Siemens Corporation

Siemens Corporate Research, Inc.

Ncr Corporation

At+t Corp.

Siemens Medical Solutions Usa, Inc.

Siemens Enterprise Communications Gmbh & Co. Kg

Andre Benjamin Bondi

Alberto Avritzer

Christoph Kuhmuench

Moshe Segal

Michael W Hasling

A method for automatic identification of execution phases in load test data includes receiving load test data indicating processor utilization for threads over a period of time divided into a plurality of intervals. For each pair of proximate intervals, it is determined whether thread-wise processor utilization for a first interval is statistically indistinguishable from that of a second interval. The pair of proximate intervals is combined when it is determined that the processor utilization for the first interval is statistically indistinguishable from that of the second interval. Each of the pair of proximate intervals is divided into subintervals when it is determined that the first interval is not statistically indistinguishable the second interval. One or more execution phases are automatically identified as occurring between proximate intervals that are not substantially equivalent.

Automatic identification of execution phases in load tests

A method for automatic identification of execution phases in load test data includes receiving load test data. A period of time of the load test data is divided into a plurality of intervals. For each pair of proximate intervals of the plurality of intervals, a first interval is compared with a second interval to determine whether application execution characteristics for each of the pair of proximate intervals are substantially equivalent. The pair of proximate intervals is combined into a single interval when it is determined that the pair is substantially equivalent. Each of the pair of proximate intervals is divided into subintervals when it is determined that the pair of proximate intervals is not substantially equivalent. One or more execution phases are automatically identified as occurring between proximate intervals that are not substantially equivalent.

A computer-implemented method for triggering a software rejuvenation system and/or method includes receiving a request for resources, determining an estimated response time to the request for resources, determining that the estimated response time is greater than a first threshold, determining that a number of estimated response times greater than the first threshold is greater than or equal to a dynamic second threshold, and triggering the software rejuvenation system and/or method.

Dynamic tuning of a software rejuvenation method using a customer affecting performance metric

A method for analyzing load run test results of a computer system includes providing a plurality of performance measure sets derived from a first series of load run tests performed over a same time period. Each set comprises a plurality of records and each record has a timestamp and a value. Performance measure records are sorted by value in ascending order, for each performance measure set. Plateau regions are identified in the sorted performance measure sets, for each performance measure. Performance measure records are sorted within each plateau region by time stamp in ascending order, for each performance measure. Overlapping time intervals covered by the plateau regions are identified for the plurality of performance measure sets, and averages of performance measures are computed over the overlapping time intervals.

System and method for automating the analysis of computer system load testing

A system for servicing a telecommunications conversation comprising one or more messages includes a sequence recognition engine for processing an incoming message, a queue for storing a sequence of messages of said conversation, a length indicator for a message subsequence eligible for processing, a state indicator of said message sequence, and an indicator of a most recently transmitted message of said message sequence. When a new message arrives the sequence recognition engine is invoked to place the message in a correct position in the queue and to determine a next state for the message sequence.

System and method for recognition and cyclic service of partial sequences of communication requests

A computer-implemented method for triggering a software rejuvenation system or method in a computer-system including a plurality of replicates includes receiving a request for resources at a first replicate, determining an estimated response time to the request for resources, and determining that the estimated response time is greater than a first threshold. The method further includes determining that a number of estimated response times greater than the first threshold is greater than or equal to a second threshold, wherein the second threshold of the first replicate is different than a second threshold of a second replicate of the computer-system, and triggering the software rejuvenation system or method for the first replicate.

Inducing diversity in replicated systems with software rejuvenation

A computer-implemented method for triggering a software rejuvenation system and/or method includes receiving a plurality of requests for resources, determining to take a sample of the requests with a probability, determining an estimated average response time to the sample of the requests for resources, determining that the estimated average response time is greater than a first threshold, wherein the first threshold is a mean of an assumed value, determining that a number of estimated average response times greater than the first threshold is greater than or equal to a second threshold, and triggering the software rejuvenation system and/or method.

System and method for triggering software rejuvenation using a customer affecting performance metric

An apparatus and method for routing calls to communication processing centers based on an optimal destination plan. Calls are allocated on the basis of area codes, regions, or other predetermined categories to particular communication processing centers. Each communication processing center's free capacity is calculated for all time intervals and this result is used to ascertain the smallest free capacity of any time interval for each communication processing center. The calls are then reallocated through linear programming methods such that the amount of smallest free capacity is maximized in order to determine the optimal call allocation scheme.

Method and apparatus for assigning incoming communications to communications processing centers

A method and system for monitoring nodes in a network having at least one network management station and a plurality of nodes is provided. Initially, polling messages are sent from the network management station to the plurality of nodes. Preferably, the polling messages are sent in a sequence whose emission is regulated at a predetermined rate. As the polling messages are sent, an unacknowledged poll table is updated with the appropriate information indicating that a set of nodes have been sent a polling message. The unacknowledged poll table is preferably indexed by an IP address of each node and is indexed by the time of the next scheduled timeout associated with each node. This method allows an arbitrary number of unacknowledged polls to be outstanding at any instant. Once the polling messages are sent and the queue is updated, the network management station then determines if a node has failed. That is, once the polling messages are sent and the queue is updated, the method verifies whether an acknowledgement is received from each of the managed nodes within a predetermined timeout period. If an acknowledgement is not received, another polling message is sent and the network management station determines if the polling message has been acknowledged after a predetermined timeout period has elapsed. This process is repeated for each node on the outstanding poll list until a predetermined number of polling messages are sent to the same target node. When the network management station sends out the last of the predetermined number of polling messages and has waited for the associated time-out period to expire, then the target node is determined to have failed.

Growing community of inventors

Red Bank, NJ, United States of America

Andre Benjamin Bondi