703 results about "Memory pool" patented technology

A method of building a computer system is provided. The method comprises determining an average memory storage per virtual machine, determining an average number of virtual machines per host computer, and determining an amount of memory storage per memory pool. The method also comprises determining a maximum number of host computers per memory pool based on the average memory storage per virtual machine, the average number of virtual machines per host computer, and the amount of memory storage per memory pool. The method also includes assembling the appropriate number of host computers and memory storage, organized around memory pools, to handle a specific number of virtual machines.

A method, article of manufacture and apparatus for automatically predicting out of memory exceptions in garbage collected environments are disclosed. One embodiment provides a method of predicting out of memory events that includes monitoring an amount of memory available from a memory pool during a plurality of garbage collection cycles. A memory usage profile may be generated on the basis of the monitored amount of memory available, and then used to predict whether an out of memory exception is likely to occur.

An ATM switching system architecture of a switch fabric-type is built of, a plurality of ATM switch element circuits and routing table circuits for each physical connection to/from the switch fabric. A shared pool of memory is employed to eliminate the need to provide memory at every crosspoint. Each routing table maintains a marked interrupt linked list for storing information about which ones of its virtual channels are experiencing congestion. This linked list is available to a processor in the external workstation to alert the processor when a congestion condition exists in one of the virtual channels. The switch element circuit typically has up to eight 4-bit-wide nibble inputs and eight 4-bit-wide nibble outputs and is capable of connecting cells received at any of its inputs to any of its outputs, based on the information in a routing tag uniquely associated with each cell.

In a multi-QOS level queuing structure, packet payload pointers are stored in multiple queues and packet payloads in a common memory pool. Algorithms control the drop probability of packets entering the queuing structure. Instantaneous drop probabilities are obtained by comparing measured instantaneous queue size with calculated minimum and maximum queue sizes. Non-utilized common memory space is allocated simultaneously to all queues. Time averaged drop probabilities follow a traditional Weighted Random Early Discard mechanism. Algorithms are adapted to a multi-level QOS structure, floating point format, and hardware implementation. Packet flow from a router egress queuing structure into a single egress port tributary is controlled by an arbitration algorithm using a rate metering mechanism. The queuing structure is replicated for each egress tributary in the router system.

The invention provides a memory control method for embedded systems. The method includes: applying for memory in an operating system, and using part of the memory as a memory pool and the other part as a reserved memory area; establishing a cache for each thread by pool management of the memory pool and thread caching technology; subjecting the reserved memory area to management by TLSF (two-level segregated fit) algorithm; dividing the memory pool into memory blocks different in size, connecting the memory blocks in the same size to form a double linked list, adding a memory management unit for each memory block, and placing the memory management unit and the memory blocks in different memory areas respectively; and establishing a memory statistical linked list for each thread to connect the memory blocks applied by the thread so as to facilitate troubleshooting of memory leak. In addition, a memory coverage checking mechanism is added without increasing cost of the memory control method.

FIFOs may be implemented in shared memory using linked lists and interleaved linked lists such that any individual FIFO can dynamically use any free memory location. The system may be implemented in hardware efficiently and does not pre-allocate memory to any FIFO, so that the whole shared memory may be used by any one of the FIFOs. The maximum physical size of a FIFO is then limited only by the size of the physical memory and not by the number of FIFOs sharing the same physical memory or by any mechanism for pre-allocating physical memory to the FIFOs. The linked lists contain data words, each data word containing a data field and a reference to a subsequent free memory location. A subsequent data word may be enqueued to the FIFO at the last referenced free memory location. A dequeue operation returns free memory locations to the free memory pool. Multiple linked lists may be interleaved to accelerate reading of data words from the FIFO.

The invention discloses a memory management method and a device for an embedded type system. The method of the invention comprises the following steps: firstly, a memory is divided into a plurality of large memory zones with unequal memory capacities; secondly, when an application program begins to run, a large memory zone which is matched with a maximum memory requirement value of the application program is allocated for the application program; thirdly, a small memory is allocated for the application program in the matched large memory zone by utilization of heap means; fourthly, when the small memory is released by the application program by utilization of heap means, the small memory is combined with adjacent and free small memories; when the application program stops running, the matched large memory zone is released. The invention avoids a large amount of memory fragmentations, quickens speed of memory fragmentation defrag and avoids the necessity in a memory pool means of calculation of the size and number of each memory which needs for dynamic allocation of each application program.

