Hewlett-packard Development Company, L.p.

Magnus Karlsson

Christos Karamanolis

Zhichen Xu

Mallik Mahalingam

Xiaoyun Zhu

Michele Mae Covell

Ludmila Cherkasova

John Wilkes

Ira Cohen

Chunqiang Tang

Chengdu Huang

Qian Wang

According to one embodiment, a method comprises receiving, by an adaptive controller, performance measurement for a computing system. The method further comprises estimating a performance model for use by the adaptive controller, and determining whether the estimated performance model has a correct sign for approaching performance desired for the computing system. When determined that the estimated performance model has an incorrect sign, the adaptive controller takes action to determine a performance model having a correct sign for approaching performance desired for the computing system.

System and method for determining correct sign of response of an adaptive controller

A method for diagnosing system health with system event logs is provided. The method includes receiving a plurality of event logs and health indicator states from a system; transducing the plurality of event logs into numeric-based metrics of the system; and deriving, based on the transduced numeric-based metrics, at least one model of the system that correlates the plurality of event logs to the corresponding health indicator states.

Diagnosis of system health with event logs

An embodiment of a method of controlling access to a computing resource within a shared computing environment begins with a first step of determining a plurality of controller functions for a plurality of operating ranges for workloads accessing the computing resource. Each of the controller functions comprises a mathematical operator which takes an input and provides an output. The method continues by iteratively performing second through fifth steps. In the second step, the method measures performance parameters for the workloads to determine a performance parameter vector for the workloads. In the third step, the method compares the performance parameter vector to a reference performance parameter vector to determine an error parameter. In the fourth step, the method applies a particular controller function selected from the plurality of controller functions to the error parameter to determine a target throughput for each of the workloads. In the fourth step, the method adjusts access to the computing resource for each of the workloads having a throughput limit different from about the target throughput for the workload by reducing or increasing the throughput limit for the workload to about the target throughput.

Method of controlling access to computing resource within shared computing environment

According to one embodiment, a method comprises receiving at a scheduler a change to a weight assigned to a consumer. The method further comprises utilizing, by the scheduler, a weighted proportional-share scheduling algorithm to maintain fairness in allocating shares of a resource to competing consumers in accordance with the changed weight. According to another embodiment, a system comprises at least one resource, and a plurality of competing consumers desiring to use the resource(s). A scheduler allocates shares of the resource(s) to the competing consumers according to a weighted proportional-share algorithm. A controller monitors at least one of performance of the competing consumers and utilization of the resource(s), and controls the performance and/or utilization by dynamically changing a scheduler parameter. The scheduler maintains fairness in allocating shares of the resource(s) to the competing consumers in accordance with the dynamically changed scheduler parameter.

Weighted proportional-share scheduler that maintains fairness in allocating shares of a resource to competing consumers when weights assigned to the consumers change

An embodiment of a method of controlling access to a computing resource within a shared computing environment begins with a first step of measuring performance parameters for workloads accessing the computing resource to determine a performance parameter vector for the workloads. The method continues with a second step of estimating a controller function for the computing resource by analysis of recent performance parameters and recent throughputs. The controller function comprises a mathematical operation which takes an input and provides an output. In a third step, the method compares the performance parameter vector to a reference performance parameter vector to determine an error parameter. In a fourth step, the method applies the controller function to the error parameter to determine a target throughput for each of the workloads. The method concludes with a fifth step of adjusting access to the computing resource for each work load having a throughput limit different from about the target throughput by reducing or increasing the throughput limit for the workload to about the target throughput.

A data model represents semantic information associated with objects stored in a file system. The data model includes a first object identifier, a second object identifier and a relation identifier. The first object identifier identifies a first object stored in the file system. The second object identifier identifies a second object stored in the file system, wherein the second object is related to the first object. The relation identifier identifies a relationship between the first object and the second object.

Semantic file system

Described herein is a method for creating route summaries in an auxiliary network for a peer-to-peer (P2P) overlay network for routing a data packet. The method comprises dividing a d-dimensional Cartesian space of nodes in the P2P overlay network into multiple virtual grids that form the auxiliary network, numbering each virtual grid, summarizing default overlay network zones of the P2P overlay network corresponding to each of the virtual grids that form the auxiliary network, receiving a data packet, and routing the data packet to a packet destination using one of the P2P overlay network and the auxiliary network.

