摘要:
Embodiments of the present invention provide a method, system and computer program product for collaborative route reservation in an HPC fabric. A method for collaborative route reservation in an HPC fabric includes selecting a target node in a cluster of nodes to receive a payload from a source node of the cluster over an HPC fabric and computing a route over the HPC fabric for transferring the payload from the source node to the target node, and also a duration of time requisite to transferring the payload. The method also includes notifying other nodes in the cluster of a reservation of the computed route for the duration of time and utilizing the computed route during the duration of time to transfer the payload. Finally, the method includes responding to completing transfer of the payload by notifying the other nodes that the computed path is no longer reserved.
摘要:
Systems and methods of presenting diagnostic information of an electronic device at diagnostic equipment are disclosed. According to an aspect, a system includes an input/output interface configured to communicatively connect to diagnostic equipment and configured to receive electrical power from the diagnostic equipment. Further, the system includes a diagnostic module that stores diagnostic information of an electronic device. The diagnostic module determines that the diagnostic equipment is connected to the input/output interface. The diagnostic module is configured to determine that electrical power is not available from a power supply of the electronic device. Further, the diagnostic module is configured to receive the electrical power from the diagnostic equipment and use the electrical power from the diagnostic equipment to communicate the stored diagnostic information to the diagnostic equipment in response to the diagnostic equipment being connected to the input/output interface and the electrical power not being available from the power supply.
摘要:
A method includes a client device forming a link-layer protocol packet having a field that includes first credentials of the client device and sending the link-layer protocol packet to a directly attached network device while the client device is in a first configuration. The method further includes a management system obtaining the first credentials from the network device and using the first credentials to access the client device and change the client device from the first configuration to a second configuration. Optionally, a computer system may include a plurality of client devices that implement the method to facilitate securely configuring the entire computer system. Preferably, each client device generates system-unique first credentials.
摘要:
Embodiments of the present invention provide a method, system and computer program product for collaborative route reservation in an HPC fabric. A method for collaborative route reservation in an HPC fabric includes selecting a target node in a cluster of nodes to receive a payload from a source node of the cluster over an HPC fabric and computing a route over the HPC fabric for transferring the payload from the source node to the target node, and also a duration of time requisite to transferring the payload. The method also includes notifying other nodes in the cluster of a reservation of the computed route for the duration of time and utilizing the computed route during the duration of time to transfer the payload. Finally, the method includes responding to completing transfer of the payload by notifying the other nodes that the computed path is no longer reserved.
摘要:
A computer program product includes program instructions configured to be executable by a processor of a networking hardware device to cause the processor to perform various operations. The operations include monitoring network communications on a first port of the networking hardware device to identify a first network address of a first device attached to the first local port and determining that the first network address is incompatible with communication on an external network. The operations further include automatically configuring, in response to determining that the first network address is incompatible with communication on an external network, independent network address translation for the first local port to translate the incompatible first network address to a compatible network address for all egress traffic from the first device and to translate the compatible network address to the incompatible first network address for all ingress traffic to the first device.
摘要:
Loading an operating system of a diskless compute node using a single Virtual Protocol Interconnect (‘VPI’) adapter, including: setting, by a VPI firmware module during startup of the compute node, an operational mode of the VPI adapter to operate in accordance with a first data communications protocol, the VPI adapter including a preboot execution environment module that supports the first data communications protocol prior to loading an operating system and a driver for the first data communications protocol; retrieving, by the VPI firmware module from a network source via the VPI adapter in accordance with the first data communications protocol, the operating system and a driver for the second data communications protocol; and responsive to loading the operating system and the driver for the second data communications protocol, switching the operational mode of the VPI adapter to operate in accordance with the second data communications protocol.
摘要:
Systems and methods for a resource advisor for automated bare-metal operating system installation is disclosed herein. According to an aspect, a method comprising using at least one processor and memory for detecting available resources in a computing device. Further, the method comprises validating resource criteria specified by a user. Further, the method, in response to validating the resource criteria, generating a list of recommended resources for installing an operating system based on the detected available resources.
摘要:
A system management node may perform various operations including receiving a particular event code from a monitored device, wherein the particular event code represents an event that occurred on the monitored device. The operations may further include sending a knowledge base query to a server hosting a knowledge base, wherein the knowledge base query includes the particular event code, and sending a value of a device parameter specific to the monitored device to the server hosting the knowledge base, wherein the value of the device parameter specific to the monitored device enables the knowledge base to refine a determination of a service action to be included in a response. Still further, the operations may include receiving the response from the server hosting the knowledge base, wherein the response identifies a service action to implement on the monitored device to remediate the event that occurred on the monitored device.
摘要:
A method includes a client device forming a link-layer protocol packet having a field that includes first credentials of the client device and sending the link-layer protocol packet to a directly attached network device while the client device is in a first configuration. The method further includes a management system obtaining the first credentials from the network device and using the first credentials to access the client device and change the client device from the first configuration to a second configuration. Optionally, a computer system may include a plurality of client devices that implement the method to facilitate securely configuring the entire computer system. Preferably, each client device generates system-unique first credentials.
摘要:
Loading an operating system of a diskless compute node using a single Virtual Protocol Interconnect (VPI) adapter, including: setting, by a VPI firmware module during startup of the compute node, an operational mode of the VPI adapter to operate in accordance with a first data communications protocol, the VPI adapter including a preboot execution environment module that supports the first data communications protocol prior to loading an operating system and a driver for the first data communications protocol; retrieving, by the VPI firmware module from a network source via the VPI adapter in accordance with the first data communications protocol, the operating system and a driver for the second data communications protocol; and responsive to loading the operating system and the driver for the second data communications protocol, switching the operational mode of the VPI adapter to operate in accordance with the second data communications protocol.