59results about How to "Reduce the number of migrations" patented technology

The invention discloses a hybrid main storage architecture based efficient dynamic page scheduling method applied to a memory controller of a horizontal storage architecture which comprises a first storage medium and a second storage medium. The hybrid main storage architecture based efficient dynamic page scheduling method comprises the steps of configuring an access list respectively in each storage medium; configuring a stack respectively for each storage medium in the memory controller; carrying out dynamic management fore recently accessed pages in the storage medium based on the page state records recorded in the stack and the reading-writing access time recorded in the access list, placing the pages with high writing hot-degree in the first storage medium, storing the frequently read pages in the second storage medium, wherein the first storage medium is a DRAM, and the second storage medium is NVM. The hybrid main storage architecture based efficient dynamic page scheduling method can master the historical access information more precisely to adapt to the application scenes with relatively weak locality, can greatly reduce frequent and ineffective page migration quantity between the hybrid memory mediums, and can improve the utilization rate and the access hit rate of a main memory.

The invention discloses a distributed storage-based dynamic metadata optimizing method and system. The method comprises the following steps: obtaining the load of mds (metadata server) nodes; obtaining the average load of mds clusters; calculating the processing power on the basis of the memory of the mds nodes and a CPU (Central Processing Unit) to obtain corresponding weighting coefficients; and judging whether the time when the load of the mds nodes is greater than the average load exceeds the threshold or not, and if yes, migrating data of the mds nodes of which the load is greater than the average load to the mds nodes of which the load is lower than the average load. The processing power is calculated on the basis of the memory of the mds nodes and the CPU to obtain the corresponding weighting coefficients, so that the data processing power of all the mds nodes is evaluated more accurately; regional division is carried out on the metadata server; when sub-trees are migrated, the load and the machine performance are comprehensively considered; the selection strategy of sub-tree migration targets is optimized on the basis of the data processing power and the current load of each mds node; and therefore, the network flow and the migration time are reduced, and the migration efficiency is increased.

The invention discloses a method for selecting a garbage collection target block, and a solid state disk. The method is characterized in that the recovery factor score of each block is calculated according to the proportion and the erasing frequencies of effective pages in each block, and the block with a lowest score is selected as a recovery target block. The effective page ratio and the erasingfrequency weighting score value of each block on the SSD (Solid State Disk) are calculated to comprehensively consider an influence on the read-write performance and the service life of the SSD as far as possible, and the block with a low effective page proportion and few erasing frequencies is selected as the target recovery block of a GC. A wear-leveling thought is used for selecting an optimalblock as a recovery block, block erasing frequencies in the SSD are reduced, data migration frequencies are lowered, write amplification is reduced, and the service life of the SSD is improved.

The embodiment of the invention discloses a metadata management method and device and a computer readable storage medium. The metadata management method comprises the steps of obtaining load values and heat values corresponding to all metadata servers respectively in a clustering system; selecting overload metadata servers and light load metadata servers according to a preset threshold value and the load values corresponding to all the metadata servers; calculating an average load value according to the load values corresponding to all the metadata servers; migrating a corresponding number ofcatalogue subtrees in the overload metadata server to the selected target metadata server according to the average load value, the preset threshold value, the load values and the heat values corresponding to the overload metadata servers and the load values and the heat values corresponding to the light load metadata servers. In the process, the heat conditions of all the metadata servers are fully considered, therefore the catalogue subtrees can be more reasonably migrated, the frequency of migrating the catalogue subtrees is effectively reduced, and the efficiency of load balance is improved.

The invention discloses an application deployment method in a mobile edge cloud, and the method combines fog computing with a mobile edge cloud network, and constructs a three-layer tree edge network structure consisting of a remote radio unit (RRU), a baseband processing unit (BBU) and a central controller. On the basis, an application deployment problem in the mobile edge cloud is converted into a binary integer programming problem, and an application deployment method for solving the problem is provided. According to the method, a descending optimal adaptation algorithm (BFD algorithm) is combined with a PRA algorithm, an application deployment mode capable of reducing the total cost is found, and the application is reasonably deployed in an edge server. In the aspect of the redeployment frequency of the application program, a heuristic redeployment method is provided, and the method utilizes a hit rate monitoring function to heuristic trigger redeployment operation, so that the system can reasonably determine the redeployment frequency, and meets the time-varying requirement.

