Dell USA L P

Dell Products Inc.

Other

Kenneth L Jeffries

Craig S Jones

Terry J Parks

Daniel L Bouvier

Victor K Pecone

Joseph M Maurin

A serial storage architecture (SSA) storage subsystem which includes an SSA initiator and a series of SSA targets which include disk drives housed in a storage pod and an intelligent backplane of the storage pod which houses the other SSA targets. The intelligent backplane includes a controller, a series of status registers, each indicating status of an operating parameter for the SSA target drives, and a series of control registers, each configured to transmit a respective command to selected ones of said SSA target drives. The controller polls each of the status registers and, based on the contents thereof, determines whether an event relating to the operating parameter has occurred. If so, the controller reports the event to the SSA initiator. The SSA initiator may also control the target drives using the control registers.

Intelligent backplane for serial storage architectures method and system

Intelligent backplane for serial storage architectures

Intelligent backplane for collecting and reporting information in an SSA

A computer system including a bus master performs a method for automatically assigning addresses to agents on a bus. Addresses are automatically assigned so that a computer system user does not set physical or logical switches, either manually or through software programming. The system and method also automatically assign unique addresses to new devices that are inserted on the bus while the bus is operating, thereby supporting 'hot pluggable' devices. Slave agents are originally configured to operate at a class address. At the beginning of the method, a master determines whether any of the slaves reside at the class address. If so, then the master determines a new unique address and issues a Get Bitwise UID command to the slaves residing at the class address. Each of the slaves receives the Get Bitwise UID and responsively transmits a hardware identification (UID) in a loop of bitwise byte transmissions. One slave successfully transmits the hardware identification (UID) while the other individual slaves detect transmission errors during transmission of the hardware identification (UID) and terminate transmission upon the error detection. The unsuccessful slaves, if previously residing at a unique address, revert to the class address. The master receives the hardware identification (UID) of the successfully transmitting slave and issues a Set Address command in combination with the hardware identification (UID) and the new unique address to all slave agents at the class address. Only the slave agent with a hardware identification (UID) matching the UID of the Set Address command processes the Set Address command to completion and is assigned the address. The master uses the Get Bitwise UID command followed by the Set Address command in a loop to locate, identify and assign addresses to agents responding to a predetermined class address.

System and method for assigning unique addresses to agents on a system

A disk controller for a disk drive array which maintains two representations of all drive defects. The controller maintains a logical defect list that is used to maintain the sector remapping structure when reconstructing redundancy information. The controller also maintains a physical defect list that is used to preserve known defect information on a physical disk basis. The physical defect list stores the defects even if the logical configuration of the disks changes. When the controller of the present invention determines that a block of data is bad, the controller allocates space for the respective stripe in an alternate block, recovers the data in the stripe and writes the recovered data to the newly allocated stripe. The controller then updates the remap tables in memory with the remap information. On each disk access, the controller searches the logical defect list to determine if the access involves one or more bad blocks. When a failed disk is replaced, the controller rebuilds the data from the failed drive using the remaining data and parity. The controller also uses both the logical and physical defect lists to unmap remapped sectors which were originally remapped due to defective sectors on the replaced disk drive.

Method and controller for defect tracking in a redundant array

A serial storage architecture (SSA) storage subsystem includes a backplane, first and second storage devices, each equipped with a link resiliency circuit, connected to the backplane, an SSA initiator and a series of SSA targets, which include a backplane of the storage pod and the first and second storage devices, serially connected to the SSA initiator. Control and status registers are coupled to the link resiliency circuits. The control register transmits control signals, such as time required for the link resiliency circuits to heal line faults, to the link resiliency circuits in response to commands issued by the SSA initiator. The status register is polled by a controller to the presence of line faults detected by the link resiliency circuits. If so, the detected line faults are reported to the SSA initiator.

