1876 results about "Pool" patented technology

A data network may include an upstream router having one or more data handling queues, a downstream router, and a policy server. In one embodiment, the policy server includes processing resources, a communication interface in communication with the processing resources, and data storage that stores a configuration manager executable by the processing resources. The configuration manager configures data handling queues of the upstream router to provide a selected bandwidth to one or more of a plurality of service classes of data flows. In addition, the configuration manager transmits to the downstream router one or more virtual pool capacities, each corresponding to a bandwidth at the upstream router for one or more associated service classes among the plurality of service classes. In one embodiment, the configuration manager configures the data handling queues on the upstream router only in response to acknowledgment that one or more virtual pool capacities transmitted to the downstream router were successfully installed.

It is provided a resource management server comprising: a resource management module for managing virtual machines and use amounts of resources used by the virtual machines for deployment; a pool management module for managing a total amount of the resources and a total amount of available resources; a requirement specification reception module for receiving an allocation requirement for a new virtual machine along with requirement specifications including amounts of resources; and a search module for searching the virtual machines for deployment included in the resource management information for virtual machines for deployment having amounts of resources that satisfy the requirement specifications, and for virtual machines for deployment which do not have the amounts of resources that satisfy the requirement specifications but which allow the amounts of resources that satisfy the requirement specifications to be secured by adding the available resources included in the pool management information.

An apparatus, system, and method are disclosed for managing contents of a cache. A storage request module monitors storage requests received by a cache. The storage requests include read requests and write requests. A read pool module adjusts a size of a read pool of the cache to maximize a read hit rate of the storage requests. A dirty write pool module adjusts a size of a dirty write pool of the cache to maximize a dirty write hit rate of the storage requests.

Techniques for fast recovery and/or balancing of connections to a clustered computing system provide management of such connections by determining a number of connections to load balance across nodes and by triggering creation of such connections. In one aspect, a notification of an event regarding the clustered computing system is received by a connection pool manager, a pool of connections to the system is identified based on the notification, and one or more connections from the pool are processed in response to the event. According to an embodiment, the notification comprises the identities of the service, database, server instance and machine that correspond to the event.

A management system manages a storage system and a server computer. The management system (A) acquires the storage level pool information that includes the information of a capacity of a storage level pool from the storage system, (B) acquires the server level pool information that includes the information of a capacity of a server level pool from the server computer, (C) determines a first risk degree that indicate a risk of a depletion of a free capacity of the storage level pool based on the storage level pool information, (D) determines a second risk degree that indicate a risk of a depletion of a free capacity of a server level pool based on the server level pool information, and (E) displays the information that is associated with the first risk degree that indicate a risk of a depletion of a free capacity of the storage level pool and the second risk degree that indicate a risk of a depletion of a free capacity of the server level pool.

A cloud computing environment can be configured to allow third party, user-controlled resources to be included in the pool of resources available in the cloud, regardless of the type of system supporting the user-controlled resources. A cloud management system can identify a particular type of management system supporting the user-controlled resources and the commands necessary to communicate with the particular type of management system. Once the particular type of management system is identified, the cloud management system can communicate with the particular type of management system, utilizing the determined commands and protocols, to coordinate access to the user-controlled resources. The cloud management system can identify, allocate, and manage a set of resources from the user-controlled resources utilizing the commands and protocols.

A methodology and algorithm for managing resources from classes within a pool of resources to determine whether and what resources may be allocated upon a request for resources.

A buffer manager that manages blocks of memory amongst multiple levels of buffer pools. For instance, there may be a first level buffer pool for blocks in first level memory, and a second level buffer pool for blocks in second level memory. The first level buffer pool evicts blocks to the second level buffer pool if the blocks are not used above a first threshold level. The second level buffer pool evicts blocks to a yet lower level if they have not used above a second threshold level. The first level memory may be dynamic random access memory, whereas the second level memory may be storage class memory, such as a solid state disk. By using such a storage class memory, the working block set of the buffer manager may be increased without resorting to lower efficiency random block access from yet lower level memory such as disk.

In a distributed network hosting a shared pool of configurable computing resources there is provided one or more system instances of a blockchain. Each system instance comprises a virtual machine and a set of containers. The containers are members of the blockchain in that the blockchain's public ledger records encrypted copies of at least selected directories of each container. Each container that is in the set can thus verify whether any other container also belongs to the same set with reference to the public ledger, since the transactions recorded on the public ledger are encrypted copies of the set's containers. The use of a blockchain thus allows a system boundary to be defined around a set of containers by the initial specification of the blockchain. The system boundary can be defined to ensure that the set of containers comply with legal requirements, such as a geographic restriction.

Data is placed in tiered storage with a suitable granularity according to application characteristics. The storage apparatus comprises a controller for managing storage areas, provided by storage media of a plurality of types of varying performance, as pools, and for assigning the storage areas in page units to a virtual volume from any tiered storage among a plurality of types of tiered storage which the pool comprises in response to a data write request from the host computer, wherein, for specific data which is managed by the host computer, the controller specifies an area with a high referencing frequency among the specific data on the basis of organization information of the specific data, and moves this area to another of the tiered storage with a higher performance than an already assigned tiered storage.

