Abstract: A failure prediction subsystem may obtain metrics information from a set of servers, the metrics information including measurements of the operation of the set of servers. The failure prediction subsystem may then determine a mean time between failures for at least one server of the set of servers by at least providing a portion of the metrics information as an input to a machine learning algorithm. The machine learning algorithm may output a mean time between failure for at least one server of the set of servers and the failure prediction subsystem may determine if mitigating action should be taken based at least in part on the mean time between failure.

Abstract: Systems and methods are presented for enabling a user to provide rules for the placement of computing resources at a data center for an entity that employs or is associated with the user. The data center can use the placement rules to select a data center computer system to host computing resources for a user. The rules can be used to establish diversity in computing resource placement at the data center thereby reducing the number of users who lose access to computing resources when a specific data center computer suffers a failure. Further, the placement rules can be used to facilitate configuration of the computer resources for the user based, for example, on the user's employment responsibilities.

Abstract: A data set may be partitioned according to relative differences indicated by a cover tree. A cover tree may be generated for a data set. Items in the data set may be stored at the same or different nodes according to the relative difference between the items indicated in the cover tree. Portions of the cover tree may be assigned to different nodes storing the data set. Access requests for the data set may be performed by sending the access requests to nodes identified according to the assigned portions of the cover tree.

Abstract: A distributed system may implement prioritizing resource hosts for resource placement. A placement request for a resource to be placed in a distributed system at a resource host may be received. Available resource hosts may be filtered out according to placement constraints. The resource hosts may be assigned to different priority sets according to a priority scheme for placing resources at resource hosts. Evaluations of resource hosts from the different priority sets may be performed to select candidate resource hosts to provide as possible placement locations.

Abstract: A computer implemented system and method is provided for managing sources of data transfer from a content delivery network (CDN) service provider to a client device. Illustratively, the data transfer is initialized between a first endpoint of the CDN and the client device. While the data transfer is ongoing between the first endpoint and the client device, the system receives performance information regarding alternative endpoints which may replace the first endpoint as the source for the ongoing data transfer. The performance information may be generated by the client device that probes alternative endpoints, or may be generated by dedicated monitoring nodes deployed on the network. Should a change of data transfer source be desirable, the system may cause the first endpoint to stop data transmission and cause a selected alternative endpoint to continue the ongoing data transfer to the client in accordance with a current progress of the data transfer.

Abstract: Systems and methods are provided for increasing the overall network performance experienced by a group of devices by forming a dynamic and collaborative cluster of computing devices. In particular, the computing devices within the cluster collectively may identify and leverage the current capabilities of each of the individual members of the cluster to respond efficiently to network resource requests from computing devices inside or outside the cluster. As such, various embodiments provide for a dynamic cluster of computing devices that tailor the responsibilities of the members of the cluster to the current capabilities, capacities, and state of these computing devices. In particular, devices in the cluster may participate dynamically in the cluster to ensure that a device in the cluster that is currently most suited to performing a task is the device selected to perform that task.

Abstract: A power cable includes electrical connectors at a first end and a second end of the power cable. The electrical connectors individually or collectively include a built-in reserve power source, such as batteries or supercapacitors, and a control circuit configured to provide power to one or more downstream devices from the reserve power source when power received at one or more of the electrical connectors of the power cable is interrupted. In some embodiments, an electrical connector includes one or more indicators configured to indicate a state of the electrical connector and/or display one or more indications in response to receiving a signal from an external source, such as a management system of a facility.

Abstract: A computer system associated with a certificate authority receives a request to obtain information that can be used to determine a validity status of a digital certificate. In response to the request, the computer system provides the information and updates usage information for the digital certificate to incorporate information obtained from the request. The usage information may be generated based at least in part on previous requests to obtain the information. Based at least in part on the usage information, the computer system will perform at least one operation associated with the digital certificate.

