Abstract: A method and system for predicting performance of an application on a machine of a predetermined hardware configuration simulates the performance of the application under a plurality of different simulated hardware configurations; builds a predictive model of the performance of the application based on the results of the simulations; obtains the performance of the application on a plurality of actual machines, each of the machines having a different hardware configuration; and Bayesian reinterprets the predictive model built from the results of the simulations using the performance of the application on the plurality of actual machines, to obtain a final predictive model of the performance of the application having an accuracy greater than the predictive model built from the results of the simulations.

Abstract: This invention provides a new mechanism for “Hot-Tracing” using a novel placeholder mechanism and binary rewriting techniques, which leverages existing compiler flags in order to enable light-weight and highly flexible dynamic instrumentation. Broadly, I-Probe can be divided in 2 distinct workflows—1. Pre-processing (ColdPatch), and 2. Hot Tracing. The first phase is a pre-processing mechanism to prepare the binary for phase 2. The second phase is the actual hot-tracing mechanism, which allows users to dynamically instrument functions (more specifically symbols) of their choice.

Abstract: A system and method are provided for estimating physical resource utilization information for virtual machines. The system includes a model which indicates how virtual resource activities are transformed into physical resource activities for each virtual machine running on a server. A run-time calibrator utilizes the model to estimate physical resource utiliation information for each individual virtual machine on the server.

Abstract: Systems and method for modeling system dynamics, including extracting features representative of a temporal evolution of a dynamical system, further including deriving one or more vector trajectories by performing sliding window segmentation of one or more time series; applying a linear test to determine whether the one or more vector trajectories are linear or nonlinear; and performing linear or nonlinear subspace decomposition on the vector trajectory based on the linear test. The system and method may generate a system evolution model from the extracted features of the dynamical system and determine a fitness score of the system evolution model.

Abstract: A system and method for analysis of complex systems which includes determining model parameters based on time series data, further including profiling a plurality of types of data properties to discover complex data properties and dependencies; classifying the data dependencies into predetermined categories for analysis; and generating a plurality of models based on the discovered properties and dependencies. The system and method may analyze, using a processor, the generated models based on a fitness score determined for each model to generate a status report for each model; integrate the status reports for each model to determine an anomaly score for the generated models; and generate an alarm when the anomaly score exceeds a predefined threshold.

Abstract: A computer implemented method provides efficient monitoring and analysis of a program's memory objects in the operation stage. The invention can visualize and analyze a monitored program's data status with improved semantic information without requiring source code at runtime. The invention can provide higher quality of system management, performance debugging, and root-cause error analysis of enterprise software in the production stage.

Abstract: A computer implemented method for temporal ranking in invariant networks includes considering an invariant network and a set of broken invariants in the invariant network, assuming, for each time point inside a window W, that each metric with broken invariants is affected by a fault at that time point, computing an expected pattern for each invariant of a metric with assumed fault, said pattern indicative of time points at which an invariant will be broken given that its associated metric was affected by a fault at time t, comparing the expected pattern with the pattern observed over the time window W; and determining a temporal score based on a match from the prior comparing

Abstract: The invention efficiently provides user code information for kernel level tracing approaches. It applies an advanced variation of stack walking called multi-mode stack walking to the entire system level and generates the unified trace where the user code and kernel events are integrated. The invention uses runtime stack information and internal kernel data structures. Therefore, source code for user level code and libraries are not required for inspection. The invention introduces the mechanism to narrow down the monitoring focus to specific application software and improve monitoring performance.

Abstract: A method and apparatus for consolidating a plurality of applications into one or more servers. The method and apparatus organizes consolidation constraints representing preferences about placing applications into the one or more servers, and allocates the applications into the one or more servers in a manner that maximally satisfies the consolidation constraints.

Abstract: A system for automatically instrumenting and tracing an application program and related software components achieves a correlated tracing of the program execution. It includes tracing of endpoints that are the set of functions in the program execution path that the developers are interested. The tracing endpoints and related events become the total set of functions to be traced in the program (called instrument points). This invention automatically analyzes the program and generates such instrumentation points to enable correlated tracing. The generated set of instrumentation points addresses common questions that developers ask when they use monitoring tools.