The invention discloses a block storage method and a block storage device in a blockchain. The block storage method in the blockchain is implemented by a miner node, and comprises the steps of: packing transaction information in a memory pool into a block; broadcasting first block header information in the block to a blockchain network, wherein the first block header information comprises a block transaction serial number hash value; generating a first hash table corresponding to the first block header information according to the first block header information in the block and transaction serial number information; generating a second hash table corresponding to the first block header information according to the transaction serial number information in the block and transaction content information; and uploading the first hash table and the second hash table corresponding to the first block header information to a distributed storage network. By utilizing the block storage method and the block storage device provided by the invention, the pressure of network transmission and storage pressure of the single node are effectively alleviated.

Methods and systems for estimating the hypothetical performance of a messaging application are disclosed. A number of pool sizes may be identified, each pool size being a potential size for the memory allocated to the messaging application. An online simulation is running during the execution of the messaging application. The online simulation tracks the requests made by the messaging application and predicts the operation of the messaging application for each pool size. The data predicted includes the number of spill and unspill operations that read and write to disk. In addition, a method for calculating the age of the oldest message in a memory pool is disclosed. The age is used in determining the number of spill and unspill operations.

A switched architecture is provided to allow controllers to manage physically independent memory systems as a single, large memory system. The switched architecture includes a path between switches of controllers for inter-controller access to memory systems and input/output interfaces in a redundant controller environment. Controller memory systems are physically independent of each other; however, they are logically managed as a single, large memory pool. Cache coherency is concurrently maintained by both controllers through a shared locking mechanism. Volume Logical Block Address extents or individual cache blocks can be locked for either shared or exclusive access by either controller. There is no strict ownership model to determine data access. Access is managed by the controller in the pair that receives the access request. When a controller is removed or fails, a surviving controller may take appropriate action to invalidate all cache data that physically resides in the failed or missing controller's memory systems. Cached write data may be mirrored between redundant controllers to prevent a single point of failure with respect to unwritten cached write data.

The present invention provides an application virtualization framework that allows dynamic allocation and de-allocation of compute resources, such as memory. Runtime applications are pooled across multiple application servers and compute resources are allocated and de-allocated in such a way that resource utilization is optimized. In addition, objects are either stored in local memory or in a non-local memory pool depending on certain criteria observed at runtime. Decision logic uses decision rules to determine whether an object is to be stored locally or in a memory pool. Resource management logic monitors the memory pool to determine which locations are available for storing objects and how much memory is available overall in the memory pool for storing objects.

A method for raising search speed of information includes generating Haxi value by key word in searching information, judging whether corresponding search result is buffer-stored or not by utilizing Haxi value to check Haxi table, obtaining array element index used for identifying storage unit in array internal memory cell from Haxi table and fetching search result information identified storage unit of said cell then feeding it back to request party if it is or otherwise using search engine to obtain search result through key word and setting up mapping relation of Haxi value to index value then storing said result information in identified storage unit of said cell and feeding it back to request party.

The invention discloses a memory management method. The method comprises the following steps of: creating a memory pool comprising a plurality of memory groups, wherein the memory blocks in the same memory group are the same in size and idle memory blocks are sequentially linked into an idle memory linked list, and memory blocks in different memory groups are different in size; selecting a first memory group according to the size of a requested memory when a memory allocation request is received, and allocating a first memory block in the idle memory linked list of the selected first memory group to a memory requester; and selecting a second memory group according to the address of a memory required to be released when a memory release request is received, and linking the memory block which is required to be released and is taken as the first memory block into the idle memory linked list of the second memory group. The embodiment of the invention also provides a corresponding memory management system. In the method disclosed by the invention, the memory is allocated or released in the corresponding memory group, thus no memory fragment is produced and the memory allocation efficiency can be improved.

The invention relates to a method for allocating internal memory of real time operating system, which comprises the following steps: applying a big internal memory area from operating system, dividing the big internal memory area into internal memory polls with different sizes and the same internal memory poll comprises internal memory block with the same size, initialing each internal memory poll controlling bar. When it needs internal memory block, it finds the corresponding internal memory poll according to the size of internal memory block, and quotes weather the internal memory poll has free block, if it has free block, it fetches a block from the head of the team and modifies the referent information of internal memory poll controlling bar, and then the allocation finishes; if not, it dose dynamic adjustment to internal memory; when it releases the internal memory, it finds the corresponding internal memory poll according to the size of internal memory and returns it into the tail of the team and rewrites the referent information of internal memory poll controlling bar.

A method of scheduling jobs in real time using a database management system is provided. An application task classifies jobs as any one transaction type of a hot type and a normal type. A processing area in a memory pool that is a common resource is allocated to the application task, and the job is transferred to a database job manager through a client application program interface (API). The job manager loads a request node of the job in a list of a transaction type corresponding to the job, of a mailbox of the DB task, which classifies job request nodes as the hot type and the normal type with respect to the type of transaction and manages the nodes, so that the job request node can be scheduled in units of transactions. The job manager transfers the job request nodes loaded in the mailbox for the DB task, one by one to the DB task so that the job request nodes can be processed in units of transactions in a manner in which a job corresponding to a hot-type transaction is processed with a priority over a job corresponding to a normal-type transaction, and between jobs of an identical truncation type, jobs are processed in order of job requesting time. The DB task loads through the job manager, the processing result of the job in a mailbox of the application task which requested the processing of the job, so that the corresponding application task can use the processing result in the future.

