Ncr Corporation

Oracle America, Inc.

Kit M Chow

P Keith Muller

Michael W Meyer

Alan P Adamson

Weinan William Tseng

George R Cameron

Blake Alan Jones

Gary Lee Boggs

Chinh Kim Nguyen

Gregory DuVall Bruno

Morten Sk&oslash;ien With

Gregory Du Bruno

Niels Haahr Hornek&aelig;r

A method involving receiving an indication of a requirement to allocate at least one page for a process, where pages are associated with cache colors; generating a selection bitmap by performing a logical operation of a system available colors bitmap and a process bitmap, where the system available colors bitmap and the process bitmap each include one bit corresponding to each cache color, where each bit of the system available colors bitmap indicates whether a number of pages associated with a corresponding cache color that are available to be allocated is above a minimum threshold, and where each bit of the process bitmap indicates whether any pages associated with the corresponding cache color have been recently allocated for the process. The method also includes selecting, using the selection bitmap, a cache color; and allocating a page for the process, wherein the allocated page is associated with the selected cache color.

System available cache color map

A method and apparatus for write-back caching in a data storage and processing system has been described. The method comprises the steps of receiving a write request including write data from a compute node in a first I/O node, forwarding the write data from the first I/O node to a second I/O node, and sending an acknowledgment message to the compute node from the second I/O node after the write data is received by the second I/O node. After the data is written into non-volatile storage of the first I/O node, a purge request or command is sent to the second I/O node to purge the write data from the volatile memory of the second I/O node. In one embodiment, the purge request is not sent until the first I/O node receives a second write request, in which case, the purge request is sent in the same interrupt as the write data for the second write request. The processing system comprises a first and a second I/O node, each with means for receiving a write request from the compute node and forwarding that write data to the other I/O node. Each I/O node also comprises a means for sending an acknowledgment message back to the compute node directly, without sending the acknowledgment through the I/O node that sent the write data. The result is an I/O protocol that reduces the number of interrupts required to store data, while still implementing write back caching to improve storage speed and turnaround. The invention also can be described in terms of a program storage device, such as a hard disk, floppy disk, or a CD, which tangibly embodies instructions stored thereon for performing the instructions to practice the invention.

Method and apparatus for write-back caching with minimal interrupts

A method, apparatus, and article of manufacture for dynamically binding shared resources among I/O nodes is disclosed. The method comprises the steps of de-allocating resources requested by an initiating node from a responding node, allocating resources not requested by the initiating node and reachable by the responding node to the responding node, de-allocating resources allocated to the second node from the first node, and allocating unallocated resources reachable by the first node to the first node. The article of manufacture comprises a program storage device tangibly embodying program steps executable by a computer for performing the foregoing method steps. The apparatus comprises a data storage resource having a plurality of storage resources, a first I/O node and a second I/O node. The first and second I/O nodes have an I/O processor for transceiving resource ownership negotiation messages, and for de-allocating and allocating resources as indicated in the information received in the ownership negotiation messages.

Protocol for dynamic binding of shared resources

A distributed service subsystem comprises one or more cabinets containing one or more computer systems, one or more administration consoles for controlling the computer systems in the cabinets, and a network interconnecting the administration consoles and the cabinets. Each of the cabinets includes a cabinet module interface controller (CMIC) coupled to and controlling one or more chassis management boards (CMBs). The CMBs are each adapted to communicate with one or more managed components in the cabinet through a component specific interface.

Distributed service subsystem architecture for distributed network management

A method and apparatus for communicating data in a highly distributed parallel processing computer architecture is described. The method comprises the steps of generating a globally unique ID in the I/O node for a data extent physically stored in the plurality of storage devices, binding the globally unique ID to the data extent, and exporting the globally unique ID to the compute nodes via the interconnect fabric. In one embodiment, the globally unique ID is generated from a globally unique I/O node identifier and a locally unique data extent identifier. A local entry point is generated in the compute node for the data associated with the globally unique ID, thereby presenting the globally unique ID as a device point in the compute node. In one embodiment, the step of exporting the globally unique ID to the compute nodes comprises the step of receiving a message from the compute node comprising a signature securely identifying it to the I/O node, authenticating the source of the message using the signature, and transmitting the globally unique ID comprising data specifying local access rights to the data represented by the globally unique ID from the I/O node to the compute node.

Name service for multinode system segmented into I/O and compute nodes, generating guid at I/O node and exporting guid to compute nodes via interconnect fabric

A highly-scalable parallel processing computer system architecture is described. The parallel processing system comprises a plurality of compute nodes for executing applications, a plurality of I/O nodes, each communicatively coupled to a plurality of storage resources, and an interconnect fabric providing communication between any of the compute nodes and any of the I/O nodes. The interconnect fabric comprises a network for connecting the compute nodes and the I/O nodes, the network comprising a plurality of switch nodes arranged into more than g(log,N) switch node stages, wherein b is a total number of switch node input/output ports, and g(x) indicates a ceiling function providing the smallest integer not less than the argument x, the switch node stages thereby providing a plurality of paths between any network input port and network output port.

Highly-scalable parallel processing computer system architecture

The present invention describes a parallel processing system. The system comprises a plurality of compute nodes for executing applications via a storage application interface having system input/output calls, a plurality of I/O nodes, and a file system implemented in the compute node, for storing information mapping API system input/output calls for the data object with the globally unique identification for the data object. Each I/O node manages a communicatively coupled plurality of storage resources and each has a means for generating a globally unique identification for a data object stored on the storage resource and transmits the globally unique identification and the data object to the compute node via at least one interconnecting fabric providing communication between any of the compute node and any of the I/O nodes.

Dynamic and consistent naming of fabric attached storage by a file

Distributed service subsystem protocol for distributed network management

A method for transferring data from a first node to a second node in a multi-processor system is described. The multi-processor system comprises a plurality of nodes coupled to an interconnect fabric via an interconnect fabric interface, the nodes comprising a compute node and an I/O node, the I/O node coupled to a plurality of data storage devices. The method comprises the steps of generating a I/O request packet in a first node in response to an I/O request from an application executing in a first node, transmitting the data request packet to the second node via the interconnect fabric, and executing the destination interconnect channel program to extract the debit ID to transfer the data request to the second node buffer. In one embodiment, the I/O packet includes a data transfer request, an interconnect destination channel program, a first debit ID designating a second node buffer where the data request will be transmitted, and a first credit ID designating a first node buffer where the data responsive to the I/O request will be transmitted.

I/O protocol for highly configurable multi-node processing system

An enhanced interface between a processor and an Asynchronous Transfer Mode (ATM) segmentation controller. A software driver communicates with the controller to efficiently segment protocol data units (PDUs) into ATM cells for transmission. The software driver creates a series of data structures in memory for managing the PDU, wherein the data structures include a linked list of buffer descriptors that describe the size and location of subportions of the PDU. The ATM controller is capable of accessing and interpreting these data structures. As a result, only a single I/O operation is required for the software driver to initiate direct memory access (DMA) operations by the ATM controller to retrieve the data structures from memory and subsequently transmit the PDU.

Growing community of inventors

Carlsbad, CA, United States of America

Kit M Chow