Abstract: A system and method are disclosed for managing a data center in terms of power and performance. The system includes at least one system positioning application for managing power costs and performance costs at a data center. The at least one system positioning application may determine a status of a data center in terms of power costs and performance costs or generate configurations to automatically implement a desired target state at the data center. A system configuration compiler is configured to receive a request from the system positioning application associated with a data center management task, convert the request into a set of subtasks, and schedule execution of the subtasks to implement the data center management task.

Abstract: A method and system for coordinating energy management in a virtualized data center including a plurality of physical servers and a plurality of virtual machines (VMs), includes analyzing status information about the virtualized data center; determining server utilization target settings for server consolidation from the analyzed status information; and executing the server consolidation according to the determined server utilization target settings. Server consolidation can be executed by determining an effective size of each of the VMs and placing the VMs on the servers in a selective manner using an independent workload VM placement process, a correlation-aware VM placement process, or a migration-cost and correlation-aware VM placement process.

Abstract: A system and method are provided for estimating physical resource utilization information for virtual machines. The system includes a model which indicates how virtual resource activities are transformed into physical resource activities for each virtual machine running on a server. A run-time calibrator utilizes the model to estimate physical resource utiliation information for each individual virtual machine on the server.

Abstract: A system and method are disclosed for managing a data center in terms of power and performance. The system includes at least one system positioning application for managing power costs and performance costs at a data center. The at least one system positioning application may determine a status of a data center in terms of power costs and performance costs or generate configurations to automatically implement a desired target state at the data center. A system configuration compiler is configured to receive a request from the system positioning application associated with a data center management task, convert the request into a set of subtasks, and schedule execution of the subtasks to implement the data center management task.

Abstract: Systems and methods are disclosed that profile event sequences by creating a mixture model from the event sequences; estimating parameters for the mixture model; and applying the mixture model to profile the event sequences.

Abstract: A method system for diagnosing a detected failure in a computer system, compares a failure signature of the detected failure to an archived failure signature contained in a database to determine if the archived failure signature matches the failure signature of the detected failure. If the archived failure signature matches the failure signature of the detected failure, an archived solution is applied to the computer system that resolves the detected failure, the archived solution corresponding to a solution used to resolve a previously detected computer system failure corresponding to the archived failure signature in the database that matches the detected failure.

Abstract: A system and method for prioritizing alerts includes extracting invariants to determine a stable set of models for determining relationships among monitored system data. Equivalent thresholds for a plurality of rules are computed using an invariant network developed by extracting the invariants. For a given time window, a set of alerts are received from a system being monitored. A measurement value of the alerts is compared with a vector of equivalent thresholds, and the set of alerts is ranked.

Abstract: A system and method for optimizing system performance includes applying sampling based optimization to identify optimal configurations of a computing system by selecting a number of configuration samples and evaluating system performance based on the samples. Based on feedback of evaluated samples, a location of an optimal configuration is inferred. Additional samples are generated towards the location of the inferred optimal configuration to further optimize a system configuration.

Abstract: A method and system for predicting performance of an application on a machine of a predetermined hardware configuration simulates the performance of the application under a plurality of different simulated hardware configurations; builds a predictive model of the performance of the application based on the results of the simulations; obtains the performance of the application on a plurality of actual machines, each of the machines having a different hardware configuration; and Bayesian reinterprets the predictive model built from the results of the simulations using the performance of the application on the plurality of actual machines, to obtain a final predictive model of the performance of the application having an accuracy greater than the predictive model built from the results of the simulations.

Abstract: Disclosed is a method and system for modeling invariant relationships between flow intensity measurements in a distributed system. In the method, a measurement is randomly selected from a plurality of flow intensity measurements. The method searched for relationships between the randomly selected measurement and each remaining one of the plurality of flow intensity measurements, and each of the flow intensity measurements having a relationship with the randomly selected measurement is grouped into a cluster with the randomly selected measurement. The method than determines relationships between all of the flow intensity measurements in the cluster. This method is repeated with the remaining flow intensity measurements until all of the flow intensity measurements are grouped into a cluster.