A method is disclosed to identify a port that is associated with a faulty cable. In one embodiment, such a method identifies a cable to replace. The cable provides a path between a first port, residing on a first component, and a second port, residing on a second component. The method further identifies whether an alternative path, bypassing the first cable, exists between the first component and the second component. In the event the alternative path exists, the method sends, over the alternative path, from the first component to the second component, a command to activate an indicator on the second port. This command is received and executed by the second component to activate the indicator. A corresponding apparatus and computer program product are also disclosed.

A power delivery system for data center. Servers are divided into server clusters, and each server cluster is served by multiple power source with multiple inputs from a power supply module. Each power supply module includes a local bus and a local controller which controls connections on the local bus to deliver utility power to its assigned server cluster or backup power to any of the server clusters. Each of the power supply modules has a battery storage system with switchable connections to the local bus and to an inter-system bus, and a PV system with switchable connections to the local bus and to an inter-system bus. The battery storage system may be charged from the utility power or the PV system. The inter-system bus is connected to all of the server clusters, such that power flowing in the inter-system bus can be delivered and dispatched to power any server cluster.

A system is provided and includes a securable resource, a locking element configured to assume a locked condition in which the securable resource is locked and an unlocked condition in which the securable resource is unlocked, a first controller, which is receptive of an instruction to authorize users to unlock the securable resource, and a common interface to which the first controller and additional controllers, which are independent from the first controller and one another, are tied. The first controller is configured to authenticate the users and to perform operating system (OS) level control of the locking element in accordance with the instruction to authorize users and an authentication of the users by the common interface.

A system is provided and includes a securable resource, a locking element configured to assume a locked condition in which the securable resource is locked and an unlocked condition in which the securable resource is unlocked, a controller and a physical authentication interface. The controller is receptive of an instruction to authorize users to unlock the securable resource and configured to perform operating system (OS) level authentication of the users and OS level control of the locking element in accordance with the instruction to authorize users and the OS level authentication. The physical authentication interface is configured to enable or disable a capability of the controller to perform the OS level authentication.

A system includes at least one storage tank configured to store at least one of first compressed air or liquid air. The system also includes a power supply system comprising a turbine, a generator, and a flywheel. The power supply system is configured to receive second compressed air from the at least one storage tank, wherein the second compressed air comprises either the first compressed air or the liquid air which has been heated into a gaseous state; spin the turbine and the flywheel using the second compressed air, wherein the spinning of the turbine generates electrical energy at the generator; provide the electrical energy to a data center for powering electronic devices of the data center; and provide at least a portion of the second compressed air exhausted by the turbine to the data center for cooling the electronic devices of the data center.

Technologies for dynamically managing resources in disaggregated accelerators include an accelerator. The accelerator includes acceleration circuitry with multiple logic portions, each capable of executing a different workload. Additionally, the accelerator includes communication circuitry to receive a workload to be executed by a logic portion of the accelerator and a dynamic resource allocation logic unit to identify a resource utilization threshold associated with one or more shared resources of the accelerator to be used by a logic portion in the execution of the workload, limit, as a function of the resource utilization threshold, the utilization of the one or more shared resources by the logic portion as the logic portion executes the workload, and subsequently adjust the resource utilization threshold as the workload is executed. Other embodiments are also described and claimed.

A system is provided that includes a rack including mount positions to accept electronic equipment modules that are rack mountable and equipped with respective RFID tags, and a RFID reader array including RFID readers each of which is associated with a respective mount position, and has a read field that among the mount positions extends into only the respective mount position. Processing circuitry causes an RFID reader to radiate an interrogation signal within its read field, and in response, receive module identifiable information from an RFID tag of an electronic equipment module mounted at the mount position. A signal including the module identifiable information and mount position identifiable information is transmitted to the processing circuitry, and a second signal to a computer configured to identify the electronic equipment module from the module identifiable information, and locate the electronic equipment module at the mount position from the mount position identifiable information.

A compute device for operation in a rack of a data center includes a substrate, a solid state drive array, a fan array, and a plurality of physical resources. The components of the compute device are arranged on the substrate to prevent or reduce shadowing of heat-producing physical resources, such as a processor, by other heat-producing physical resources relative to an airflow generated by the fan array. To do so, ruler solid state drives are used to facilitate positioning of the fan array toward a front edge of the substrate.

A data center includes: a server including a control plane; a data plane that is configured to receive network connection information from the control plane; and a storage group including a plurality of first storage devices. The data plane may be configured to set connections between the server and the plurality of first storage devices based on the network connection information corresponding to each first storage device of the plurality of first storage devices.

A system to facilitate automatic data center assembly is described. The system includes a first enclosure, including a first set of components, each component including a component manager to perform a component function and a first enclosure manager, communicatively coupled to each component manager, to perform a discovery of each of the first set of components and discover one or more adjacent enclosures coupled to the enclosure via link cables. The system also includes a system manager to automatically discover the first enclosure via the first enclosure manager and discover each of the first set of plurality of components via an associated component manager.