摘要:
Provided are techniques for determining a link speed. When a link between two computing devices is operational, a link speed for use in communicating across the link is stored and a remembered indicator is set to TRUE. After any event occurs that causes the link to become inoperational, in response to determining that the remembered indicator is TRUE, the stored link speed is used when attempting to make the link become operational.
摘要:
Provided are a computer program product, system, and method for supporting non-delivery notification between a switch and device in a network. A login with a switch is initiated including a test service parameter requesting a non-delivery notification service to cause the switch to return a non-delivery notification in response to a test path request indicating a destination node that is unavailable. Indication is made that the test service is available for the login in response to receiving a reply from the switch that the switch accepts the login with the non-delivery notification service. The test path request is sent to the switch to determine whether the destination node is available in response to determining that the non-delivery notification service is available for the switch login. The non-delivery notification is received from the switch in response to the switch determining that the destination node to which the test path request is directed is unavailable.
摘要:
A method for combining partial records into a single direct memory access (DMA) operation for a count key data (CKD) protocol in a computer environment is provided. In an initiator processor of the computer environment, a number of the partial records to be prefetched is determined by gathering a plurality of descriptor information for a command according to a predetermined algorithm having a plurality of assumptions for the command. The number of partial records is prefetched. At least one of record headers and record keys of the number of partial records are concatenated into the single DMA operation. The DMA operation is forwarded to a receiver process to be completed.
摘要:
Provided are a computer program product, system, and method for supporting non-delivery notification between a switch and device in a network. A login with a switch is initiated including a test service parameter requesting a non-delivery notification service to cause the switch to return a non-delivery notification in response to a test path request indicating a destination node that is unavailable. Indication is made that the test service is available for the login in response to receiving a reply from the switch that the switch accepts the login with the non-delivery notification service. The test path request is sent to the switch to determine whether the destination node is available in response to determining that the non-delivery notification service is available for the switch login. The non-delivery notification is received from the switch in response to the switch determining that the destination node to which the test path request is directed is unavailable.
摘要:
An apparatus, system, and method are disclosed for updating a code image for a communication adapter. The apparatus includes a save module, an update module, and a restore module. The save module saves configuration data associated with a connection at a communication adapter. The update module updates a code image on the communication adapter. The restore module restores the configuration data in response to the code image update. Advantageously, the apparatus eliminates the need for conventional host recovery, including relogin and reestablishment of paths and connections, after the code image update.
摘要:
Provided are a method, system, and article of manufacture, wherein a primary storage control unit receives an information unit from a remote host over a fibre channel connection, wherein persistent information unit pacing is implemented over the fibre channel connection. Information is maintained on how many large writes have been received at the primary storage control unit over at least one logical path established over the fibre channel connection between the primary storage control unit and the remote host, wherein a large write is an input/output (I/O) operation for which a number of data information units that are processed exceeds a default value of an information unit pacing credit. The primary storage control unit adjusts an information unit pacing parameter included in a response sent from the primary storage control unit to the remote host, wherein the adjusting is based at least on the information maintained on how many large writes have been received at the primary storage control unit over the at least one logical path.
摘要翻译:提供了一种方法,系统和制品,其中主存储控制单元通过光纤通道连接从远程主机接收信息单元,其中持续信息单元起搏在光纤通道连接上实现。 维持在通过在主存储控制单元和远程主机之间的光纤通道连接上建立的至少一个逻辑路径在主存储控制单元处接收到多少大写入的信息,其中大写是输入/输出( I / O)操作,其中处理的多个数据信息单元超过信息单元起搏信用的默认值。 主存储控制单元将从主存储控制单元发送的响应中包括的信息单元起搏参数调整到远程主机,其中调整至少基于在主存储器上已经接收到多少大写入的信息 所述至少一个逻辑路径上的控制单元。
摘要:
An apparatus, system, and method are provided for facilitating port testing of a multi-port host adapter. The present invention includes a scheduler that schedules execution of a plurality of threads to test a first port and a plurality of threads to test a second port of a multi-port adapter. The port test routine is divided into threads such that execution time and switching overhead is minimized. A multithreading module provides multithreaded execution of the plurality of threads such that the port test of the first port and the port test of the second port are performed in parallel. The apparatus further includes a communication module that takes the first port and the second port off-line. A third port remains on-line for Input/Output (I/O) communications that are multithreaded with the plurality of threads involving the first port and the second port.
摘要:
A Transfer Progress Alert Module and a method for optimizing processing of a data transfer load, in a data communication system is provided. The data transfer load is divided in individual data blocks. The device and method simultaneously perform pipelined operations on different individual data blocks, thus optimizing the overlap of pipelined operations. The method includes initializing the transfer by selecting a pre-defined individual data block size and determining a key for selecting and monitoring transfers with transfer addresses within a pre-determined address region. The method then continuously repeats following steps until all monitored individual data blocks from the data transfer load are processed. First, the incoming individual data blocks are transferred on a bus between a peripheral device and a memory, and the Transfer Progress Alert module is used for monitoring the individual data blocks having transfer addresses determined to belong in the pre-determined address region. The TPA module is used to determine when each monitored transferred individual data block is ready for a post-processing operation, at least one post-processing operation is performed on the data, and the processed data is transferred to a peripheral device. The method and device may also be adapted for performing an error detection operation on the monitored individual data blocks transferred on the bus, for ascertaining the integrity of the transferred data.
摘要:
Provided are a method, system, and article of manufacture, wherein a control unit receives a request to establish a relationship over a fiber channel connection, wherein a first indicator associated with the request indicates that the relationship supports persistent information unit pacing across a plurality of command chains. The control unit sends a response indicating an acceptance of the relationship, wherein a second indicator associated with the response indicates that the control unit supports persistent information unit pacing across the plurality of command chains. An information unit pacing parameter value is retained across the plurality of command chains, in response to determining that the second indicator indicates that the control unit supports persistent information unit pacing across the plurality of command chains.
摘要:
Provided are a method, system, and article of manufacture, in which a logical path is established between a control unit and a channel over a fiber channel connection. Code for persistent information unit pacing is loaded into the control unit and the channel. An indicator is set in node descriptors of the control unit and the channel to indicate concurrent enablement of persistent pacing while retaining the established logical path between the control unit and the channel.