The invention provides a computing unloading scheduling method based on depth reinforcement learning, which provides a method for computing unloading to make unloading decision for Internet of Thingsequipment, including making decisions on various aspects needing to be unloaded according to the basic model of computing unloading. Based on different optimization objectives, different optimizationobjectives can be achieved by changing the value function. TheDeep-SARSA algorithm is similar to DQN algorithm, which combines reinforcement learning and depth learning. It can effectively change theunloading state and unloading action into training samples of depth learning when cooperating with experience pool. The invention can effectively carry out machine learning on an unloading state modelof an unlimited dimension, reducing the complexity of learning, this method uses neural network as the linear approximator of Q value, which can effectively improve the training speed and reduce thesample required for training. This method can effectively make the best decision through deep reinforcement learning under the given model and optimization objective.

Guest domains (virtual machines) may execute at a virtualization server. During execution, the guest domains may request various services to facilitate their operation. The virtualization server may disaggregate the services requested by the guest domains among various service domains. A service domain may be configured to provide one the services. The virtualization server may also maintain a pool of partially initialized service domains. When a guest domain requests a particular service, one of the partially initialized service domains may be selected from the pool, and the partially initialized service domain may complete initialization to become an initialized service domain. The initialized service domain may thus be configured to provide the service requested by a guest domain. The virtualization server may periodically replenish the pool of partially initialized service domains to ensure that a partially initialized service domain is available upon demand from one of the guest domains.

A visual tool may be provided to display information associated with computer job allocation and to allow a user to explore different job configurations. Jobs executing on a computing environment comprising a shared pool of configurable computing resources may be monitored. Cost and duration estimates may be determined with uncertainty associated with the cost and duration estimates. Sandbox environment may be provided that allow users to manipulate one or more different job configuration options for executing the jobs in the computing environment.

The invention discloses a method and a device using a vector instruction to process a file index in a parallel mode. The method comprises the following steps of calculating and sorting index key values of data in a data table, distributing needed storing space of all leaf nodes, using an SIMD (single instruction multiple data) instruction to load a plurality of index key values once, sequentially filling the data in all the leaf nodes, distributing the needed storing space of the internal nodes of the upper layer, building the internal nodes, continuously building the internal nodes of the upper layer until the node number of the upper layer is one, regarding the node as a root node, and finishing the building of an index tree; building a buffer pool in internal storage, using an inquiring queue to temporarily store all received inquiring requests, taking the inquiring requests out of the buffering pool when the length of the inquiring queue reaches a preset threshold value, sorting the inquiring requests, and forming a sorted to-be-inquired queue; sequentially obtaining a to-be-inquired term in the sorted inquiring queues, sequentially searching from the root node of the built index tree, and using the SIMD method to search the to-be-inquired term in a parallel mode inside the internal nodes and the leaf nodes.

Exemplary embodiments provide a technique to manage object based tier to improve allocation of media to unallocated area. In one embodiment, a method of allocating an area of a logical volume to an unallocated area of a virtual volume for a write command comprises: calculating an object location of an object based on the write command and an object allocation information, the write command containing a virtual volume name and a virtual volume address of a virtual volume; selecting a tier from a plurality of tiers based on the calculated object location and an object and tier definition information; selecting a media type from a plurality of media types based on the selected tier and a tier and media definition information; and selecting a logical volume from a plurality of logical volumes based on the virtual volume specified by the write command, the object allocation information, a pool information, and the selected media type.

The invention discloses a system and method for automatically deploying an application system in a cloud resource pool, and relates to the field of cloud computation. An application system template is analyzed and divided into a single server template, an internal private network setting, a public network setting, an application setting and a gateway setting; a VLAN and a subnet are allocated to the application system according to the internal private network setting, and a DHCP server is configured so that an IP designated for each server can be kept; all the servers are set up according to the single server template in the sequence of logic entities; a network device control interface is called according to the public network setting so that a private network routing strategy, a security strategy, a firewall strategy of an exit gateway, and an address mapping of the private network and the public network can be set; a service issuing interface is called to send an application configuration parameter list to an application configuration system so that application initialization can be completed, an application service gateway is set up according to the application external service type and the gateway setting, and therefore the application system is automatically deployed in the cloud resource pool and can adapt to deploying of various different servers.

Embodiments of the disclosure relate to archiving data in a storage system. An exemplary embodiment comprises making a flash copy of data in a source volume, compressing data in the flash copy wherein each track of data is compressed into a set of data pages, and storing the compressed data pages in a target volume. Data extents for the target volume may be allocated from a pool of compressed data extents. After each stride worth of data is compressed and stored in the target volume, data may be destaged to avoid destage penalties. Data from the target volume may be decompressed from a flash copy of the target volume in a reverse process to restore each data track, when the archived data is needed. Data may be compressed and uncompressed using a Lempel-Ziv-Welch process.

A management method and management apparatus that can prevent deterioration of response performance of an entire system is suggested.

A management method and apparatus for managing a storage apparatus to which a hierarchical pool configuration technique is applied is designed to: select a migration-candidate virtual volume(s) from among virtual volumes to which a page(s) is allocated from a pool constituted from a storage tier(s) whose number of accesses per unit time has reached a performance threshold and whose used capacity has not reached a capacity threshold; select a pool(s) from which a page(s) is allocated to the virtual volume, as a migration source pool; also select a pool(s) constituted from a storage tier whose number of accesses per unit time has not reached the performance threshold, as a migration destination pool; decide a migration candidate pair(s) by combining the migration-candidate virtual volume(s) with the migration destination pool(s); calculate response performance of the entire system before and after migration respectively; and decide a migration candidate pair which is a migration object, based on the calculation result.