Abstract:
Technologies are generally described for systems, devices and methods effective to provision computing resources based on deduplication. In an example, a resource cost may be produced. The resource cost may relate to execution of a first virtual machine at a data center. A processor may receive a first representation related to an image of the first virtual machine. The processor may receive a second representation related to one or more instances of second virtual machines being executed by the data center. The processor may compare the first and second representations to produce a deduplication value for the first virtual machine. The processor may transform the deduplication value to produce the resource cost.
Abstract:
Technologies are generally described for monitoring application health in datacenters through load rebalancing. In some examples, a domain name server (DNS) module may cyclically vary load provided to an application instance. A monitoring module may then measure application output data and/or parameters associated with the processing of the load by the application based on a difference or a comparison between parts of the cyclical load variation. The monitoring module may use the measured data/parameters to both determine the health of the application instance and to cause the DNS module to adjust the periodic variation of the load.
Abstract:
Technologies are generally described for controlling a flight path of a UAV based image capture system for solid modeling. Upon determining an initial movement path based on an estimate of a structure to be modeled, images of the structure to be modeled may be captured and surface hypotheses formed for unobserved surfaces based on the captured images. A normal vector and a viewing cone may be computed for each hypothesized surface. A set of desired locations may be determined based on the viewing cones for the entire structure to be modeled and a least impact path for the UAV determined based on the desired locations and desired flight parameters.
Abstract:
Techniques described herein generally include methods for the management of hardware accelerator images in a processor chip that includes one or more programmable logic circuits. Hardware accelerator images may be optimized by swapping out which hardware accelerator images are implemented in the one or more programmable logic circuits. The hardware accelerator images may be chosen from a library of accelerator programs downloaded to a device associated with the processor chip. Furthermore, the specific hardware accelerator images that are implemented in the one or more programmable logic circuits at a particular time may be selected based on which combination of accelerator images best enhances performance and power usage of the processor chip.
Abstract:
Technologies are generally described for providing application packages with hardware accelerators. In some examples, an application package for a datacenter may include an application and multiple hardware accelerators associated with the application. Each hardware accelerator may be configured for a different datacenter hardware configuration. When a datacenter receives the application package, it may select the appropriate hardware accelerator for implementation based on its hardware configuration.
Abstract:
Technologies and implementations for providing network coverage by cycling through beam shape coverage configurations are generally disclosed.
Abstract:
Technologies are presented for automatically generating accelerator code for datacenter users, detecting multiple hardware tenants with overlapping accelerator needs, and managing the accelerator needs of the tenants so that they can share use of programmable accelerator hardware. In some examples, the accelerator code may also be customer supplied. In other examples, a delay that simulates accelerator programming for a user first accessing an accelerator that was virtualized from a pre-existing one may be applied. By simulating the delay, co-location detection (a form of side channel attack) may be prevented.
Abstract:
Technology is described for two parties, by leveraging previously established secure connections with third parties, to obtain a shared secret for generating a secure connection with each other in a way that reduces vulnerability to man-in-the-middle attacks. In some examples, the technology can include generating a session identifier; coordinating use of the session identifier by the two parties; finding an available secure communication channel to a third party; transmitting the session identifier to the third party via the available secure communication channel; receiving, via the available secure communication channel, a third party identifier and a session identifier-specific secret; sharing information about the received third party identifier; determining that the received third party identifier matches a third party identifier received by the second party; and using the session identifier-specific secret received with the matching third party identifier to generate a cryptographic key to secure communication between the two parties.