International Business Machines Corporation

Glenn David Gilda

Patrick James Meaney

John Steven Dodson

Vesselina K Papazova

Eric Eugene Retter

Mark R Hodges

Gary A Van Huben

Christian Jacobi

Stephen J Powell

Arthur J O'Neill

David D Cadigan

Lawrence D Curley

Timothy Carl Bronson

Brad William Michael

Irving G Baysah

Lawrence Jones

Phillip G Williams

Bruce Leroy Beukema

Paul Allen Ganfield

Steven Lee Gregor

Ronald Edward Fuhs

Jentje Leenstra

Scot H Rider

James Charles Gregerson

Ashutosh Mishra

Logan I Friedman

Robert J Sonnelitter, Iii

Craig Raymond Walters

Barry Marshall Trager

Eric M Foster

Anthony J Bybell

Gregory J Fredeman

Hubert Harrer

Manuel Joseph Alvarez, Ii

Adrian Stephen Butter

Thomas J Dewkett

Michael B Spear

Rajat Rao

Stefan P Jackowski

William Garrett Verdoorn, Jr

Hussain Shaukatullah

Jason Andrew Thompson

Thomas E Miller

John M Sheplock

Terrence A Quinn

Don L Baker

Yvonne Hanson Kleppel

Xiaomin Duan

Michael Todd Fisher

Jamee Abdulhafiz

Michael Jason Cade

Concurrent servicing of a first cable of a cable pair while a second cable of the cable pair remains operational improves multi-processor computer system availability and serviceability. A pair of processing chips in different processing drawers may continue operation by way of the second cable while the first cable is degraded or serviced. Upon the servicing of the first cable, the serviced cable may transition to a fully operational state with no interruptions to the operations between the processing drawers by way of the second cable. Since cable faults are typically more common than processing chip or processing drawer faults, identification of cable faults and the ability to maintain operations of the processing drawers connected therewith is increasingly important.

Cable pair concurrent servicing

Embodiments of the invention are directed to a computer-implemented method of operating a data transmission system. The data transmission system includes a transmitter and a receiver. The computer-implemented method includes using the transmitter to send serialized data from the transmitter through a plurality of lanes to the receiver. The transmitter sends the serialized data at a first serialization ratio. The receiver is configured to receive and load the serialized data at a second deserialization ratio, wherein the first serialization ration is greater than the second deserialization ratio.

Low-latency deserializer having fine granularity and defective-lane compensation

A computer-implemented method includes using a transmitter to send data from the transmitter through a plurality of lanes to a receiver using a synchronous operation mode that includes sending the data from the transmitter through the plurality of lanes to the receiver in a synchronous transmission manner that relies on an alignment between a transmitter clock frequency and a receiver clock frequency. A synchronous operation performance analysis (SOPA) is performed during the synchronous operation mode. A switch from the synchronous operation mode to an asynchronous operation mode is made based on a result of performing the SOPA. The asynchronous operation mode includes sending the data from the transmitter through the plurality of lanes to the receiver without requiring alignment between the transmitter clock frequency and the receiver clock frequency.

Low-latency, high-availability and high-speed SerDes interface having multiple synchronization modes

A computer-implemented method includes fetching, by a controller, data using a plurality of memory channels of a memory system. The method further includes detecting, by the controller, that a first memory channel of the plurality of memory channels has not returned data. The method further includes marking, by the controller, the first memory channel from the plurality of memory channels as unavailable. The method further includes, in response to a fetch, reconstructing, by the controller, fetch data based on data received from all memory channels other than the first memory channel.

Low latency availability in degraded redundant array of independent memory

A computer-implemented method includes refreshing a set of memory channels in a memory system substantially simultaneously, each memory channel refreshing a rank that is distinct from each of the other ranks being refreshed. Further, the method includes marking a memory channel from the set of memory channels as being unavailable for the rank being refreshed in the memory channel. In one or more examples, the method further includes blocking a fetch command to the memory channel for the rank being refreshed in the memory channel.

Refresh-hiding memory system staggered refresh

A computer-implemented method for memory macro disablement in a cache memory includes identifying a defective portion of a memory macro of a cache memory bank. The method includes iteratively testing each line of the memory macro, the testing including attempting at least one write operation at each line of the memory macro. The method further includes determining that an error occurred during the testing. The method further includes, in response to determining the memory macro as being defective, disabling write operations for a portion of the cache memory bank that includes the memory macro by generating a logical mask that includes at least bits comprising a compartment bit, and read address bits.

Cache array macro micro-masking

Embodiments include methods, systems and circuits for operating an error trapping logic circuit in a memory device. Aspects include receiving, during a first clock cycle, data and check bits for the data from a memory location and determining, during the first clock cycle, whether the data includes any error by calculating an error syndrome from the data and the check bits. Aspects also include determining, during a second clock cycle, a type of the error based on a full decoding of the error syndrome. Aspects further include determining whether to store the data, the check bits and the error syndrome in trap registers of the error trapping logic circuit based on an operating mode of the error trapping logic circuit and the type of the error.

Error trapping in memory structures

A processor implemented method for spreading data traffic across memory controllers with respect to conditions is provided. The processor implemented method includes determining whether the memory controllers are balanced. The processor implemented method includes executing a conditional spreading with respect to the conditions when the memory controllers are determined as unbalanced. The processor implemented method includes executing an equal spreading when the memory controllers are determined as balanced.

Conditional memory spreading for heterogeneous memory sizes

Systems, methods, and computer-readable media are disclosed for performing reduced latency error decoding using a reduced latency symbol error correction decoder that utilizes enumerated parallel multiplication in lieu of division and replaces general multiplication with constant multiplication. The use of parallel multiplication in lieu of division can provide reduced latency and replacement of general multiplication with constant multiplication allows for logic reduction. In addition, the reduced symbol error correction decoder can utilize decode term sharing which can yield a further reduction in decoder logic and a further latency improvement.

Reduced latency error correction decoding

Embodiments include techniques used for a power-reduced redundant array of independent memory RAIM system. The technique includes blocking commands to one or more memory modules of the RAIM system and reading data from one or more unblocked memory modules. The technique also includes applying a power channel mark for one or more blocked memory modules, the power channel mark indicating the one or more blocked memory modules to a decoder for error correction.

Power-reduced redundant array of independent memory (RAIM) system

In some embodiments, a computer-implemented method includes maintaining two or more error indicators for correctable errors occurring at two or more memory components. Each of the error indicators may be associated with a corresponding memory component. A correctable error may be detected as occurring during a first memory fetch operation at a first memory component. A first error indicator corresponding to the first memory component may be set, responsive to the correctable error at the first memory component. An uncorrectable error may be detected during a second memory fetch operation. It may be detected that the first error indicator is set. The first memory component may be marked, responsive to the uncorrectable error and to detecting that the first error indicator is set. The two or more error indicators for correctable errors may thus determine which memory component to mark due to the uncorrectable error.

Reducing uncorrectable errors based on a history of correctable errors

A computer-implemented method for managing a memory control unit includes receiving a command at the memory control unit. The command includes a command type that either requires or does not require buffering resources. The method further includes determining, via the memory control unit, a number of available memory tags from a first set of memory tags that are associated with the buffering resources. The method includes determining, via the memory control unit, a number of available memory tags from a second set of memory tags that are not associated with the buffering resources. The method also includes dynamically adjusting, via the memory control unit, assignment of memory tags for use in the second set of memory tags based on the command type, the number of available memory tags from the first set of memory tags, and the number of available memory tags from the second set of memory tags.

Dynamic load based memory tag management

Growing community of inventors

Binghamton, NY, United States of America

Glenn David Gilda