Hewlett-Packard

About: Hewlett-Packard is a company organization based out in Palo Alto, California, United States. It is known for research contribution in the topics: Signal & Substrate (printing). The organization has 34663 authors who have published 59808 publications receiving 1467218 citations. The organization is also known as: Hewlett Packard & Hewlett-Packard Company.

Enforcing performance isolation across virtual machines in Xen

Abstract: Virtual machines (VMs) have recently emerged as the basis for allocating resources in enterprise settings and hosting centers. One benefit of VMs in these environments is the ability to multiplex several operating systems on hardware based on dynamically changing system characteristics. However, such multiplexing must often be done while observing per-VM performance guarantees or service level agreements. Thus, one important requirement in this environment is effective performance isolation among VMs. In this paper, we address performance isolation across virtual machines in Xen [1]. For instance, while Xen can allocate fixed shares of CPU among competing VMs, it does not currently account for work done on behalf of individual VMs in device drivers. Thus, the behavior of one VM can negatively impact resources available to other VMs even if appropriate per-VM resource limits are in place.In this paper, we present the design and evaluation of a set of primitives implemented in Xen to address this issue. First, XenMon accurately measures per-VM resource consumption, including work done on behalf of a particular VM in Xen's driver domains. Next, our SEDF-DC scheduler accounts for aggregate VM resource consumption in allocating CPU. Finally, ShareGuard limits the total amount of resources consumed in privileged and driver domains based on administrator-specified limits. Our performance evaluation indicates that our mechanisms effectively enforce performance isolation for a variety of workloads and configurations.

Inheritance is not subtyping

Abstract: In typed object-oriented languages the subtype relation is typically based on the inheritance hierarchy. This approach, however, leads either to insecure type-systems or to restrictions on inheritance that make it less flexible than untyped Smalltalk inheritance. We present a new typed model of inheritance that allows more of the flexibility of Smalltalk inheritance within a statically-typed system. Significant features of our analysis are the introduction of polymorphism into the typing of inheritance and the uniform application of inheritance to objects, classes and types. The resulting notion of type inheritance allows us to show that the type of an inherited object is an inherited type but not always a subtype.

Template-driven approach for generating models on network services

Abstract: A method and system of modeling a selected service within a network environment includes forming a service model template that is not specific to the network environment, but identifies anticipated network elements and network services that cooperate to enable the selected service. The service model template includes specifications related to linking the anticipated network elements and network services. When the service model template is combined with discovered instance information that is specific to actual network elements and actual network services, a service model instance is generated for the selected service. The service model instance includes identifications of dependencies among the elements and services. Preferably, the service model instance also includes identification of the “health” of the different elements and services. A view generator may be used to configure the service model instance as a hierarchical graph of nodes that are linked to identify the dependencies among the nodes. The discovered instance information may be acquired using a discovery template that orchestrates the process. The discovery template may identify auto-discovery modules that are to be used to detect actual network elements and services and may identify dependencies among the modules. A user-configurable discovery engine accesses the discovery template and deploys the modules in order to acquire the discovered instance information.

Distributed workflow resource management system and method

Abstract: A system and method for distributed resource management in a computer network operating under control of workflow management software system (WFMS) to manage plural resources to perform a workflow process that includes multiple process activities, uses two-step resource assignment to manage resources at two levels. The resources are grouped by capabilities into resource groups controlled by a local resource manager (LRM), which maintains dynamic status of each resource, such as availability and load, as well as their roles and addresses. A global resource manager (GRM) stores resource data only about the groups: capabilities and status. The WFMS invokes the GRM, requesting a specified resource activity. The GRM checks the stored capabilities and status among the resource groups, selects the resource group capable of the specified activity and available, and forwards the request to the LRM for the selected group. The LRM selects one of the resources in the group to perform the specified activity and assigns the activity to that resource.