The invention discloses a virtual machine placement method based on multi-objective optimization during cloud computing. Based on current information and forecast future information of resources demands of all virtual machines, new mapping is set up among the multiple virtual machines and multiple physical nodes. Therefore, service quality of all applications is ensured and stability of all the physical nodes is greatly enhanced in a novel virtual machine placement scheme. In the meantime, migration of a large number of virtual machines is avoided during the process of shifting from an old placement state to a new placement state for virtual machines. By adoption of the scheme, quality of service is ensured when the multiple virtual machines are replaced. The migration frequency of the virtual machines is low. The long stabilization time of all the physical nodes is ensured in the meantime.

The invention discloses a container dynamic migration method and system based on a minimum migration quantity and designs a dynamic container migration model. The invention aims to provide a containerdynamic migration optimization method based on the minimum migration quantity. According to the method, by fitting increase speeds of containers in a source server, the containers for dynamic migration are selected according to the principle of selecting the containers with high migration increase speeds and high occupation rates. In this way, the migration quantity is reduced under the conditionof guaranteeing cluster load balance, and therefore the utilization rate of resources is increased.

The invention relates to a virtual machine energy saving migration method oriented to cloud computing. Compared with an existing mainstream virtual machine migration method, the method disclosed by the invention has a small influence on service performance since a virtual machine selection algorithm of minimum migration time is adopted when a small-capacity server is overloaded, and service performance loss generated by virtual machine migration and server overload can be maintained at a low level; when a large-capacity server is overloaded, minimum virtual machine migration algorithms are adopted to select the virtual machine to be migrated, the overload problem of the server is quickly solved, virtual machine migration frequencies are reduced, and the migration efficiency of the virtualmachine can be improved.

An embodiment of the invention provides a cached data migration method and device and aims to solve the problem that frequent data migration in the prior distributed caching service cluster uses greater network bandwidth, resulting in decreased caching service performance. The method includes: after determining that a corresponding caching server fails and is offline, back-upping the caching server in a back-up server as back-up data, serving as data of a client visiting the caching server; if the caching server which fails and is offline is online within a set time, restoring the back-up data to the caching server. After the caching server fails and is offline and is online again within the set time, the corresponding back-up server only restores the back-up data to the caching server and avoids data migration; therefore, data migrations are fewer, the network bandwidth is never excessively used, and the performance of the caching service is improved.

The invention relates to a cloud computing platform virtual machine dispatching method based on operation duration. The cloud computing platform virtual machine dispatching method can reduce SLA violation times, empty a low-load physical computer and switch the physical computer to be in a low energy consumption state, increase CPU utilization rate of the physical computer, and can achieve the beneficial effects of low energy consumption and high energy efficiency by reducing migration times of a virtual machine and avoiding unreasonable effect of virtual machine migration.

The invention discloses an energy-saving control method and device for (optical line terminal) ports in a TWDM (time-wavelength division multiplexing) PON (passive optical network) and belongs to the optical access network field. The energy-saving control method includes the following steps that: whether an OLT port can be selected from an OLT system to enter a sleep mode is judged according to the total service load of the OLT system; when it is judged that the OLT port can be selected to enter the sleep mode, the service load condition of each OLT port is put into statistics, and an OLT port with minimum service load is selected as an OLT port requiring sleep; and an ONU (optical network unit) under the OLT port is migrated to another OLT port, and then, the OLT port is controlled to enter the sleep mode. According to the energy-saving control method and device of the present invention, based on the actual demands of existing network service operation, the OLT port is selected to enter the sleep mode according to the magnitude of the load of the OLT port, and therefore, the scale of service migration can be decreased.

Disclosed are a state transition method and apparatus based on robust header compression (ROHC), and a computer storage medium. The method comprises: determining that a state machine of a compressor is in an initial IR state, and when it is learned that a decompressor corresponding to the compressor can only successfully parse a static field of a service stream packet header, transiting the state of the state machine of the compressor from the IR state to a first order FO state; and when it is learned that the decompressor corresponding to the compressor can successfully parse the static field and a dynamic field of the service stream packet header, transiting the state of the state machine of the compressor from the IR state to a second order SO state; determining that the state machine of the compressor is in the FO state, and when it is learned that the decompressor corresponding to the compressor can successfully parse the dynamic field of the service stream packet header, transiting the state of the state machine of the compressor from the FO state to the SO state; and determining that the state machine of the compressor is in the SO state, and when it is learned that the decompressor corresponding to the compressor fails to parse the dynamic field of the service stream packet header within a preset time threshold, transiting the state of the state machine of the compressor from the SO state to the FO state.

