706 results about "Memory pool" patented technology

The present invention discloses devices Including a transparent client-connection manager for exchanging client data between an application server and the device: a request analyzer for analyzing query requests from at least one application server; a data-retrieval / command-execution module for executing query requests; a database connection manager for exchanging database data between at least one database server and the device: a cache-memory pool for storing data items from at least one database server: a cache-policy module for determining cache criteria for storing the data Items In the cache-memory pool; and a data-consistency invalidation module for determining invalidated data Items based on invalidation criteria for removing from the cache-memory pool. The cache-memory pool Is configured to utilize memory modules residing in data proxy devices and distributed cache management utility, enabling the memory capacity to be used as a cluster to balance workloads.

A method and system for verifying computer system drivers such as kernel mode drivers. A driver verifier sets up tests for specified drivers and monitors the driver's behavior for selected violations that cause system crashes. In one test, the driver verifier allocates a driver's memory pool allocations from a special pool bounded by inaccessible memory space for testing the driver's accessing memory outside of the allocation. The driver verifier also marks the space as inaccessible when it is deallocated, detecting a driver that accesses deallocated space. The driver verifier may also provide extreme memory pressure on a specific driver, or randomly fail requests for pool memory. The driver verifier also checks call parameters for violations, performs checks to ensure a driver cleans up timers when deallocating memory and cleans up memory and other resources when unloaded. An I / O verifier is also described for verifying drivers use of I / O request packets.

The invention disclosed a multi-thread processor dynamic memory management system and method. The system includes a message treament module, a main control memory management module and a memory monitoring module. The method includes the steps: a main kernel thread applies to one-stage memory as a memory pool from the memory of system; all kernel threads tranmit the message for applying to the memory for main kernal thread, and main kernel thread distributes free memory in the memory pool to the kernel thread for applying to the memory after receiving the message for applying to the memory, and transmits the message for distributing the memory for the kernel thread for applying to the memory; main control kernel thread inquires the use of memory which has been applied by the kernel thread, and transmits the message of callbacking memory; the kernel thread returns the memory after receiving the medssage of callbacking memory, and transmits the message of releasing the memory for main control thread. The use of memory in the system ia able to be improved using the inventive multi-thread processor dynamic memory management system and method.

The invention discloses a method for realizing a local file system through an object storage system. A metadata cache algorithm of the file system and a memory description structure are adopted, so that the interactive frequency of an application and a background of a swift storage system is reduced and the performance of accessing the swift storage system by the application is improved; a policy of pre-allocating a memory pool and recovering idle memory blocks in batches in a delayed manner is adopted, so that the efficiency of traversing a directory comprising a large amount of subdirectories and files is improved; a memory description structure of an open file handle is adopted, so that the application can efficiently perform file reading-writing operation; a pre-reading policy is adopted, so that the frequency of network interaction between an application server and a swift storage back end is effectively reduced and the reading performance of the file system is improved; and a zero copying and block writing policy is adopted, so that no any data copying and caching exist in a file writing process, system call during each write is a complete block writing operation, and the file writing efficiency is improved.

The invention discloses a thread for high-performance computer NUMA perception and a memory resource optimizing method and system. The system comprises a runtime environment detection module used for detecting hardware resources and the number of parallel processes of a calculation node, a calculation resource distribution and management module used for distributing calculation resources for parallel processes and building the mapping between the parallel processes and the thread and a processor core and physical memory, a parallel programming interface, and a thread binding module which is used for providing the parallel programming interface, obtaining a binding position mask of the thread according to mapping relations and binding the executing thread to a corresponding CPU core. The invention further discloses a multi-thread memory manager for NUMA perception and a multi-thread memory management method of the multi-thread memory manager. The manager comprises a DSM memory management module and an SMP module memory pool which manage SMP modules which the MPI processes belong to and memory distributing and releasing in the single SMP module respectively, the system calling frequency of the memory operation can be reduced, the memory management performance is improved, remote site memory access behaviors of application programs are reduced, and the performance of the application programs is improved.

The invention discloses a data processing method and device used for improving the security of data. The method includes the steps that to-be-secured data sent by a target user is received; the first target Hash value of the to-be-secured data is calculated according to a preset Hash algorithm; a target transaction recording the first target Hash value is constructed; the target transaction is sent to a preset node in a target block chain; when information, in the target block chain, of the target transaction meets a first preset condition, the target transaction is sent to other nodes, except the preset node, in the block chain; when confirmation information sent by other nodes is received, the amount of the confirmation information is counted; when the amount of the confirmation information reaches a first preset amount, the state of the target transaction stored in a preset node memory pool is modified into a confirmed state. By the adoption of the scheme, multiple nodes can monitor the to-be-secured data jointly, and thus the security of the data is improved.

The invention provides a memory management method used in a Linux system. A large page memory is used in a Linux environment, and the memory configuration working process, the memory application process and memory release process are executed based on the large page memory; the memory configuration working process comprises the steps of calculating the relation between virtual addresses and physical addresses to determine the numa node a mapped hugepage belongs to and performing sorting according to the physical addresses; the memory application process comprises the steps of memory pool configuration application and common memory application. The memory management method keeps the high efficiency advantage of static memory allocation and adopts the 2MB hugepage to replace a 4KB page, the page query time is saved, and the TLB miss probability is reduced. In addition, a hugepage memory is not exchanged to a disk, it is ensured that the memory is always used by an applied application program, priority is given to use a local memory during memory application, and the memory access speed is improved.

The embodiment of the invention discloses a method and device for distributing memory. The method for distributing the memory comprises the following steps that a memory request of a target task is received; the capacity of the distributed memory is determined according to the capacity of requested memory; whether a free memory block meeting the capacity of the distributed memory exists in a first memory pool or not is determined; if yes, the memory block with the capacity of the distributed memory is distributed to the target task in the first memory pool; otherwise, the memory block with the capacity of the distributed memory is distributed to the target task in a preserved second memory pool under the condition that the priority of the target task is higher than a preset priority threshold value. According to the technical scheme, when the memory in the first memory pool is used up, the system can distribute the preserved memory to a critical task with higher priority, the basic management capacity and diagnosis capacity of the system are guaranteed, and the safety and reliability of the system are improved.