摘要:
Methods and systems for vehicle-to-vehicle communications where vehicles can communicate one or more of speed, position, direction, acceleration and state in order to coordinate or control their operation are disclosed herein. In some embodiments of the disclosed subject matter, methods for vehicle-to-vehicle communications where vehicles can communicate one or more of speed, position and state in order to coordinate or control their operation. An exemplary method includes creating moving broadcasting groups which groups can move with the flow of traffic, adjust the span of the group and maintain minimum overlap size between adjacent broadcasting groups, combining two or more vehicles within one or more broadcasting groups, within some specified distance into neighborhoods, transmitting information about one or more of speed, position and state of vehicles in the neighborhood to the other vehicles in said neighborhood; receiving information about speed, position and state by vehicles in the neighborhood from the other vehicles in said neighborhood, filtering out duplicate information and information received from vehicles that are not in the same neighborhood, providing that two or more vehicles in the same neighborhood have substantially continuous communication capacity by maintaining the size of overlap between adjacent neighborhoods, avoiding delays in entering the new neighborhood by extending the overlap size, and utilizing one or more guarantees provided by an underlying reliable broadcast communication layer by transferring said guarantees to corresponding one or more neighborhoods.
摘要:
A method and system of managing resources in a host bridge by determining whether resources are deficient, preventing a second device from obtaining further resources and if this measure does not enable a first device to obtain resources, guaranteeing all resources to the first device.
摘要:
A method, system, and device for energy efficient job scheduling in a datacenter computing environment includes a master node. The master node can periodically receive energy data from slave nodes and dynamically assign computing tasks to be executed by the slave nodes based on the energy data.
摘要:
An embodiment may include circuitry to facilitate, at least in part, a first network interface controller (NIC) in a client to be capable of accessing, via a second NIC in a server that is remote from the client and in a manner that is independent of an operating system environment in the server, at least one command interface of another controller of the server. The command interface may include at least one controller command queue. Such accessing may include writing at least one queue element to the at least one command queue to command the another controller to perform at least one operation associated with the another controller. The another controller may perform the at least one operation in response, at least in part, to the at least one queue element. Many alternatives, variations, and modifications are possible.
摘要:
A method, system, and device for energy efficient job scheduling in a datacenter computing environment includes a master node. The master node can periodically receive energy data from slave nodes and dynamically assign computing tasks to be executed by the slave nodes based on the energy data.
摘要:
In one embodiment of the invention, a transaction queue stores a transaction request and issues a stream transaction for the stored transaction request when a slot in a set of active stream contexts becomes available. A token generator provides a token representative of a pre-fetched request. A queue selector transfers the transaction request to the transaction queue based on a selection condition.
摘要:
A computer network is provided for handling ordered transactions between a chipset and a memory controller. The chipset provides an interface with a first bus segment and a second bus segment. The chipset may include logic to attach a destination code to ordered transactions transmitted from the chipset. The memory controller may also include logic to parse the destination code from ordered transactions and apply a fence with respect to a first queue and a second queue of the memory controller.
摘要:
An apparatus and method for predicting quantities of data required by requesting devices capable of requesting unspecified quantities of data from storage device, in which prediction of quantities that will be required are made based on past patterns of quantities of data required in past transfers.
摘要:
An embodiment may include circuitry to facilitate, at least in part, a first network interface controller (NIC) in a client to be capable of accessing, via a second NIC in a server that is remote from the client and in a manner that is independent of an operating system environment in the server, at least one command interface of another controller of the server. The command interface may include at least one controller command queue. Such accessing may include writing at least one queue element to the at least one command queue to command the another controller to perform at least one operation associated with the another controller. The another controller may perform the at least one operation in response, at least in part, to the at least one queue element. Many alternatives, variations, and modifications are possible.
摘要:
An apparatus and method for predicting quantities of data that will be requested by requesting devices capable of initiating requests for data from storage devices, in which prediction of quantities that will be requested are made based on past patterns of quantities of data requested.