Abstract: A hierarchical distributed routing architecture including at least three levels, or layers, for receiving, processing and forwarding data packets between network components is provided. The core level router components receive an incoming packet from a network component and identify a distribution level router component based on processing a subset of the destination address associated with the received packet. The distribution level router components receive a forwarded packet and identify a transit level router component based on at least a subset of the destination address associated with the received packet. The transit level router components receive the forwarded packet and forward the packet to a respective network. The mapping of destination addresses to router components of may be managed by a router management component. In some embodiments, mapping of destination address to router components may be based, at least in part, on traffic volumes associated with the mapped destination addresses.

Abstract: A hierarchical distributed routing architecture including at least two levels, or layers, for receiving, processing and forwarding data packets between network components is provided. The core level router components receive an incoming packet from a network component and identify a distribution level router component based on processing a subset of the destination address associated with the received packet. The distribution level router components receive a forwarded packet and forward the packet to a respective network. The mapping, or other assignment, of portions of destination addresses to router components of the distributed routing environment may be managed by a router management component. In some embodiments, mapping of destination address to router components may be based, at least in part, on traffic volumes associated with the mapped destination addresses.

Abstract: A hierarchical distributed routing architecture including at least three levels, or layers, for processing data packets is provided. The core level router components receive an incoming packet and identify a distribution level router component based on processing a subset of the destination address associated with the received packet. The distribution level router components receive a forwarded packet and identify a transit level router component based on at least a subset of the destination address associated with the received packet. The distribution level router components maintain address routing information a first and second memory. The transit level router components receive the forwarded packet and forward the packet to a respective network. The mapping of destination addresses to router components of may be managed by a router management component. Mapping of destination address to router components may be based, at least in part, on traffic volumes associated with the mapped destination addresses.

Abstract: A data center for executing a data processing application includes processing units, sub-units or servers. Each of the processing units, sub-units or servers can execute a part or all of the data processing application. The processing units, sub-units or servers are electrical disjoint with respect to data communications, but can communicate with each other over free space optical links.

Abstract: Input and output (I/O) operations performed by a data storage device are managed dynamically to balance aspects such as throughput and latency. Sequential read and write requests are sent to a data storage device whereby the corresponding operations are performed without time delay due to extra disk revolutions. In order to minimize latency, particularly for read operations, random read and write requests are held in a queue upstream of an I/O controller of the data storage device until the buffer of the data storage device is empty. The queued requests can be reordered when a higher priority request is received, improving the overall latency for specific requests. An I/O scheduler of a data server is still able to use any appropriate algorithm to order I/O requests, such as by prioritizing reads over writes as long as the writes do not back up in the I/O queue beyond a certain threshold.

Abstract: When providing a user with native access to at least a portion of device hardware, the user can be prevented from modifying firmware and other configuration information by controlling the mechanisms used to update that information. In some embodiments, an asymmetric keying approach can be used to encrypt or sign the firmware. In other cases access can be controlled by enabling firmware updates only through a channel or port that is not exposed to the customer, or by mapping only those portions of the hardware that are to be accessible to the user. In other embodiments, the user can be prevented from modifying firmware by only provisioning the user on a machine after an initial mutability period wherein firmware can be modified, such that the user never has access to a device when firmware can be updated. Combinations and variations of the above also can be used.

Abstract: An exemplary component for managing requests for resources in a data center includes a service request module for receiving requests for resources from a plurality of services and a resource module to monitor resources in a data center and to match received requests to resources. Such a component optionally includes an application programming interface (API) that provides for sending information in response to an API call made by a service. Other methods, devices and systems are also disclosed.

Abstract: A power management system is provided. The system includes a permanent storage medium that has a plurality of storage segments that are individually controllable. A power manager analyzes requirements of programs that access the permanent storage medium and selectively enables or disables a subset of the storage segments in order to mitigate power consumption of the storage medium.

Abstract: A data center for executing a data processing application includes processing units, sub-units or servers. Each of the processing units, sub-units or servers can execute a part or all of the data processing application. The processing units, sub-units or servers are electrical disjoint with respect to data communications, but can communicate with each other over free space optical links.

Abstract: A data center for executing a data processing application includes processing units, sub-units or servers. Each of the processing units, sub-units or servers can execute a part or all of the data processing application. The processing units, sub-units or servers are electrical disjoint with respect to data communications, but can communicate with each other over free space optical links.

