Abstract: A network storage appliance includes a chassis, enclosing a storage controller and first and second servers. The storage controller has first and second I/O ports for coupling to first and second I/O links. The storage controller controls a plurality of physical disk drives and presents the plurality of physical disk drives as one or more logical disk drives on the first and second I/O links. The servers each have an I/O port for coupling to a respective one of the first and second I/O links. Each of the servers transmits packets to the storage controller over the respective I/O link. The packets include block-level protocol disk commands each identifying one of the logical disk drives, such as SCSI block level protocol commands each identifying one of said logical disk drives as a SCSI logical unit. The I/O links may be FibreChannel, Ethernet, or Infiniband links, for example.

Abstract: A method for transferring data within a network storage appliance is disclosed. The method includes transmitting a packet on an I/O link from a server to a first portion of a storage controller. Transmitting the packet on the I/O link is performed within a single blade module in a chassis enclosing the storage appliance. The method also includes forwarding a data transfer command within the packet from the first portion of the storage controller to a second portion of the storage controller. Forwarding the data transfer command is performed via a local bus on a backplane of the chassis through a connector of the blade connecting the blade to the backplane.

Abstract: A server blade includes a printed circuit board (PCB), including a connector for connecting the blade to a backplane comprising a local bus, and a removal mechanism for use by a person to disconnect the connector from the backplane for removal of the blade from a chassis while the chassis is powered up. The server blade also includes an I/O link and a server, each affixed on the PCB. The server transmits packets on the I/O link to a storage controller enclosed in the chassis. The packets include commands to transfer data to at least one storage device controlled by the storage controller. A portion of the storage controller, affixed on the PCB, receives the packets from the server on the I/O link, and forwards the commands on the backplane local bus to another portion of the storage controller affixed on a separate PCB enclosed in the chassis.

Abstract: A high data availability write-caching storage controller has a volatile memory with a write cache for caching write cache data, a non-volatile memory, a capacitor pack for supplying power for backing up the write cache to the non-volatile memory in response to a loss of main power, and a CPU that determines whether reducing an operating voltage of the capacitor pack to a new value would cause the capacitor pack to be storing less energy than required for backing up the current size write cache to the non-volatile memory. If so, the CPU reduces the size of the write cache prior to reducing the operating voltage. The CPU estimates the capacity of the capacitor pack to store the required energy based on a history of operational temperature and voltage readings of the capacitor pack, such as on an accumulated normalized running time and warranted lifetime of the capacitor pack.

Abstract: A storage controller has a capacitor pack for storing energy to supply power during a main power loss, a temperature sensor that senses the capacitor pack temperature, and a CPU, which detects that the temperature of the capacitor pack has risen above a predetermined threshold while operating at a first voltage value and determines whether a projected lifetime of the capacitor pack is less than the warranted lifetime. If the projected lifetime is less than the warranted lifetime, the CPU reduces the operating voltage of the capacitor pack to a second value, in order to increase the capacitor pack lifetime. In one embodiment, the CPU reduces the voltage if an accumulated normalized running time of the capacitor pack is greater than an accumulated calendar running time. In another embodiment, the CPU reduces the voltage if a percentage capacitance drop of the capacitor pack is greater than a calendar percentage capacitance drop.

Abstract: A storage controller has a capacitor pack for storing energy to supply during a main power loss, a temperature sensor that senses the capacitor pack temperature, and a CPU, which repeatedly: receives the temperature during an interval over which the capacitor pack is operated, determines a lifetime over which the capacitor pack would have a capacity to store at least a predetermined amount of energy if operated at the temperature during the lifetime, normalizes the interval by a ratio of a warranted lifetime of the capacitor pack relative to the determined lifetime, and adds the normalized interval to an accumulated normalized running time. The operating voltage of the capacitor pack may also sampled and used to determine the lifetime. The predetermined amount of energy may be for backing up a volatile write cache to a non-volatile memory in response to the loss of main power.

Abstract: A method for adopting an orphaned I/O port of a storage controller is disclosed. The storage controller has first and second redundant field-replaceable units (FRU) for processing I/O requests and a third FRU having at least one I/O port for receiving the I/O requests from host computers coupled to it. Initially the first FRU processes the I/O requests received by the I/O port and the third FRU routes to the first FRU interrupt requests generated by the I/O port in response to receiving the I/O requests. Subsequently, the second FRU determines that the first FRU has failed and is no longer processing I/O requests received by the I/O port, and configures the third FRU to route the interrupt requests from the I/O port to the second FRU rather than the first FRU, in response to the determining that the first FRU has failed.