Hypervisor managed memory paging is provided in a data processing system having multiple logical partitions. The data processing system includes a shared memory pool defined within physical memory. The shared memory pool includes a volume of physical memory with dynamically adjustable sub-volumes or sets of physical pages associated with the multiple logical partitions. Each sub-volume or set is associated with a particular logical partition and includes mapped logical memory pages for that logical partition. A hypervisor memory manager interfaces the multiple logical partitions and the shared memory pool, and manages access to logical memory pages within the shared memory pool. The hypervisor memory manager further manages page-out and page-in of logical memory pages from the shared memory pool to one or more external paging devices. This page-out and page-in managing by the hypervisor memory manager is transparent to the multiple logical partitions.

Mechanisms to implement memory management to enable protocol-aware asynchronous, zero-copy transmits. A transport protocol engine exposes interfaces via which memory buffers from a memory pool in operating system (OS) kernel space may be allocated to applications running in an OS user layer. The memory buffers may be used to store data that is to be transferred to a network destination using a zero-copy transmit mechanism, wherein the data is directly transmitted from the memory buffers to the network via a network interface controller. The transport protocol engine also exposes a buffer reuse API to the user layer to enable applications to obtain buffer availability information maintained by the protocol engine. In view of the buffer availability information, the application may adjust its data transfer rate.

A memory management service occupies a configurable portion of an overall memory system in a disaggregate compute environment. The service provides optimized data organization capabilities over the pool of real memory accessible to the system. The service enables various types of data stores to be implemented in hardware, including at a data structure level. Storage capacity conservation is enabled through the creation and management of high-performance, re-usable data structure implementations across the memory pool, and then using analytics (e.g., multi-tenant similarity and duplicate detection) to determine when data organizations should be used. The service also may re-align memory to different data structures that may be more efficient given data usage and distribution patterns. The service also advantageously manages automated backups efficiently.

A method, system, and program for dynamic memory management of unallocated memory in a logical partitioned data processing system. A logical partitioned data processing system typically includes multiple memory units, processors, I/O adapters, and other resources enabled for allocation to multiple logical partitions. A partition manager operating within the data processing system manages allocation of the resources to each logical partition. In particular, the partition manager manages allocation of a first portion of the multiple memory units to at least one logical partition. In addition, the partition manager manages a memory pool of unallocated memory from among the multiple memory units. Responsive to receiving a request for a memory loan from one of the allocated logical partitions, a second selection of memory units from the memory pool is loaned to the requesting logical partition. The partition manager, however, is enabled to reclaim the loaned selection of memory units from the requesting logical partition at any time.

A distributed memory switch system for transmitting packets from source ports to destination ports, comprising: a plurality of ports including a source port and a destination port wherein a packet is transmitted from the source port to the destination port; a memory pool; and an interconnection stage coupled between the plurality of ports and the memory pool such that the interconnection stage permits a packet to be transmitted from the source port to the destination port via the memory pool.

The invention discloses a management method of memory operation, and a thread level memory pool maintained by a thread is arranged. The method further comprises the following steps: a business module applies to the thread when needing to use a memory block; the thread calculates the size of the allocated memory block, and searches whether the memory block with needed size exists in the self thread level memory pool; if exists, the thread allocates the memory block with corresponding size from the thread level memory pool to the business module and completes the memory application flow; otherwise, the thread is locked and applies for a certain amount of memory blocks with the needed size to a process; or the memory block with the size of one level bigger than the needed size is allocated to the business module and the memory application flow is completed. The invention further discloses a memory operation management system comprising a first-level memory management unit, a second-level memory management unit and a third-level memory management unit. With the invention, the running efficiency of programs can be improved.

Allocation of memory is optimized across multiple pools of memory, based on minimizing the time it takes to successfully retrieve a given data item from each of the multiple pools. First data is generated that indicates a hit rate per pool size for each of multiple memory pools. In an embodiment, the generating step includes continuously monitoring attempts to access, or retrieve a data item from, each of the memory pools. The first data is converted to second data that accounts for a cost of a miss with respect to each of the memory pools. In an embodiment, the second data accounts for the cost of a miss in terms of time. How much of the memory to allocate to each of the memory pools is determined, based on the second data. In an embodiment, the steps of converting and determining are automatically performed, on a periodic basis.