage array controller comprising the steps:

a. ceasing to emit a normal operating signal or heartbeat by the storage array controller of the second adjacent single RAID subsystem having a faulty storage array controller,

b. noting the cessation of emission of a normal operating signal by the storage array controller of the second adjacent single RAID subsystem by the first single RAID subsystem, and emitting a close signal by the first single RAID subsystem to the core connecting the channels of the first single RAID subsystem with the channels of the second adjacent single RAID subsystem,

c. closing switching means in the core thereby establishing electrical connections between the channels of the first single RAID subsystem and the channels of the second adjacent single RAID subsystem,

d. identifying the channels of the second adjacent single RAID subsystem electrically connected to the channels of the first single RAID subsystem by the storage array controller of the first RAID subsystem,

e. appropriating the flow of data to and from the host computer of the second adjacent single RAID subsystem by the storage array controller of the first single RAID subsystem, and

f. controlling the direct access storage units in the channels of both the first single RAID subsystem and of the second adjacent single RAID subsystem by the storage array controller of the first single RAID subsystem.

17. In a redundant RAID system having a first and a second single RAID subsystem, each subsystem comprising a storage array controller and multiple channels of direct access storage devices, and a core with repeater switching means connecting the channels of the first single RAID subsystem with the corresponding channels of the second single RAID subsystem, the process of controlling the arrays of direct access storage units of both single RAID subunits by the first single RAID subsystem when the second single RAID subsystem has a faulty storage array controller comprising the steps:

a. ceasing to emit a normal operating signal or heartbeat by the storage array controller of the second single RAID subsystem having a faulty storage array controller,

b. noting the cessation of emission of a normal operating signal or heartbeat by the storage array controller of the second single RAID subsystem having a faulty storage array controller by the first single RAID subsystem, and emitting a close signal by the first single RAID subsystem to the core,

c. closing switching means in the core thereby establishing electrical connections between the channels of the first single RAID subsystem and the channels of the second single RAID subsystem,

d. appropriating the flow of data to and from the host computer of the second single RAID subsystem by the host computer of the first single RAID subsystem, and

e. controlling the direct access storage units in the channels of both the first single RAID subsystem and of the second single RAID subsystem by the storage array controller of the first single RAID subsystem.