The invention discloses a dynamic group system for layered rollback recovery protocols based on MPI (Message Passing Interface) high performance computing. The system comprises a message monitoring module, a message analysis module and a process migration module, and belongs to the fields of high performance computing and system fault tolerance. The message monitoring module is used for monitoring a message passing record between processes in an MPI high performance computing application, storing the record in a format of a triad (source process, target process and message size) and submitting the message passing record to the message analysis module; the message analysis module is used for analyzing the message passing record collected by the message monitoring module, analyzing to obtain a message passing mode of the current application as a basis for the application to execute the process migration module at the next time while utilizing the last message passing mode to judge whether to execute the process migration module at this time; the process migration module is used for carrying out migration operation on the changed process when the message passing mode of the application changes, so that the performance optimization of the layered rollback recovery protocols is realized.

The invention discloses a process migration method, a computing device and a storage medium, and the process migration method comprises the following steps of calculating the predicted running time of each process in a processor according to the running information of the processes; according to the predicted running time of the processes, determining the process with the predicted running time smaller than a running time threshold value as a short running process; acquiring the running information of the processor, and determining whether the short running process is used as a to-be-migrated process according to the running information of the processor; if the short running process is the to-be-migrated process, determining a target processor for migrating the to-be-migrated process according to the running information of other processors; and migrating the to-be-migrated process to the target processor for operation. According to the present invention, the migration of all the short running processes can be avoided, the migration frequency of the short running processes is reduced, the problem of invalidation of cache and memory affinity caused by migration is reduced, and it is guaranteed that when the short running processes are migrated, the system performance can be improved, and the use efficiency of the processor is improved.

The embodiment of the invention relates to the technical field of document management, and discloses a document processing method, a document processing system and a Notes server. The method includes: allowing the Notes server to acquire Notes document data; performing standardization conversion on the Notes document data according to a preset data standardization rule, and acquiring a standard document; and receiving an access request of an access client, and sending the standard document to the access client in response to the access request. According to the method, direct interaction between the Notes system and the access client can be achieved, an involvement process of addition servers and relation type databases can be avoided in the prior art, the data migration frequency is lowered, the document processing efficiency is improved, and the data loss risk in a migration process can be lowered.

The invention discloses a server integration method based on a virtual machine. The server integration method based on the virtual machine comprises a server local integration algorithm based on a dynamic threshold value and a server global integration algorithm based on a genetic algorithm. The method is scientific and reasonable, is safe and convenient to use, and can effectively realize an important function of server integration according to the load change dynamic state of the cloud data center through local integration and global integration, thereby improving the service support capability and the operation and maintenance efficiency, reducing the investment and decision-making risk, and achieving the purposes of saving the investment and saving the energy. The objective of the provided server local integration algorithm is to carry out local integration in a small range on servers in a high-load state and a low-load state, and the energy-saving requirement is met. The objectiveof the server global integration algorithm is to integrate the servers in a larger range, and the virtual machines run on the servers as few as possible while the service quality is ensured, so thatthe overall resource utilization rate is improved, and the energy consumption is reduced.

The embodiment of the invention provides a method for processing cloud platform resource fragments and related equipment, and belongs to the technical field of computers and communication. The cloud platform comprises a plurality of nodes, and each node comprises a first resource and a second resource; the method comprises the following steps: acquiring a current first resource allocation rate anda current second resource allocation rate of each node; according to the current first resource allocation rate and the current second resource allocation rate of each node, determining that each node is currently in one of a resource balance state, a first resource imbalance state and a second resource imbalance state; and migrating the stock hosts causing the first resource allocation imbalancein the nodes in the first resource imbalance state to the nodes in the second resource imbalance state or the resource balance state. According to the technical scheme, the method for processing theresource fragments of the cloud platform is provided, the resource fragments of the cloud platform can be effectively reduced, and the resource utilization rate is increased.

