Hewlett-packard Development Company, L.p.

At&t Intellectual Property I, L.p.

University of Massachusetts

Washington University

Timothy Wood

Ludmila Cherkasova

Kadangode K Ramakrishnan

Jacobus Erasmus Van Der Merwe

Prashant Shenoy

Wei Zhang

Abhigyan

Horacio Andres Lagar-Cavilla

Software-based data planes for network function virtualization may use a modular approach in which network functions are implemented as modules that can be composed into service chains. Infrastructures that allow these modules to share central processing unit resources are particularly appealing since they support multi-tenancy or diverse service chains applied to different traffic classes. Systems, methods, and apparatuses introduce schemes for load balancing considering central processing unit utilization of a next hop device when processing a packet that uses a service chain.

Model-based load balancing for network data plane

Pipelined data replication for disaster recovery is disclosed. An example pipelined data replication method for disaster recovery disclosed herein comprises sending replicated first data from a primary processing environment to a secondary processing environment for backup by the secondary processing environment, the replicated first data being a replica of first data in the primary processing environment, processing the first data in the primary processing environment prior to the backup of the replicated first data by the secondary processing environment being confirmed, and preventing a result of the processing of the first data from being released by the primary processing environment until the backup of the replicated first data by the secondary processing environment is confirmed.

Pipelined data replication for disaster recovery

Methods and apparatus to migrate virtual machines between distributive computing networks across a network are disclosed. A disclosed example method includes establishing a data link across a network between a first distributive computing network and a second distributive computing network, the first distributive computing network including a virtual machine operated by a first host communicatively coupled to a virtual private network via a first virtual local area network, communicatively coupling a second host included within the second distributive computing network to the virtual private network via a second virtual local area network, and migrating the virtual machine via the data link by transmitting a memory state of at least one application on the first host to the second host while the at least one application is operating.

Methods and apparatus to migrate virtual machines between distributive computing networks across a wide area network

Methods and apparatus to migrate virtual machines between distributive computing networks across a wide area network are disclosed. A disclosed example method includes establishing a data link across a wide area network between a first distributive computing network and a second distributive computing network, the first distributive computing network including a virtual machine operated by a first host communicatively coupled to a virtual private network via a first virtual local area network, communicatively coupling a second host included within the second distributive computing network to the virtual private network via a second virtual local area network, and migrating the virtual machine via the data link by transmitting a memory state of at least one application on the first host to the second host while the at least one application is operating.

Described herein is a method for scaling a prediction model of resource usage of an application in a virtual environment, comprising: providing a predetermined set of benchmarks, wherein the predetermined set of benchmarks includes at least one of: a computation-intensive workload, a network-intensive workload, and a disk-intensive workload; executing the predetermined set of benchmarks in a first native hardware system in which the application natively resides; executing the predetermined set of benchmarks in the virtual environment; generating at least one first prediction model that predicts a resource usage of the application running in the virtual environment based on the executions of the predetermined set of benchmarks in the first native hardware system and the virtual environment; determining a resource usage of the application running in a second native hardware system in which the application also natively resides; generating at least one second prediction model based on a scaling of the at least one first prediction model by a predetermined constant; and predicting a resource usage of the application running in the virtual environment based on the resource usage of the application running in the second native hardware system and the at least one second prediction model.

Scaling a prediction model of resource usage of an application in a virtual environment

Predicting resource usage of applications running in a virtual environment may include executing benchmarks in a native system and collecting first traces of resource utilization metrics; executing the same benchmarks in a virtual environment and collecting second traces of resource utilization metrics; generating a first prediction model that maps selected first traces to the second traces; generating a second prediction model that maps second selected first traces to the second traces; combining first and second application traces generated from running first and second applications on the native system; and predicting a combined resource usage of the first and second applications running in the virtual environment by applying the first and second prediction models to the combined application traces.

Optimizing a prediction of resource usage of multiple applications in a virtual environment

Described herein is a method for optimizing a prediction of resource usage of an application running in a virtual environment, comprising: providing a predetermined set of benchmarks, wherein the predetermined set of benchmarks; executing the predetermined set of benchmarks in a native hardware system in which the application natively resides; executing the predetermined set of benchmarks in the virtual environment; collecting first traces of first resource utilization metrics in the native hardware system based on the execution of the predetermined set of benchmarks in the native hardware system; collecting second traces of second resource utilization metrics in the virtual environment based on the execution of the predetermined set of benchmarks in the virtual environment; generating a first prediction model and a second prediction model; generating a third prediction model that maps all of the first traces of the selected first metric to the second traces of resource utilization metrics; comparing the first and second prediction models against the third prediction model; and predicting a resource usage of the application running in the virtual environment with either a) a combination of the first and second prediction models or b) the third prediction model based on the comparing.

Optimizing a prediction of resource usage of an application in a virtual environment

Described herein is a system for predicting resource usage of an application running in a virtual environment. The system comprises a first hardware platform implementing a native hardware system in which an application natively resides and executes, the native hardware system operating to execute a predetermined set of benchmarks that includes at least one of: a computation-intensive workload, a network-intensive workload, and a disk-intensive workload; a second hardware platform implementing a virtual environment therein, the virtual environment operating to execute the predetermined set of benchmarks; a third hardware platform operating to collect first resource usage traces from the first hardware platform and second resource usage traces from the second hardware platform; wherein the third hardware platform further operating to generate at least one prediction model that predicts a resource usage of the application executing in the virtual environment based on the collected first and second resource usage traces.

Predicting resource usage of an application in a virtual environment

Described herein is a system for improving accuracy in a prediction of resource usage of an application running in a virtual environment. The system comprises a first hardware platform implementing a native hardware system in which an application natively resides and executes, the native hardware system operating to execute a predetermined set of benchmarks that includes at least one of a network-intensive workload and a disk-intensive workload, a second hardware platform implementing a virtual environment therein, the virtual environment operating to execute the predetermined set of benchmarks, and a third hardware platform operating to collect first resource usage traces from the first hardware platform that result from the execution of the predetermined set of benchmarks in the native hardware system and second resource usage traces from the second hardware platform that result from the execution of the predetermined set of benchmarks in the virtual environment. The third hardware platform further operating to perform a linear regression computation to generate at least one prediction model that predicts a resource usage of the application executing in the virtual environment based on the collected first and second resource usage traces.

Accuracy in a prediction of resource usage of an application in a virtual environment

Described herein is a method for detecting an error in a prediction of resource usage of an application running in a virtual environment, comprising: providing a plurality of benchmark sets, executing the plurality of benchmark sets in a native hardware system in which the application natively resides, executing the plurality of benchmark sets in the virtual environment, collecting first traces of first resource utilization metrics in the native hardware system based on the execution of each of the plurality of benchmark sets in the native hardware system, collecting second traces of second resource utilization metrics in the virtual environment based on the execution of each of the plurality of benchmark sets in the virtual environment, generating at least one initial prediction model that maps the first traces of first resource utilization metrics to the second traces of second resource utilization metrics, computing a plurality of mean squared errors (MSE's) based on the at least one initial prediction model, each of the MSE's is further based on and associated with the collected first and second traces for a different one of the plurality of benchmark sets, and determining whether to use the initial prediction model to predict a resource usage of the application running in the virtual environment based on the plurality of MSEs.

Growing community of inventors

Amherst, MA, United States of America

Timothy Wood