Abstract: A trusted virtualization platform protects sensitive customer data during operation of virtual machines in a multi-tenant cloud computing center. The trusted virtualization platform limits administrator access to the data and state of the virtual machines running thereon, reports any changes made thereto, and requires keys provided by the customer or a trusted third party of the customer to perform management operations on the virtual machines. By requiring cloud computing centers to use such trusted virtualization platforms, customers uploading their virtual machines into the cloud computing center can be assured that cloud administrators will not be able to access or tamper with their private data. Furthermore, customers can directly audit all important state or configuration changes for their virtual machines as the trusted virtualization platform can be configured to report all such changes according to a security policy set by the customer.

Abstract: In one embodiment, a latency value is determined for an input/output IO request in a host computer of a plurality of host computers based on an amount of time the IO request spent in the host computer's issue queue. The issue queue of the host computer is used to transmit IO requests to a storage system shared by the plurality of host computers. The method determines a host specific value assigned to the host computer based in proportion on a number of shares assigned to the host in a quality of service policy for IO requests. The size for the host computer's issue queue is determined based on the latency value and the host specific value to control a number of IO requests that are added to the host computer's issue queue where other hosts in the plurality of hosts independently determine respective sizes for respective issue queues.

Abstract: Embodiments associate software applications with computing resource containers based on placement rules. A placement rule indicates that a first software application is to be co-located with a second software application during execution of the first and second software applications and second placement rule indicates that the first software application is to be separated from the second software application when the second placement rule is violated by enforcing the first placement rule. The placement rule also indicates that the first software application is to be separated from the second software application during execution of the first and second software applications and the second placement rule indicates the first software application is to be co-located with the second software application when the second placement rule is violated by enforcing the first placement rule.

Abstract: Embodiments associate software applications with computing resources based on failure correlation information and an anti-affinity rule. An anti-affinity rule indicates that a first software application is to be separated from a second software application during execution. A management device determines failure correlations between a first computing resource that is associated with the first software application and a plurality of computing resources other than the first computing resource. The management device selects the computing resource that corresponds to the lowest failure correlation and associates the second software application with the selected computing resource based on the anti-affinity rule.

Abstract: A resource management system for a virtual machine computing environment includes a software component that optimizes capacity between server clusters or groups by monitoring the capacity of server clusters or groups and automatically adding and removing host systems to and from server clusters or groups. The software component may be implemented at a server cluster management level to monitor and execute host system moves between server clusters and/or at a higher level in the resource management hierarchy. At the higher level, the software component is configured to monitor and execute host system moves between sets of server clusters being managed by different server cluster management agents.

Abstract: Cache utility curves are determined for different software entities depending on how frequently their storage access requests lead to cache hits or cache misses. Although possible, not all access requests need be tested, but rather only a sampled subset, determined by whether a hash value of each current storage location identifier (such as an address or block number) meets one or more sampling criteria. The sampling rate is adaptively changed so as to hold the number of location identifiers needed to be stored to compute the cache utility curves to within a set maximum limit.

Abstract: One or more embodiments of the present invention provide a technique for effectively managing virtualized computing systems with an unlimited number of hardware resources. Host systems included in a virtualized computer system are organized into a scalable, peer-to-peer (P2P) network in which host systems arrange themselves into a network overlay to communicate with one another. The network overlay enables the host systems to perform a variety of operations, which include dividing computing resources of the host systems among a plurality of virtual machines (VMs), load balancing VMs across the host systems, and performing an initial placement of a VM in one of the host systems.

Abstract: An anomaly in a shared input/output (IO) resource that is accessed by a plurality hosts or clients is detected when a host that is not bound by any QoS policy presents large workloads to a shared IO resource that is also accessed by hosts or clients that are governed by QoS policy. The anomaly detection triggers a response from the hosts or clients as a way to protect against the effect of the anomaly. The response is an increase in window sizes. The window sizes of the hosts or clients may be increased to the maximum window size or in proportion to their QoS shares.

Abstract: A technique for predictive distributed resource scheduling and distributed power management includes analyzing patterns in the workload, predicting future workloads, and making recommendations for changes to the virtual computing environment. In addition, a cost-benefit analysis can be performed to determine whether the recommended change would likely result in improved performance.

Abstract: Activity level of memory pages is repeatedly classified in a virtual machine environment, so that live VM migration can be carried out more efficiently. The time intervals upon which the activity level of the memory pages are repeatedly classified can be dynamically adjusted to better align its performance with the live VM migration process.

Abstract: A shared input/output (IO) resource is managed in a decentralized manner. Each of multiple hosts having IO access to the shared resource, computes an average latency value that is normalized with respect to average IO request sizes, and stores the computed normalized latency value for later use. The normalized latency values thus computed and stored may be used for a variety of different applications, including enforcing a quality of service (QoS) policy that is applied to the hosts, detecting a condition known as an anomaly where a host that is not bound by a QoS policy accesses the shared resource at a rate that impacts the level of service received by the plurality of hosts that are bound by the QoS policy, and migrating workloads between storage arrays to achieve load balancing across the storage arrays.