Abstract: A deployment service at a remote provider network receives topology data for a local network and generates data filters for edge devices of the local network based on the topology data. The deployment service then sends the data filters to a hub device connected to the local network. The hub device deploys the data filters to respective edge devices of the local network. The data filters may be configured to discard a sufficient portion of collected data to prevent routers from being overloaded by network traffic. The data filters may also be configured to discard a sufficient portion of collected data to prevent the edge devices from consuming too much power in order to preserve energy cost or battery life.

Abstract: A roller assembly for a taping device includes a roller body defining a central bore, a bushing configured to be received within the central bore and rotationally fixed to the roller body, and a shaft configured to be disposed within the central bore. The shaft defines an outer surface that extends through the bushing. The shaft has a boss that extends radially outward from the outer surface. A resistance element is disposable within the central bore is configured to contact the bushing and the boss in a manner providing an amount of rotational resistance between the roller body and the shaft when the roller assembly is in an assembled configuration.

Abstract: An interrupt request generating process can include determining a first interrupt triggering event has occurred after a predetermined period of time in which no interrupt triggering event occurred. In response to determining that the first interrupt triggering event has occurred, the interrupt request generating process may generate a first interrupt request without adding an intentional delay, and initiating a timer configured to expire after a predetermined time interval. When a second interrupt triggering event is determined to have occurred before the timer expires, a second interrupt request is delayed from being generated until the timer expires.

Abstract: A messaging service provides a search mechanism that utilizes a protected index. The protected index is generated by converting documents maintained by the messaging service into a set of tokens or words. Each token is converted to a corresponding value using a transformation such as a cryptographic hash function. The values are placed into an index that allows the messaging service to efficiently identify a set of documents associated with each particular value. When a document search request is submitted to the messaging service, the messaging service uses the transformation to generate corresponding values for each term in the search request, and uses the index to identify sets of documents associated with the values corresponding to the search terms. The messaging service applies search logic associated with the search request to the identified sets of documents to produce a final set of documents satisfying search request.

Abstract: A metrics subsystem determines the utilization of computing resources of server computer systems. The metrics subsystem may obtain metadata associated with a plurality of computing resources used to provide a service. The metrics subsystem determines a set of references to data included in the metadata and aggregates references of the set of references to determine utilization of computing resources and provides the utilization to a metering service.

Abstract: Techniques for providing of content using pre-loaded content attribute information are described herein. In some cases, a plurality of compute nodes may be grouped into multiple different compute node groups. Each of the multiple different compute node groups may include one or more compute nodes loaded with information associated with one or more assigned content attribute values. A request for content may be received from a client. The request may be associated with one or more requested content attribute values. The request may then be processed by selecting, based, at least in part, on the requested content attribute values, a selected compute node group to handle the request.

Abstract: In a computing environment, a request for computing resources is received by an entity. A quantity of resource slots sufficient to fulfill the request is determined and associated with the entity during a portion of a recurring time period. The resource slots can be associated with other entities during other portions of the recurring time period.

Abstract: Techniques are described for optimizing streaming video encoding profiles.

Abstract: Systems, methods, apparatus and computer readable medium is disclosed for extending the Virtual Routing and Forwarding (VRF) in the network using embedded source identifiers in the packets. An example apparatus may include ports, memory and processing logic. The memory may include VRF routing tables corresponding to respective VRF domains and a list of media access control (MAC) addresses, wherein each MAC address in the list is linked to one of the respective VRF routing tables. The processing logic may be configured to receive an ingress packet at a port, identify a matching MAC address in the list that matches a source identifier of the ingress packet, and process the ingress packet according to the respective VRF domain associated with the matching MAC address that matches the source identifier.

Abstract: An unmanned aerial vehicle (UAV) landing marker that absorbs incoming radar signals emitted by a UAV and/or disperses reflected radar signals. The absorption and/or dispersion of the radar signals creates a reduced radar return in comparison to the environment about the landing marker. The UAV can detect the area of reduced radar return and determine that it is a landing marker. Additionally, the UAV can determine a position of the landing marker relative to the UAV, based on the reduced radar return, to effect delivery of an item by the UAV. The landing marker can include materials, structures and/or features to absorb and/or disperse radar signals to cause the reduced radar return.