International Business Machines Corporation

Qualcomm Incorporated

Dilma Menezes Da Silva

Alper Buyuktosunoglu

Pradip Bose

Hubertus Franke

Priyanka Tembey

Eli M Dow

Balaram Sinharoy

Saumitra Mohan Das

Hui Chao

Orran Yaakov Krieger

Robert William Wisniewski

Robert Brett Tremaine

Elmootazbellah Nabil Elnozahy

Xiaowei Shen

Hazim Shafi

Michael R Hines

Jonathan Appavoo

Behrooz Khorashadi

Shmuel Ben-Yehuda

Abel Alkon Gordon

David Breitgand

Kyung D Ryu

Marc Alan Auslander

Amir Epstein

Marcio Augusto Silva

Alexander Glikson

Vrajesh Rajesh Bhavsar

Andrew Arnott Baumann

Kevin Kin-Fai Hui

Power consumption in a microprocessor platform is managed by setting a peak power level for power consumed by a multi-core microprocessor platform executing multi-threaded applications. The multi-core microprocessor platform contains a plurality of physical cores, and each physical core is configurable into a plurality of logical cores. A simultaneous multithreading level in at least one physical core is adjusted by changing the number of logical cores on that physical core in response to a power consumption level of the multi-core microprocessor platform exceeding the peak power level. Performance and power data based on simultaneous multi-threading levels are used in selecting the physical core to be adjusted.

Power shifting in multicore platforms by varying SMT levels

A method for providing dynamic resizing of at least one virtual machine in a cloud infrastructure includes steps of: obtaining a measure of resource usage for the virtual machine and/or application performance of an application running on the virtual machine; determining, as a function of prescribed rules relating to resource usage and the measure of resource usage and/or application performance, whether the virtual machine requires resizing; and resizing the virtual machine to dynamically change resource availability for the virtual machine so as to maintain resource usage and/or application performance within prescribed limits defined by the prescribed rules.

Dynamic virtual machine resizing in a cloud computing infrastructure

An example method of synchronizing application data between a mobile device and a remote computing device includes for each network of one or more networks, collecting network data regarding a mobile device's current connectivity state to the respective network. The mobile device includes one or more mobile applications including application data. The method also includes detecting, at the mobile device, a current connection to a current network. The method further includes comparing one or more opportunity scores for synchronizing application data. The one or more opportunity scores includes an opportunity score for the current network. Each opportunity score is associated with a particular network. The method also includes determining, based on the comparing, whether the current network is a good opportunity network for syncing mobile application data.

Methods and systems for triggering efficient application synchronization

Various methods, apparatuses, devices, and articles of manufacture are provided for use by one or more electronic devices to provide or otherwise identify applications for potential use by a mobile device with respect to a particular environment. For example, a computing device may apply a context-aware filter to identify a list of applications for potential use by a mobile device; and, in response to a first event proffer one or more applications for use by the mobile device; and, in response to a second event withdraw the application for use by the mobile device. In another example, a mobile device may, in response to a first indication that an application has been proffered for use, initiate activation of the application; and, in response to a second indication that the application has been withdrawn for use, initiate deactivation of the application.

Method and apparatus for use in providing context-aware identification of mobile device applications

A method for load balancing virtual machines includes accessing current consumption data and historical consumption data for a first virtual machine running on a host hardware platform, wherein the host hardware platform is coupled to a network and accessing specification and utilization information for networked hardware platforms published on the network by each networked hardware platform, wherein the networked hardware platforms each include a hardware platform configured to run virtual machines. The method also includes selecting a target platform from the networked hardware platforms to receive the first virtual machine based on the published specification and utilization information for the networked hardware platforms and consumption data for the first virtual machine, wherein the first virtual machine will operate on the target platform in a condition that is not over-constrained.

Virtual machine load balancing

A method for load balancing virtual machines includes accessing specification and utilization information for selected networked hardware platforms published on a network, the networked hardware platforms configured to run virtual machines and scoring the selected networked hardware platforms based on parameters from the utilization information to indicate resource availability for each of the selected networked hardware platforms. The method also includes removing a networked hardware platform from the selected networked hardware platforms based on a collocation policy to produce candidate platforms to receive a first virtual machine, wherein the collocation policy prohibits placement of the first virtual machine on the networked hardware platform based on at least one of consumer rules, provider rules, and industry rules.

A memory management method in a virtualized computing environment is provided, in which a hypervisor implements at least a virtual machine (VM) over a host machine, wherein a guest operating system (OS) is executed over the VM and an application supporting memory management capabilities is executed over the guest OS. The method comprises invoking a first memory manager (java balloon) implemented by the application to deallocate memory allocated to the application for use by the hypervisor, in response to a request submitted by the hypervisor; and invoking a second memory manager (guest balloon) implemented over the guest operating system to deallocate memory allocated to the guest OS, in response to a request submitted by the hypervisor.

Dynamic memory management in a virtualized computing environment

A method for virtual machine management that includes establishing a first virtual machine on a hardware platform, performing a selected task on the first virtual machine and recording a first data set indicating a characteristic of the first virtual machine performing the selected task. The method also includes establishing a second virtual machine on the hardware platform, performing the selected task on the first and second virtual machines, recording a second data set indicating the characteristic of the first and second virtual machines performing the selected task and indicating acceptable data and unacceptable data within the first and second data sets. The method also includes creating and training a decision tree based on the acceptable and unacceptable data from the first and second data sets and inputting a third data set from a third virtual machine into the decision tree to determine if the third virtual machine is over-constrained.

Identification of over-constrained virtual machines

Methods, systems, and media for reducing memory latency seen by processors by providing a measure of control over on-chip memory (OCM) management to software applications, implicitly and/or explicitly, via an operating system are contemplated. Many embodiments allow part of the OCM to be managed by software applications via an application program interface (API), and part managed by hardware. Thus, the software applications can provide guidance regarding address ranges to maintain close to the processor to reduce unnecessary latencies typically encountered when dependent upon cache controller policies. Several embodiments utilize a memory internal to the processor or on a processor node so the memory block used for this technique is referred to as OCM.

Methods and arrangements to manage on-chip memory to reduce memory latency

To dynamically update an operating system, a new factory object may have one or more new and/or updated object instances. A corresponding old factory object is then located and its version is checked for compatibility. A dynamic update procedure is then executed, which includes (a) changing a factory reference pointer within the operating system from the old factory object to the new factory object. For the case of updated object instances, (b) hot swapping each old object instance for its corresponding updated object instance, and (c) removing the old factory object. This may be performed for multiple updated object instances in the new factory object, preferably each separately. For the case of new object instances, they are created by the new factory and pointers established to invoke them. A single factory object may include multiple updated objects from a class, and/or new object instances from different classes, and the update may be performed without the need to reboot the operating system.

Dynamic update mechanisms in operating systems

Systems, especially operating systems, are becoming more complex to the point where maintaining them by humans is becoming nearly impossible. Many corporations have recognized this trend and have begun investing in autonomic technology. Autonomic technology allows a piece of software to monitor, diagnose, and repair itself. This can be used for improved performance, reliability, maintainability, security, etc. Disclosed herein is a mechanism to allow operating systems to hot swap a piece of operating system code, while continuing to offer to the user the service which that code is providing. This can be used, for examples, to increase the performance of an application or to fix a detected security hole live without bringing the machine down. Some autonomic ability will be mandatory in next generation operating system for without it they will collapse under their own complexity. The invention offers a key component of being able to achieve autonomic computing.

Growing community of inventors

White Plains, NY, United States of America

Dilma Menezes Da Silva