Abstract: In one embodiment, a latency value is determined for an input/output IO request in a host computer of a plurality of host computers based on an amount of time the IO request spent in the host computer's issue queue. The issue queue of the host computer is used to transmit IO requests to a storage system shared by the plurality of host computers. The method determines a host specific value assigned to the host computer based in proportion on a number of shares assigned to the host in a quality of service policy for IO requests. The size for the host computer's issue queue is determined based on the latency value and the host specific value to control a number of IO requests that are added to the host computer's issue queue where other hosts in the plurality of hosts independently determine respective sizes for respective issue queues.

Abstract: Embodiments associate software applications with computing resource containers based on placement rules. A placement rule indicates that a first software application is to be co-located with a second software application during execution of the first and second software applications, or that the first software application is to be separated from the second software application during execution of the first and second software applications. A target computing resource container is selected based on the placement rule and a computing resource container that is associated with the first software application. The second software application is associated with the target computing resource container, and the placement rule may be provided to the target computing resource container.

Abstract: In one embodiment, a method receives current latency values from a plurality of host computers where a current latency value is calculated by a respective host computer based on an amount of time spent in the respective host computer's issue queue by an IO request most recently removed from the issue queue of the respective host computer. The issue queue of the respective host computer is used to transmit IO requests from the respective host computer to a storage system. The method then calculates a combined average latency value based on the current latency values and sends the combined average latency value to the plurality of host computers. Each respective host computer adjusts a size of the respective host computer's issue queue based on the combined average latency value, and the size controls a number of IO requests that are added to the respective host computer's issue queue.

Abstract: A shared input/output (IO) resource is managed in a decentralized manner. Each of multiple hosts having IO access to the shared resource, computes an average latency value that is normalized with respect to average IO request sizes, and stores the computed normalized latency value for later use. The normalized latency values thus computed and stored may be used for a variety of different applications, including enforcing a quality of service (QoS) policy that is applied to the hosts, detecting a condition known as an anomaly where a host that is not bound by a QoS policy accesses the shared resource at a rate that impacts the level of service received by the plurality of hosts that are bound by the QoS policy, and migration of workloads between storage arrays to achieve load balancing across the storage arrays.

Abstract: The Arabic bank check analysis and zone extraction method for courtesy amount, legal amount, and the date fields uses available information and performs adaptive local processing using projections and connected component analysis. The method performs skew correction and connected component analysis on the check image and removes components smaller than average size before identifying the regions using projection profiles. This leads to a highly effective and robust technique. The method is based on prior knowledge of the check zone structure.

Abstract: One or more embodiments of the present invention provide a technique for effectively managing virtualized computing systems with an unlimited number of hardware resources. Host systems included in a virtualized computer system are organized into a scalable, peer-to-peer (P2P) network in which host systems arrange themselves into a network overlay to communicate with one another. The network overlay enables the host systems to perform a variety of operations, which include dividing computing resources of the host systems among a plurality of virtual machines (VMs), load balancing VMs across the host systems, and performing an initial placement of a VM in one of the host systems.

Abstract: A method for managing an amount of IO requests transmitted from a host computer to a storage system is described. A current latency value of an IO request most recently removed from an issue queue maintained by a host bus adapter of the host computer in order to transmit IO requests from the host computer to the storage system is periodically determined. An average latency value is the calculated based on the current latency value and a size limit of the issue queue is adjusted based in part on the average latency value. Upon receiving an IO request from one of a plurality of client applications running on the host computer, it can then be determined whether a number of pending IO requests in the issue queue has reached the size limit and the IO request can be transmitted to the issue queue if the number of pending IO request falls within the size limit.

Abstract: In a computer system, a method of controls interrupts which correspond to input/output (I/O) processing. For each delivery of an I/O completion interrupt, the method provides a recordation of a delivery time; identifies I/O completions for which deliveries of corresponding I/O completion interrupts involve deliveries of inter-processor interrupts; and for each of the identified I/O completions, accesses the recordation of the most recent delivery time to determine whether a selected period of time has elapsed since a last delivery of an inter-processor interrupt. As a response to a determination that the selected period has elapsed, an inter-processor interrupt is delivers. As a response to a determination that less than the duration of the selected period has elapsed, the method refrains from delivering an inter-processor interrupt.

Abstract: Two shared data structures are maintained in a storage unit for coordinating statistic collection among multiple hosts that share the storage unit. The first data structure stores information about the number of hosts that possess slots within the second data structure and identifying information about the slots. The second data structure maintains statistics of each of the multiple hosts. By using this first data structure, hosts can be added to and deleted from the second data structure in an organized manner.