International Business Machines Corporation

Bret Wayne Holley

Roger Gregory Hathorn

Juan A Coronado

Matthew J Kalos

Harry M Yudenfriend

Matthew Richard Craig

Clint Alan Hardy

James Lincoln Iskiyan

Dinh Hai Le

Clarisa Valencia

William Frank Micka

Louis William Ricci

Lawrence Carter Blount

Mark Steven Bayus

Daniel J Perkin

John R Flanagan

Daniel Francis Casper

Brent Cameron Beardsley

Gustav E Sittmann, Iii

Scott M Carlson

Steven Edward Klein

Catherine C Huang

Mark P Bendyk

Ashwani Kumar

Timothy Alan Griffin

Tiep Quoc Pham

Marlyn Quinones

Marilyn Quinones

Asim R Qureshi

Matthew Joseph Kalos

Juan Alonso Coronado

A first device determines whether an indicator is configured to allow use of an increased value for an information unit pacing credit that is greater than a default value for the information unit pacing credit, where the information unit pacing credit indicates a number of information units that a second device is allowed to send to the first device without waiting for any additional response from the first device. In response to determining that the indicator is configured to allow use of the increased value for the information unit pacing credit that is greater than the default value for the information unit pacing credit, the first device adjusts the information unit pacing credit via a response sent from the first device to the second device, to the increased value that is greater than the default value.

Adjustment of an information unit pacing credit to increase data transmission rate

A plurality of queues including a target input queue is configured for multi-protocol operation in a driver. The target input queue is configured to receive all incoming new input/output (I/O) initiation messages, regardless of protocol, wherein messages on the target input queue include messages for first burst data that follows a command to preserve order in processing of messages, in response to allocation of identical exchange identifiers to reused messages.

Configuration of a set of queues for multi-protocol operations in a target driver

A control unit provides a number of buffer credits, to one or more channels, in response to an initiation of a startup phase of communication between the one or more channels and the control unit, where the provided number of buffer credits when used for transferring data causes transfer ready operations but no retry operations. The control unit iteratively increases the number of buffer credits by an amount that is high enough to eliminate any transfer ready operations or cause retry operations to occur within a predetermined amount of time from the initiation of the startup phase.

Adjustment of buffer credits and other parameters in a startup phase of communications between a plurality of channels and a control unit

A control unit monitors a number of transfer ready operations and a number of retry operations during a monitoring period during communication with a plurality of channels. Based, on the monitoring, a number of buffer credits for communication with the plurality of channels is adjusted.

Adjustments of buffer credits for optimizing the number of retry operations and transfer ready operations

A channel of a host computational device sends a command to transfer data to a control unit included in a storage controller. The channel of the host computational device receives a number of buffer credits from the control unit for communication with the control unit, where the number of buffer credits that is received is based on the control unit monitoring a number of transfer ready operations and a number of retry operations during a monitoring period while communicating with a plurality of channels that includes the channel.

Receiving buffer credits by a plurality of channels of one or more host computational devices for transmitting data to a control unit

A primary storage controller receives a write command from a host, wherein Fiber Channel frames corresponding to the write command have a priority indicated by the host. The primary storage controller performs a synchronous copy operation to copy data written by the write command from the primary storage controller to a secondary storage controller, wherein Fiber Channel frames corresponding to the synchronous copy operations have an identical priority to the priority indicated by the host.

Maintenance of a fabric priority during synchronous copy operations

A first device that is coupled to a second device receives a first Fiber Channel frame, wherein the first Fiber Channel frame has a priority indicated by the second device. The first device responds to the second device with a second Fiber Channel frame that has an identical priority to the priority indicated by the second device.

Maintenance of a fabric priority among operations within a fibre channel exchange

Provided are a method, a system, and a computer program that use a Fibre Connection (FICON) protocol, in which a first device that is coupled to a second device receives an outbound exchange from the second device, wherein one or more Fibre Channel frames of the outbound exchange have a priority indicated by the second device. The first device responds to the second device with an inbound exchange, wherein one or more Fibre Channel frames of the inbound exchange have an identical priority to the priority indicated by the second device in the outbound exchange. In additional embodiments, priority is maintained across related exchange pairs.

Maintenance of a fabric priority among operations in related fibre channel exchanges

A primary storage controller receives a write command from a host, wherein Fibre Channel frames corresponding to the write command have a priority indicated by the host. The primary storage controller performs a synchronous copy operation to copy data written by the write command from the primary storage controller to a secondary storage controller, wherein Fibre Channel frames corresponding to the synchronous copy operations have an identical priority to the priority indicated by the host.

A method, system and computer-usable medium are disclosed for managing storage control unit input/output (I/0) operations in a multi-host environment. A storage control unit sends a busy status message when it has exhausted its resources to process new I/O requests. Once the utilization of the storage control unit's resources has subsided to a predetermined upper utilization value, it sends an availability status message to a first predetermined number of host systems that are owed an availability status message. After the first predetermined number of availability status messages are sent, the storage control unit then waits for a predetermined period of time before sending a second predetermined number of availability status messages.

Method to prevent I/O starvation of slower hosts in a multi-host environment

A method for clearing a fault condition at a target device is disclosed herein. In selected embodiments, such a method includes detecting a fault condition at a target device and receiving N instructions before the fault condition is cleared, where the N instructions are unexecutable due to the fault condition. N fault condition indicators are transmitted in response to the N instructions. Clearing of the fault condition is detected when the fault condition no longer exists. Acknowledgments corresponding to the fault condition indicators are received, where each acknowledgment indicates that one of the fault condition indicators has been received. A fault clear indicator is transmitted only after both all N fault condition indicators have been received and clearing of the fault condition has been detected. A corresponding system and computer program product are also disclosed herein.

Growing community of inventors

Tucson, AZ, United States of America

Bret Wayne Holley