Abstract: A method and apparatus are disclosed for enabling nodes in a distributed system to share one or more memory portions. A home node makes a portion of its main memory available for sharing, and one or more sharer nodes mirrors that shared portion of the home node's main memory in its own main memory. To maintain memory coherency, a memory coherence protocol is implemented. Under this protocol, load and store instructions that target the mirrored memory portion of a sharer node are trapped, and store instructions that target the shared memory portion of a home node are trapped. With this protocol, valid data is obtained from the home node and updates are propagated to the home node. Thus, no “dirty” data is transferred between sharer nodes. As a result, the failure of one node will not cause the failure of another node or the failure of the entire system.

Abstract: Embodiments relate to enhancing a refresh PCI translation (RPCIT) instruction to refresh a translation lookaside buffer (TLB). A computer processor determines a request to purge a translation for a single frame of the TLB in response to executing an enhanced RPCIT instruction. The enhanced RPCIT instruction is configured to selectively perform one of a single-frame TLB refresh operation or a range-bounded TLB refresh operation. The computer processor determines an absolute storage frame based on a translation of a PCI virtual address in response to the request to purge a translation for a single frame of the TLB. The computer processor further performs the single-frame TLB refresh operation to purge the translation for the single frame.

Abstract: The disclosure of the present invention presents a method and system for efficiently maintaining an object cache to a maximum size by number of entries, whilst providing a means of automatically removing cache entries when the cache attempts to grow beyond its maximum size. The method for choosing which entries should be removed provides for a balance between least recently used and least frequently used policies. A flush operation is invoked only when the cache size grows beyond the maximum size and removes a fixed percentage of entries in one pass.

Abstract: According to one embodiment, a method of creating compiler-generated memory mapping hints in a computer system includes analyzing code, by a compiler of the computer system, to identify data access patterns in the code. System configuration information defining data processing system characteristics of a target system for the code is accessed. The data processing system characteristics include a plurality of processing resources and memory domain characteristics relative to the processing resources. A preferred allocation of data in memory domains of the target system is determined based on mapping the code to one or more selected processing resources and mapping the data to one or more of the memory domains based on the memory domain characteristics relative to the one or more selected processing resources. The preferred allocation is stored as compiler-generated memory mapping hints in a format accessible by a physical memory mapping resource of the target system.

Abstract: A flash memory management method and apparatus provides for the separation of the command and data paths so that communication paths may be used more efficiently, taking account of the characteristics of NAND FLASH circuits where the times to read, write and erase data differ substantially. A unique sequence identifier is assigned to a write command and associated data and association of the data and commands are validated prior to writing to the memory by comparing the unique sequence numbers of the data and command prior to executing the command. This comparison is performed after the data and command have traversed the communication paths.

Abstract: Limiting the execution of background management operations in a drive array, including: receiving a read instruction to read data from a memory drive in the drive array; determining whether the read instruction is associated with a write instruction to write data to a memory drive in the drive array; responsive to determining that the read instruction is associated with the write instruction, restricting performance of background management operations on the memory drive targeted by the write instruction; determining whether the write instruction has completed; and responsive to determining that the write instruction has completed, removing restrictions associated with the performance of background management operations on the memory drive targeted by the write instruction.

Abstract: An aspect includes coherency management between volatile memory and non-volatile memory in a through-silicon via (TSV) module of a computer system. A plurality of TSV write signals is simultaneously provided to the volatile memory and the non-volatile memory. A plurality of values of the TSV write signals is captured within a buffer of the non-volatile memory corresponding to a data set written to the volatile memory. Storage space is freed within the buffer as the data set corresponding to the values of the TSV write signals stored within the buffer is written to a non-volatile memory array within the non-volatile memory.

Abstract: An aspect includes data backup management between volatile memory and non-volatile memory in a through-silicon via module of a computer system. Data is copied data from the volatile memory to the non-volatile memory during a refresh cycle of the volatile memory. The data is written to one or more non-volatile memory cells within the non-volatile memory prior to a next refresh cycle of the volatile memory.

Abstract: An apparatus for use in telecommunications system comprises a cache memory shared by multiple clients and a controller for controlling the shared cache memory. A method of controlling the cache operation in a shared cache memory apparatus is also disclosed. The apparatus comprises a cache memory accessible by a plurality of clients and a controller configured to allocate cache lines of the cache memory to each client according to a line configuration. The line configuration comprises, for each client, a maximum allocation of cache lines that each client is permitted to access. The controller is configured to, in response to a memory request from one of the plurality of clients that has reached its maximum allocation of cache lines, allocate a replacement cache line to the client from cache lines already allocated to the client when no free cache lines in the cache are available.

Abstract: A memory system and method for differential thermal throttling are disclosed. In one embodiment, a memory system is provided comprising a memory and a controller. The controller is configured to receive a command to perform an operation in the memory and analyze the command to determine whether thermal throttling the memory system would result in an unacceptable impact on user experience. In response to determining that thermal throttling the memory system would result in an unacceptable impact on user experience, the controller executes the command. In response to determining that thermal throttling the memory system would not result in an unacceptable impact on user experience, the controller thermal throttles the memory system. Other embodiments are provided.