The invention discloses a cloud computing virtual machine placement method based on resource use curve complementation. A main idea is characterized by picking out an overload physical host, selectinga virtual machine on the overload physical machine; predicting the resource use condition of the system at t time points in the future; predicting the resource use condition of the physical host formigration at t time points in the future; and searching a physical host to enable the resource use condition of the physical host to be complementary to the resource use condition of the virtual machine; calculating the complementation degree between the physical host and the virtual machine, calculating the complementation degree between all the virtual machines on the overloaded physical host and all the migration-available physical hosts according to the process, selecting a group of virtual machines and physical hosts with the maximum complementation degree values, and migrating the virtual machines to the physical host, and if the physical machine is still overloaded after migration, removing the corresponding comprehensive complementarity from the comprehensive complementarity set and reselecting a group of virtual machines to migrate with the physical host until the physical machine is not overloaded or the comprehensive complementarity set is empty.

The invention discloses a method and a device for planning the itinerary of a mobile agent, which relates to the technical field of wireless sensor network, reduces the energy consumption of the WSN (Wireless Sensor Network) network and prolongs the service life of the WSN network. The method comprises the following steps: if the accumulated value of the information gain is not larger than the threshold value of the information gain, determining the candidate position of a target object at the next time; judging whether the distance between the position of the node of the host sensor in which the mobile agent is located at present and the candidate position is larger than a first threshold value; if the distance is not larger than the first threshold value, un-moving the mobile agent; and if the distance is larger than the first threshold value, selecting the node of a target sensor from the nodes of the neighbors sensors of the host sensor in which the mobile agent is located, and moving the mobile agent to the node of the target sensor. The embodiment of the invention is mainly applied to the process of planning the itinerary of the mobile agent in target tracking scenes.

Disclosed is a method that includes collecting first temporal statistics for a port element in a computing environment, collecting second temporal statistics for a switch element in the computing environment, collecting third temporal statistics for the computing environment generally, computing a spatial correlation between network features and network elements comprising the port element and theswitch element and computing, via a machine learning technique, a port dynamic weight for the port element and a switch dynamic weight for the switch element. The method can also include scheduling workload to consume compute resources within the compute environment based at least in part on the port dynamic weight for the port element and the switch dynamic weight for the switch element.

The invention discloses a virtual machine migration method and a virtual machine migration system, relates to the field of cloud calculation and solves problems the an existing virtual machine resource scheduling mechanism is unable to satisfy variability of actual operation status of the system and low in efficiency. The method includes: determining virtual machine filling priority of host equipment in a cluster; migrating at least one virtual machine on the host equipment low in virtual machine filling priority to the host equipment with the load not up to the limit and high in the virtual machine filling priority. The virtual machine migration method and the virtual machine migration system are applicable to the large-scale virtual machine cluster, and dynamic adjustment of virtual machine resource scheduling is realized.

The invention provides a cloud computing virtual machine migration method based on load curve similarity. obtaining residual resource utilization rate curves of all physical machines and resource utilization rate curves of all virtual machines to be migrated, according to a Pearson correlation coefficient formula, a similarity vector composed of the similarity of each type of resources is obtained, and finally the Euclidean distance is obtained according to the similarity vector and the average resource utilization rate so as to determine the virtual machine and the target physical machine needing to be migrated. According to the invention, the load of the overload physical machine within a period of time is taken as the standard of migration instead of paying close attention to the load condition at the moment of overload, so that the target physical machine is effectively prevented from overload within a period of time after migration, the migration frequency is reduced, and the energy consumption is reduced. Meanwhile, the similarity of the curves is used as one of migration standards, so that the virtual machine to be migrated and the target physical machine are complementary as much as possible, and the load of the target physical machine is more balanced.

The invention discloses virtual machine energy-saving integration method and system based on hybrid swarm intelligence and storage medium. The method comprises the following steps of: collecting resource occupation information of a data center server, resource request information of a virtual machine and mapping information between the server and the virtual machine by a resource monitor; detecting an overload server according to the CPU utilization rate under the peak performance ratios of the servers of different models; according to the migration value ratio, selecting the virtual machine to be migrated from the overload server for migration; re-setting the virtual machine to be migrated by using a hybrid differential evolution particle swarm optimization algorithm; and according to the CPU load mean value of the cluster, selecting an under-load server to turn off, and replacing the virtual machine in the under-load server by using a hybrid differential particle swarm optimization algorithm. According to the virtual machine integration method, the real-time power consumption of the cluster can be effectively reduced through multi-stage virtual machine scheduling, the number of active servers of the data center and the migration frequency of the virtual machines are reduced, and the service quality of the data center is guaranteed.