Abstract: A write-caching RAID controller is disclosed. The controller includes a CPU that manages transfers of posted-write data from host computers to a volatile memory and transfers of the posted-write data from the volatile memory to storage devices when a main power source is supplying power to the RAID controller. A memory controller flushes the posted-write data from the volatile memory to the non-volatile memory when main power fails, during which time capacitors provide power to the memory controller, volatile memory, and non-volatile memory, but not to the CPU, in order to reduce the energy storage requirements of the capacitors. During main power provision, the CPU programs the memory controller with information needed to perform the flush operation, such as the location and size of the posted-write data in the volatile memory and various flush operation characteristics.

Abstract: A network storage appliance including one or more integrated switching devices is disclosed. The appliance includes redundant storage controllers that transfer frames of data between storage devices and host computers. The integrated switching devices include a plurality of I/O ports and a data transfer path between each of the I/O ports for providing simultaneous data transfers between multiple pairs thereof. The switches enable the appliance to simultaneously transfer frames between its I/O ports and storage device I/O ports and/or host I/O ports, thereby providing increased data transfer bandwidth over arbitrated loop configurations. Additionally, the switches are intelligent and may be programmed to achieve improved fault isolation. The appliance may also include servers that include I/O ports coupled to the switches for simultaneously transferring data with the storage controllers and/or I/O ports of devices external to the appliance.

Abstract: A storage controller configured to adopt orphaned I/O ports is disclosed. The controller includes multiple field-replaceable units (FRUs) that plug into a backplane having local buses. At least two of the FRUs have microprocessors and memory for processing I/O requests received from host computers for accessing storage devices controlled by the controller. Other of the FRUs include I/O ports for receiving the requests from the hosts and bus bridges for bridging the I/O ports to the backplane local buses in such a manner that if one of the processing FRUs fails, the surviving processing FRU detects the failure and responsively adopts the I/O ports previously serviced by the failed FRU to service the subsequently received I/O requests on the adopted I/O ports. The I/O port FRUs also include I/O ports for transferring data with the storage devices that are also adopted by the surviving processing FRU.

Abstract: A network storage appliance is disclosed. The storage appliance includes a port combiner that provides data communication between at least first, second, and third I/O ports; a storage controller that controls storage devices and includes the first I/O port; a server having the second I/O port; and an I/O connector for networking the third I/O port to the port combiner. A single chassis encloses the port combiner, storage controller, and server, and the I/O connector is affixed on the storage appliance. The third I/O port is external to the chassis and is not enclosed therein. In various embodiments, the port combiner comprises a FibreChannel hub comprising a series of loop resiliency circuits, or a FibreChannel, Ethernet, or Infiniband switch. In one embodiment, the port combiner, I/O ports, and server are all comprised in a single blade module for plugging into a backplane of the chassis.

Abstract: A network storage appliance is disclosed. The appliance includes a single chassis that encloses a plurality of servers and a plurality of storage controllers coupled together via a chassis backplane. The storage controllers control the transfer of data between the plurality of servers and a plurality of storage devices coupled to the storage controllers. The servers and storage controllers include a plurality of field replaceable unit (FRUs) hot-pluggable into the backplane such that any one of the FRUs may fail without loss of availability to the storage devices' data. In various embodiments, the chassis fits in a 19? wide rack; is 1U high; the servers are standard PCs configured to execute off-the-shelf server applications and to facilitate porting of popular operating systems with little modification; the servers include disk-on-chip memory rather than a hard drive; local buses (e.g., PCIX) on the backplane interface the various FRUs.

Abstract: An apparatus and method for deterministically killing one of redundant servers on a common network is disclosed. The apparatus includes a chassis that encloses the servers and a storage controller, status indicators generated by the servers to the storage controller, and kill controls, generated by the storage controller to respective ones of the servers, each for killing a respective one of the servers. The status indicators and kill controls are wholly enclosed in the chassis. The kill controls deterministically disable the killed server on the network independently of the state of the server to be killed. That is, the server does not need to be able to respond to a command to be disabled on the network. In one embodiment, the kill controls comprise reset signals. After the storage controller deterministically kills one of the servers, the other server takes over the identity of the killed server on the network.