Summarizing nodes in route propagation in auxiliary network for P2P overlay networks

A system and method for an auxiliary network for a peer-to-peer overlay network using autonomous system level topology. Using information available through the auxiliary network, expressway connections are established amongst expressway nodes and ordinary connections are established between ordinary and expressway nodes. The connections established is unconstrained and arbitrary. After the connections are established, efficient routing of information may take place.

Autonomous system topology based auxiliary network for peer-to-peer overlay network

According to one embodiment, a method comprises scheduling, by a weighted proportional-share scheduler, access of competing flows to a computing service in accordance with respective weights assigned to the competing flows. The method further comprises autonomously determining, by a controller, based at least in part on received performance measurement for the flows, values of the respective weights assigned to the competing flows for achieving respective performance goals desired for the competing flows. When determined by the controller that the values of the respective weights assigned to the competing flows should be changed for achieving the performance goals, the controller communicates the changed values of the respective weights to the weighted proportional-share scheduler.

System and method for dynamically controlling weights assigned to consumers competing for a shared resource

An embodiment of a method of selecting a heuristic class for data placement in a distributed storage system begins by forming a general integer program which models the data placement and forming a specific integer program which models a heuristic class for the data placement. The general and specific integer programs each comprising an objective of minimizing a replication cost. The method continues with solving the general integer program which provides a general lower bound for the replication cost and solving the specific integer program which provides a specific lower bound for the replication cost. The method concludes with selecting the heuristic class if a difference between the general lower bound and the specific lower bound is within an allowable amount.

Method of selecting heuristic class for data placement

A system and method for an auxiliary network for a peer-to-peer overlay network using landmark numbering. Using information available through the auxiliary network, expressway connections are established amongst expressway nodes and ordinary connections are established between ordinary and expressway nodes. The connections established is unconstrained and arbitrary. After the connections are established, efficient routing of information may take place.

Landmark numbering based auxiliary network for peer-to-peer overlay network

A method and system for workload-aware request in cluster-based network servers. The present invention provides a web server cluster having a plurality of nodes wherein each node comprises a distributor component, a dispatcher component and a server component. In another embodiment, the present provides a method for managing request distribution to a set of files stored on a web server cluster. A request for a file is received at a first node of a plurality of nodes, each node comprising a distributor component, a dispatcher component and a server component. If the request is for a core file, the request is processed at the first node (e.g., processed locally). If the request is for a partitioned file, it is determined whether the request is assigned to be processed locally at the first node or at another node (e.g., processed remotely). If the request is for neither a core file nor a partitioned file, the request is processed at the first node. In one embodiment, the present invention provides a method for identifying a set of frequently accessed files on a server cluster comprising a number of nodes. Embodiments of the present invention operate to maximize the number of requests served from the total cluster memory of a web server cluster and to minimize the forwarding overhead and disk access overhead by identifying the subset of core files to be processed at any node and by identifying the subset of partitioned files to be processed by different nodes in the cluster.

System and method for workload-aware request distribution in cluster-based network servers

A technique for data placement in a distributed system that takes into account fault tolerance. Data placement is performed in which data objects, and possibly replicas thereof, are assigned to nodes within the distributed system. The resulting placement is then tested to determine whether the system provides desired performance under various different fault scenarios. If not, the distributed system is altered such as by altering its capacity or its capacity allocations. Performing data placement, testing for fault-tolerance and altering capacity or capacity allocations are performed repetitively, thereby increasing the system's ability to provide the desired performance under the fault scenarios. Preferably, a system and placement are eventually determined that provide the desired performance under the given fault scenarios.

Data placement for fault tolerance

Delayed caching receives an evaluation interval to delay updating the objects stored in a delayed cache, delays a time period corresponding to the evaluation interval, and updates the objects stored in the contents of the delayed cache when the time period delay has completed. The configuration operation for the delayed cache selects a time interval to sample a trace having object access frequencies for objects stored in a cache, creates a first working set of objects accessed during the time interval and a second working set of objects accessed during a subsequent time interval based on the historical trace, determines that the difference between the objects contained in the first and second working sets does not exceed a maximum threshold with the selected time interval, and sets a evaluation interval for evaluating the contents of the cache to the selected time interval.

Growing community of inventors

Mountain View, CA, United States of America

Magnus Karlsson