Abstract: An illustrative power-efficient data center is described for operating in an uncontrolled environment in one scenario. The data center includes an air moving system that applies unconditioned air to resource items. The resource items are stripped down to provide a substantially minimum set of components for performing the data center'core functions. Various illustrative techniques for managing a power-efficient data center are also described.

Abstract: Commitments against various resources can be dynamically adjusted for customers in a shared-resource environment. A customer can provision a data volume with a committed rate of Input/Output Operations Per Second (IOPS) and pay only for that commitment (plus any overage), for example, as well as the amount of storage requested. The customer can subsequently adjust the committed rate of IOPS by submitting an appropriate request, or the rate can be adjusted automatically based on any of a number of criteria. Data volumes for the customer can be migrated, split, or combined in order to provide the adjusted rate. The interaction of the customer with the data volume does not need to change, independent of adjustments in rate or changes in the data volume, other than the rate at which requests are processed.

Abstract: Customers of a shared-resource environment can provision resources in a fine-grained manner that meets specific performance requirements. A customer can provision a data volume with a committed rate of Input/Output Operations Per Second (IOPS) and pay only for that commitment (plus any overage), and the amount of storage requested. The customer will then at any time be able to complete at least the committed rate of IOPS. If the customer generates submissions at a rate that exceeds the committed rate, the resource can still process at the higher rate when the system is not under pressure. Even under pressure, the system will deliver at least the committed rate. Multiple customers can be provisioned on the same resource, and more than one customer can have a committed rate on that resource. Customers without committed or guaranteed rates can utilize the uncommitted portion, or committed portions that are not being used.

Abstract: An exemplary method for managing power consumption of a data center includes monitoring power consumption of a data center, assessing power consumption with respect to a billing equation for power, based on the assessment, deciding whether to implement a power policy where the power policy reduces instantaneous power consumption by the data center and increases a load factor wherein the load factor is an average power consumed by the data center divided by a peak power consumed by the data center over a period of time. Various other methods, devices, systems, etc., are also disclosed.

Abstract: An exemplary data stream includes value information for use by consumers of global computing resources in making requests for global computing resources. An exemplary method includes receiving information about data center resources from one or more data centers, based at least in part on the information estimating value information for consumption of computing resources of the one or more data centers and streaming the value information via a network. An exemplary medium or media includes instructions to instruct a computing device to receive, from a data stream, value information for computing resources of one or more data centers, to format the value information for display and to issue requests for consumption of at least some of the computing resources. Other methods, devices and systems are also disclosed.

Abstract: A data center for executing a data processing application includes processing units, sub-units or servers. Each of the processing units, sub-units or servers can execute a part or all of the data processing application. The processing units, sub-units or servers are electrical disjoint with respect to data communications, but can communicate with each other over free space optical links.

Abstract: A hierarchical distributed routing architecture including at least three levels, or layers, for receiving, processing and forwarding data packets between network components is provided. The core level router components receive an incoming packet from a network component and identify a distribution level router component based on processing a subset of the destination address associated with the received packet. The distribution level router components that receiving a forwarded packet and identify a transit level router component based a second processing of at least a subset of the destination address associated with the received packet. The distribution level router components maintain address routing information a first and second memory. The transit level router components receive the forwarded packet and forward the packet to a respective network. The mapping, or other assignment, of portions of the FIB associated with the distributed routing environment is managed by a router management component.

Abstract: A data center for executing a data processing application includes processing units, sub-units or servers. Each of the processing units, sub-units or servers can execute a part or all of the data processing application. The processing units, sub-units or servers are electrical disjoint with respect to data communications, but can communicate with each other over free space optical links.

Abstract: A data center for executing a data processing application includes processing units, sub-units or servers. Each of the processing units, sub-units or servers can execute a part or all of the data processing application. The processing units, sub-units or servers are electrical disjoint with respect to data communications, but can communicate with each other over free space optical links.

Abstract: A hierarchical distributed routing architecture including at least two levels, or layers, for receiving, processing and forwarding data packets between network components is provided. The core level router components receive an incoming packet from a network component and identify a distribution level router component based on processing a subset of the destination address associated with the received packet. The distribution level router components receive a forwarded packet and forward the packet to a respective network. The mapping, or other assignment, of portions of the FIB associated with the distributed routing environment is managed by a router management component.