Abstract: An application server blade for an embedded storage appliance is disclosed. The blade includes a printed circuit board (PCB) with a connector for connecting to a chassis backplane including a local bus. Affixed on the PCB is a server, a portion of a storage controller, and an I/O link coupling the server and storage controller portion. The server transmits packets on the I/O link to the storage controller portion. The packets include commands to transfer data to a storage device controlled by the storage controller. The storage controller portion receives the packets from the server on the I/O link and forwards the commands on the backplane local bus to another portion of the storage controller affixed on a separate PCB also enclosed in the chassis. The blade also includes a removal mechanism for hot-replacement of the blade in the chassis. The blade architecture facilitates software reuse.

Abstract: A network storage appliance is disclosed. The appliance includes a chassis enclosing a backplane, and a server enclosed in the chassis and coupled to the backplane. The appliance also includes storage controllers enclosed in the chassis, each coupled to the backplane, which control transfer of data between the server and storage devices coupled to the storage controllers. The storage controllers also control transfer of data between the storage devices and computers networked to the appliance and external to the appliance. The storage controllers and the server comprise a plurality of hot-replaceable blades. Any one of the plurality of blades may be replaced during operation of the appliance without loss of access to the storage devices by the computers. In one embodiment, the server executes storage application software, such as backup software for backing up data on the storage devices, such as to a tape device networked to the server.

Abstract: An apparatus is disclosed for deterministically performing active-active failover of redundant servers in response to a failure of a link on which each server provides a heartbeat to the other server. Each of the servers is configured to take over the identity of the other server on a common network in response to detecting a failure of the other server's link heartbeat. Each server provides a status indicator to a storage controller indicating whether the other server's link heartbeat stopped. The storage controller determines the link has failed if both of the status indicators indicate the other server's heartbeat stopped, and responsively kills one of the servers. The storage controller also receives a heartbeat directly from each server. If only one direct heartbeat stops when the status indicators indicate the link heartbeats stopped, then the storage controller detects one server has failed and inactivates the failed server.

Abstract: A SCSI controller card which includes a standard SCSI connector that can support either one or two SCSI channels, as desired. The connector can receive either a standard single channel cable or a specially designed dual channel "Y" cable to provide either one or two SCSI channels, respectively. The SCSI controller card includes logic that determines which cable is installed and switching logic which routes one or two SCSI channels to the connector accordingly. In the preferred embodiment, the SCSI adapter card includes two SCSI controllers which provide two SCSI channels. A first channel is connected directly to pins on the SCSI connector. A second channel is connected to switching logic, and a plurality of ground signals are also connected to the switching logic. The second channel is switched into the SCSI connector if the dual channel "Y" cable is attached to the connector, and the ground signals are switched into the SCSI connector if the "Y" cable is not detected.

Abstract: An apparatus and method for address pipelining of a computer system that reduce the average number of wait states required to access a dynamic random access memory (DRAM) subsystem. A memory controller addresses a plurality of random access memory integrated circuits in pages of addresses wherein contiguous address pages are in different ones of the plurality of dynamic random access memory integrated circuits. Transparent latches associated with each of the different ones of the plurality of dynamic random access memory integrated circuits allow pipelining of address setups for more than one memory page at substantially the same time. The apparatus and method improve the write access times of a computer system and, when used with a computer system having address pipelining, both read and write accesses are improved because address set up latency time is reduced.

Abstract: A SCSI adapter card which provides one or more internal SCSI channels and includes connectors for an optional daughter card which provides an external SCSI connector. The daughter card is a parallel mezzanine style daughter board which provides modular and upgradable SCSI bus routing options. In the preferred embodiment, the adapter card includes two SCSI controllers which provide two internal SCSI channels. The daughter board can include up to 2 SCSI controllers for additional SCSI channels. The daughter board can reroute one or more of the internal SCSI channels to the external connector according to various SCSI routing options or can include one or more SCSI controllers for additional SCSI channels. In one embodiment, the daughter board does not include a SCSI controller, but rather serves to reroute one or more of the internal SCSI controllers to the external connector.

Abstract: A computer is provided with a redundant power supply system using a plurality of power supply boxes having generally conventional, off-the-shelf configurations. The computer includes a housing having a wall opening spaced apart from and facing the side of a paralleling type circuit board having mounted thereon a plurality of AC electrical connectors and a spaced plurality of DC electrical connectors. The power supply boxes are mounted on carrier structures slidably received in the computer housing for drawer-like movement through the housing wall opening toward and away from the circuit boards. Each power supply box has an external wire bundle with blind mate DC electrical connectors mounted on the outer ends of its leads and supported by the carrier structure for mating connection with their associated circuit board connectors when the carrier structure is moved into adjacency with its associated circuit board.