摘要:
Embodiments of the present invention provide an approach for adapting an information extraction middleware for a clustered computing environment (e.g., a cloud environment) by creating and managing a set of statistical models generated from performance statistics of operating devices within the clustered computing environment. This approach takes into account the required accuracy in modeling, including computation cost of modeling, to pick the best modeling solution at a given point in time. When higher accuracy is desired (e.g., nearing workload saturation), the approach adapts to use an appropriate modeling algorithm. Adapting statistical models to the data characteristics ensures optimal accuracy with minimal computation time and resources for modeling. This approach provides intelligent selective refinement of models using accuracy-based and operating probability-based triggers to optimize the clustered computing environment, i.e., maximize accuracy and minimize computation time.
摘要:
The present invention proactively identifies hotspots in a cloud computing environment through cloud resource usage models that use workload parameters as inputs. In some embodiments the cloud resource usage models are based upon performance data from cloud resources and time series based workload trend models. Hotspots may occur and can be detected at any layer of the cloud computing environment, including the server, storage, and network level. In a typical embodiment, parameters for a workload are identified in the cloud computing environment and inputted into a cloud resource usage model. The model is run with the inputted workload parameters to identify potential hotspots, and resources are then provisioned for the workload so as to avoid these hotspots.
摘要:
Embodiments of the invention relate to dynamic application migration in a shared pool of configurable computer resources with disaster recovery support. Data from an application is replicated from local data storage to remote data storage. A consistency point of the data is created in both the local data storage and the remote data storage. The application may be migrated to a second data site with separate local data storage. The migration may be planned or unplanned. Based upon the created consistency point, a consistent set of application data may be requested to support a read operation from the migrated application.
摘要:
Embodiments of the invention are directed to streaming virtual machine boot services over a network. An aspect of the invention includes booting a first virtual machine and recording data and metadata from a virtual machine boot image into a virtual machine boot file. The data and metadata are accessed in the process of booting the first virtual machine. The virtual machine boot image has setup information of the virtual machine type of the first virtual machine. The virtual machine boot file is configured for the virtual machine type of the first virtual machine. A descriptor is added to metadata of the virtual machine boot image, which references a location of the virtual machine boot file for the virtual machine type of the first virtual machine. When subsequently booting a second virtual machine of the same type of virtual machine as the first virtual machine, data are streamed from the virtual machine boot file to a virtual machine monitor of a second virtual machine without the need to stream data from the virtual machine boot image.
摘要:
Embodiments of the present invention provide an approach for migrating virtual machines across network (e.g., WAN) separated data centers (e.g., storage clouds). Specifically, under embodiments of the present invention, a first storage system associated with a first data center is synchronized with a second storage system associated with a second data center via a storage system link. Then, a minimal state of a virtual machine is migrated from a first computer in the first data center to a second computer in the second data center via a WAN link. Using the minimal state, the virtual machine is stored in the second computer. Thereafter, the storage system link is terminated. In addition, as updated pages are received in memory of the first computer, they are migrated to the second computer via the WAN link. Once this migration is complete, the WAN link can be terminated. Therefore, embodiments of the present invention provide at least two forms of synchronization: computational synchronization and storage synchronization.
摘要:
Embodiments of the present invention provide an approach for intelligent storage planning and planning within a clustered computing environment (e.g., a cloud computing environment). Specifically, embodiments of the present invention will first determine/identify a set of storage area network volume controllers (SVCs) that is accessible from a host that has submitted a request for access to storage. Thereafter, a set of managed disk (mdisk) groups (i.e., corresponding to the set of SVCs) that are candidates for satisfying the request will be determined. This set of mdisk groups will then be filtered based on available space therein, a set of user/requester preferences, and optionally, a set of performance characteristics. Then, a particular mdisk group will be selected from the set of mdisk groups based on the filtering.
摘要:
Management of a planner through use of a middleware layer. A computer system is configured with the middleware layer in communication with both a planner and a data repository. One or more modules are provided in the middleware layer to support the functionality of the planner. Application program interface calls are employed to call the modules, thereby mitigating duplication of the functionality in separate planners.
摘要:
Server consolidation using virtual machine resource tradeoffs, is provided. One implementation involves assigning a virtual machine to a target physical server based on a plurality of virtualization parameters for maximizing utility of a plurality of virtual machines and physical servers. The assigning performs resource allocation for the virtual machine based on capabilities of the target physical server and a plurality of virtual machine resource requirements. Virtualization parameters include a reservation parameter (min) representing a minimum resources required for a VM, a limit parameter (max) representing a maximum resources allowable for the VM, and a weight parameter (shares) representing a share of spare resources for the VM.
摘要:
The invention provides a method and system for continuous optimization of a data center. The method includes monitoring loads of storage modules, server modules and switch modules in the data center, detecting an overload condition upon a load exceeding a load threshold, combining server and storage virtualization to address storage overloads by planning allocation migration between the storage modules, to address server overloads by planning allocation migration between the server modules, to address switch overloads by planning allocation migration mix between server modules and storage modules for overload reduction, and orchestrating the planned allocation migration to reduce the overload condition in the data center.
摘要:
A program, method and system are disclosed for planning the placement of a collection of applications in a heterogeneous storage area network data center. The program, method, and system disclosed deal with the coupled placement of virtual machine applications within a resource graph, with each application requiring a certain amount of CPU resources and a certain amount of storage resources from the connected resource node pairs within the resource graph. The resource nodes in the graph provide either storage resources, CPU resources, or both and can have differing degrees of affinity between different node pairs. Various placement algorithms may be used to optimize placement of the applications such as an individual-greedy, pair-greedy or stable marriage algorithm. One placement objective may be to place the programs among nodes of the resource graph without exceeding the storage and CPU capacities at nodes while keeping the total cost over all applications small.