International Business Machines Corporation

Paul D DiNicola

Chulho Kim

Rama Krishna Govindaraju

Gautam H Shah

Carl Alfred Bender

Myung Mun Bae

Kevin J Gildea

Rajeev Sivaram

Jamshed H Mirza

David Andrew Rice

John J Lawless

Omar M Rahim

Joseph C Kantz

Edward M Ruddick

Mandayam Thondanur Raghunath

Leonard W Helmer, Jr

Gregory Salyer

Steven J Martin

Patricia E Heywood

Carol L Soto

Francois N Dumas

Fai Chu

Charles L Bain

Jaroslaw Nieplocha

Globally consistent information is locally provided to communications layers. Globally consistent information is stored in a Network Availability Matrix, which is locally accessible by a communications layer. If an event is detected, the communications layer is automatically notified by the Network Availability Matrix, and is able to use the information in the Network Availability Matrix to quickly take action.

Locally providing globally consistent information to communications layers

A system and method that utilizes a dedicated transmission queue to enable expedited transmission of data messages to adaptive 'nearest neighbor' nodes within a cluster. Packet descriptors are pre-fetched by the communications adapter hardware during the transmission of the preceding data element and setup for the next transmission is performed in parallel with the transmission of the preceding data element. Data elements of a fixed length that is equal to the cache line size of the communication hardware can optionally be used to provide optimized transfer between computer memory and communications hardware. The data receiving processing can also be optimized to recognize and handle cache line size data elements.

Speculative method and system for rapid data communications

Method, system and program storage device are provided for monitoring and ameliorating congestion in a tightly coupled network. Commensurate with sending a packet into the network, a first time stamp is recorded. Upon receipt of an acknowledgment back across the network responsive to sending of the packet, a second time stamp is recorded. The round trip time of the packet is determined and an amount of congestion is estimated using the determined round trip time and a statically predetermined round trip representative of at least one of no network congestion or a known degree of network congestion. The number of flow control tokens for the destination node can be dynamically varied in response to the estimate of the amount of network congestion. If desired, monitoring and estimating of network congestion can be initiated only after identifying the existence of network congestion, for example, represented by a lack of flow control tokens at a sender node for a destination node.

Congestion monitoring and message flow control in a blocking network

Method, apparatus and program product for communicating from a node to a communications device. A Hardware Abstraction Layer (HAL) provides functions which can be called from user space in a node to access the communications device. An instance of HAL is created in the node. Device specific characteristics from the communications device and a pointer pointing to HAL functions for accessing the communications device are obtained by HAL. HAL then opens multiple ports on the communications device using the functions pointed to by the pointer, and messages are sent between the node and the communications device. The messages thus sent are optimized with respect to the communications device as determined by the obtained device specific characteristics. Multiple processes and protocol stacks may be associated with each port in a single instance of HAL. A further embodiment provides that multiple virtual ports may be associated with a port, with a multiple protocol stacks associated with each virtual port. A further embodiment provides that multiple communications devices may be associated with a single instance of HAL.

Hardware interface between a switch adapter and a communications

A method, apparatus and program product for detecting a communication event in a distributed parallel data processing system in which a message is sent from an origin to a target. A low-level application programming interface (LAPI) is provided which has an operation for associating a counter with a communication event to be detected. The LAPI increments the counter upon the occurrence of the communication event. The number in the counter is monitored, and when the number increases, the event is detected. A completion counter in the origin is associated with the completion of a message being sent from the origin to the target. When the message is completed, LAPI increments the completion counter such that monitoring the completion counter detects the completion of the message. The completion counter may be used to insure that a first message has been sent from the origin to the target and completed before a second message is sent.

Signaling communication events in a computer network

A method, apparatus and program product for message communication in a distributed parallel data processing system. A user message is sent from a sender to a receiver. The user message contains user data and a pointer to a header handler routine. The header handler routine includes a first pointer to a target user buffer and a second pointer to a completion routine. When the user message is received, a low level application program interface (LAPI) is informed which invokes the header handler routines which returns the first and second pointers. LAPI then transfers the user data to the user buffer indicated by the header handler routine, and invokes the completion routine indicated by the header handler routine to complete the transfer of the user message to the receiver.

Method and apparatus for efficient communications using active messages

In a computer graphics system, a method and apparatus for scaling a line pattern having a predetermined first number of pixels to an actual line having a predetermined second number of pixels different from said first number. The total line length is divided by the line pattern length to obtain the number of line pattern repetitions and the number of remainder pixels in the actual line per line pattern repetition. An error term is incremented by the number of remainder pixels per repetition for each pixel of a given line pattern repetition that is processed. When the cumulative error term equals the line pattern length, the error term is decremented by the line pattern length and an extra pixel is generated. For lines that are shorter than the line pattern, a similar procedure is used except that when the cumulative error term equals the line pattern length, a pixel is deleted. Lines are drawn upon the completion of pattern segments rather than as successive pixel positions are generated to increase the drawing efficiency.

Method and apparatus for scaling line patterns

In a graphics subsystem, a highly interactive two-dimensional (2D) data stream and a computationally intensive three-dimensional (3D) data stream are processed concurrently in such a manner that processing of the 2D data stream is not held up by processing of the 3D data stream. A 3D geometry subsystem having a parallel pipeline architecture is used to process the 3D data stream, while a 2D subsystem concurrently processed the 2D data stream in parallel with the 3D subsystem. A reordering device couples the processed 2D and 3D data streams to a common raster subsystem. The reordering device, which contains an internal buffer, reorders any order-dependent elements of the 3D data stream appearing at the output of the 3D geometry subsystem in an order different from the order in which they were supplied to the input end. The reordering device prioritizes the 2D data stream relative to the 3D data stream so that elements of the 2D data stream arriving from the 2D subsystem are passed to the raster subsystem almost immediately, without having to wait for elements of the 3D data stream.

Method and apparatus for processing two graphics data streams in parallel

The data processing system includes processing nodes and a graphics display device for processing a graphics data stream. The data processing system partitions a graphics data stream into a data segments or work groups for processing by the processing nodes. Next, the data segments are distributed for processing to the processing nodes. In response to receiving a data segment at a processing node, the data segment is processed to produce a processed data segment. The processed data segments are recombined into a processed graphics data stream. This processed graphics data stream is coupled to the graphics display device.

Method and system for processing graphics data streams utilizing

A graphic display system with the ability to use mulitiple memory buffers to produce images with a wide range of colors through bit plane encoding or to present independent application displays or to combine display images through the use of lateral bit encoding. When operated in the lateral bit encoded state, application programs can be associated with independent memory buffers or an application can use the separate buffers to create a display with animation or apparent movement. Each memory buffer can be independently associated with the display device or the images contained in the memory buffers can be mixed through the use of hardware or software image mixing to create a composite display. The combined image is use to directly control the display device and does not require the creation of an intermediate frame buffer image. This display system provides the capability for animation or image movement through the designation of one or more planes to contain the objects and the designation of display priority among the memory buffers. The image mixer combines the images according to an established display priority so that portions of the highest priority image are always displayed. The images in the memory buffers may, alternatively, be linked to create a single large image which can be scrolled across the display monitor. A method for image mixing and for displaying objects with apparent motion is provided through the use of the multiple memory buffers and display priority assignment.

Growing community of inventors

Hurley, NY, United States of America

Paul D DiNicola