Abstract: A method includes reading, by a processor, one or more configuration values from a storage device or a memory management unit. The method also includes loading the one or more configuration values into one or more registers of the processor. The one or more registers are useable by the processor to perform address translation.

Abstract: A data storage system implements out-of-order data transfer. In one embodiment, the data storage system can retrieve from a host system a scatter gather list (SGL) associated with a data read command and generate a memory access table based on the retrieved SGL. The data storage system can further retrieve data from memory, and at least some data may be retrieved out of order. Retrieved data can be provided to the host system using the memory access table, and at least some data may be provided out of order. Data retrieval performance can be increased.

Abstract: This disclosure includes a method for securing a memory of an electronic system that includes initializing the memory, creating a security key, transmitting the security key to memory, storing the security key in the memory, transmitting the current security key and a a new security key to the memory by the memory controller. If the current security key transmitted is the same as the security key stored in memory, then access to the memory is enabled and the current security key in the memory is replaced with the new security key. If the current security key transmitted is not the same as the security key stored in the memory, then access to the memory is disabled.

Abstract: This disclosure describes techniques for supporting inter-task communication in a parallel computing system. The techniques for supporting inter-task communication may use hardware-based atomic operations to maintain the state of a pipe. A pipe may refer to a First-In, First-Out (FIFO)-organized buffer that allows various tasks to interact with the buffer as data producers or data consumers. Various pipe implementations may use multiple state parameters to define the state of a pipe. The hardware-based atomic operations described in this disclosure may modify multiple pipe state parameters in an atomic fashion. Modifying multiple pipe state parameters in an atomic fashion may avoid race conditions that would otherwise occur when multiple producers and/or multiple consumers attempt to modify the state of a pipe at the same time. In this way, pipe-based inter-task communication may be supported in a parallel computing system.

Abstract: An aggregation module combines a plurality of logical address spaces to form a conglomerated address space. The logical address spaces comprising the conglomerated address space may correspond to different respective storage modules and/or storage devices. An atomic aggregation module coordinates atomic storage operations within the conglomerated address space, and which span multiple storage modules. The aggregation module may identify the storage modules used to implement the atomic storage request, assign a sequence indicator to the atomic storage request, and issue atomic storage requests (sub-requests) to the storage modules. The storage modules may be configured to store a completion tag comprising the sequence indicator upon completing the sub-requests issued thereto. The aggregation module may identify incomplete atomic storage requests based on the completion information stored on the storage modules.

Abstract: A copy technique involves generating, upon receiving a command to copy at least a portion of a source file, a set of pointers for a destination file that point to a set of pointers for the source file (“physical block pointers”), which in turn point to physical blocks storing data of the file in a storage system. In response to the copy command, a storage processor allocates an inode from an inode table of a file system for the destination file and stores a set of inode pointers in the inode of the destination file that point to the set of pointers for the source file. Thus, instead of creating a redundant copy of the data of the source file, the improved creates technique creates pointers that point to the data stored for the source file.

Abstract: A method of enhancing a refresh PCI translation (RPCIT) operation to refresh a translation lookaside buffer (TLB) includes determining, by a computer processor, a request to perform at least one RPCIT instruction for purging at least one translation from the TLB. The method further includes purging, by the computer processor, the at least one translation from the TLB in response to executing the at least one RPCIT instruction. The computer processor selectively performs a synchronization operation prior to completing the at least one RPCIT instruction.

Abstract: Embodiments relate to enhancing a refresh PCI translation (RPCIT) instruction to refresh a translation lookaside buffer (TLB). A computer processor determines a request to purge a translation for a single frame of the TLB in response to executing an enhanced RPCIT instruction. The enhanced RPCIT instruction is configured to selectively perform one of a single-frame TLB refresh operation or a range-bounded TLB refresh operation. The computer processor determines an absolute storage frame based on a translation of a PCI virtual address in response to the request to purge a translation for a single frame of the TLB. The computer processer further performs the single-frame TLB refresh operation to purge the translation for the single frame.

Abstract: A method of enhancing a refresh PCI translation (RPCIT) operation to refresh a translation lookaside buffer (TLB) includes determining, by a computer processor, a request to perform at least one RPCIT instruction for purging at least one translation from the TLB. The method further includes purging, by the computer processor, the at least one translation from the TLB in response to executing the at least one RPCIT instruction. The computer processor selectively performs a synchronization operation prior to completing the at least one RPCIT instruction.

Abstract: A system, processor, and method to record the interleavings of shared memory accesses in the presence of complex multi-operation instructions. An extension to instruction atomicity (IA) is disclosed that makes it possible for software to infer partial information about a multi-operation execution if the hardware has recorded a dependency due to an instruction atomicity violation (IAV). By monitoring the progress of a multi-operation instruction, the need for complex multi-operation emulation is unnecessary.