Processor controlled link resiliency circuit for serial storage

Apparatus and method for implementing a local proactive hot plug request/acknowledge scheme is disclosed. In a preferred embodiment, each hot pluggable device installable on a computer bus, such as a SCSI bus, is provided with a physical user interface comprising a mechanical request initiator, such as a button or two-position switch, for allowing a user to generate a hot swap request to a controller associated with the bus prior to actual installation of the device on, or removal of the device from, the bus. Upon receipt of the request, the controller determines whether the requested action may be performed, provides a visual indication of its determination to the user via an LED on the user interface and, if installation or removal is determined to be prudent, performs the hot installation/removal in an orderly manner so as not to adversely affect ongoing system operations.

Local proactive hot swap request/acknowledge system

A system and method for automatically assigning addresses to agents on a system management bus in a computer system without requiring user intervention, i.e., without requiring the user to manually or programmatically set physical or logical switches. The computer system includes a system management bus which preferably uses the I.sup.2 C serial protocol. The bus includes at least one SMB master and a plurality of slaves for performing desired monitoring and control functions in the computer system. According to the present invention, the SMB master assigns unique addresses to each of the SMB slaves automatically and without user intervention. This provides a simpler, more efficient, and less error prone method for assigning addresses to SMB agents. In addition, the system of the present invention automatically assigns unique addresses to new devices inserted on the bus while the bus is operating and thus allows for hot pluggable devices.

Systems and method for assigning unique addresses to agents on a system

A disk drive array with a controller which provides: dynamic remapping for grown defects in the disk drives, multi-thread request processing with a variable number of forkings, defect tracking with both logical and physical lists, guarded writes of less than a full stripe optimized by selectably using the redundancy to limit the number of sectors involved, association of multiple operations with a single disk request in order to facilitate error handling, use of an access hiatus as indication of further opportunity to rebuild data in background, and scatter/scatter (bidirectional scatter/gather) operations.

System for scheduling read ahead operations if new request is sequential

Disk drive controller accepting first commands for accessing composite

An improved read ahead strategy that improves the performance of a disk array subsystem. The disk controller keeps track of the last n reads to the array. If a new read request is received that is adjacent to any of the last n reads, the controller performs a look ahead read because a sequential read may be in progress. The parameter n is preferably set comparable to or greater than the number of maximum independent activites being performed by the computer system. Therefore, in a multithread system, the controller performs a readahead if any one thread is doing a sequential read.

System for scheduling readahead operations if new request is within a

A disk drive array including a controller which provides scatter/scatter (bi-directional scatter/gather) operations between noncontiguous host memory address locations and noncontiguous disk address locations. The host provides a single request to launch a scatter/scatter transfer. The single data request includes a pointer to a list of transfer counts and addresses, the length of the list, and the starting logical address on the disk transfer. Skipped blocks in a scatter/scatter request are specified by data address value of -1, and a no-operation (no-op) request is enqueued for each skipped block. Thus, during reads the controller extends the scattered read from the disk into a single large read of contiguous sectors and suppresses the unwanted data by inserting 'no-operation' commands in place of the read commands during the transfer to the host.

Controller for receiving transfer requests for noncontiguous sectors and

A disk controller which fragments host requests into atomic requests or one cycle operations prior to execution. The disk controller is coupled to a disk array and receives requests from a host CPU. Host requests are fragmented down to atomic operations. This is highly advantageous for error handling because the controller is not required to determine which phase of an operation is being requested, since every operation is single phase. The present invention includes use of fence markers or execution-control markers to 'fence' of mark a block of tasks and ensure that the atomic operations are executed in sequence, if necessary. These markers ensure that related sequences of atomic operations are kept together without, for example, possibly disruptive intervening writes.

Disk drive array with request fragmentation

An apparatus and method for a computer system to rapidly access at least two IDE disk drives. Use of standard forty pin connectors and forty wire ribbon cable having certain pairs of wires uniquely twisted so as to allow the system to independently access the IDE drives.

Growing community of inventors

Leander, TX, United States of America

Kenneth L Jeffries