scispace - formally typeset
Search or ask a question
Topic

Memory management

About: Memory management is a research topic. Over the lifetime, 16743 publications have been published within this topic receiving 312028 citations. The topic is also known as: memory allocation.


Papers
More filters
Patent
02 Feb 2009
Abstract: Memory mapping techniques for non-volatile memory are disclosed where logical sectors are mapped into physical pages using data structures in volatile and non-volatile memory. In some implementations, a first lookup table in non-volatile memory maps logical sectors directly into physical pages. A second lookup table in volatile memory holds the physical address of the first lookup table in non-volatile memory. In some implementations, a cache in volatile memory holds the physical addresses of the most recently written logical sectors. Also disclosed is a block TOC describing block content which can be used for garbage collection and restore operations.

87 citations

Patent
16 Nov 1994
TL;DR: An integrated memory controller (IMC) as discussed by the authors is a single logical unit that combines memory, graphics, and audio processing capabilities in a common logical unit. The IMC includes a pointer-based display list architecture which includes windows workspace areas spaces which define the format of the data and the data type to read or write.
Abstract: An integrated memory controller (IMC) which incorporates novel memory, graphics, and audio processing capabilities in a single logical unit. The IMC includes numerous significant advances which provide greatly increased performance over prior art systems. The integrated memory controller (IMC) includes one or more symmetric memory ports for connecting to system memory. The IMC also includes video outputs, preferably RGB (red, green, blue) outputs as well as horizontal and vertical synchronization signal outputs, to directly drive the video display monitor. The IMC transfers data between the system bus and system memory and also transfers data between the system memory and the video display output, thereby eliminating the need for a separate graphics subsystem. The IMC also improves overall system performance and response using main system memory for graphical information and storage. The IMC system level architecture reduces data bandwidth requirements for graphical display data since the host CPU is not required to move data between main memory and the graphics subsystem as in conventional computers, but rather the graphical data resides in the same subsystem as the main memory. Therefore the host CPU or DMA master is not limited by the available bus bandwidth. The IMC includes compression and decompression engines for compressing and decompressing data within the system. The IMC also includes a novel pointer-based display list architecture which includes windows workspace areas spaces which define the format of the data and the data type to read or written.

87 citations

Patent
17 Nov 2003
TL;DR: In this article, the authors present a data processing apparatus and a method for managing access to a memory within the data processing system, which consists of a processor operating in a plurality of modes and domains, including at least one nonsecure mode being a mode in the non-secure domain and at least two secure modes being a modes in the secure domain, and a processor being operable such that when executing a program in a secure mode, said program has access to secure data which is not accessible when said processor is operating in an non secure mode.
Abstract: The present invention provides a data processing apparatus and method for managing access to a memory within the data processing apparatus. The data processing apparatus comprises a processor operable in a plurality of modes and a plurality of domains, said plurality of domains comprising a secure domain and a non-secure domain, said plurality of modes including at least one non-secure mode being a mode in the non-secure domain and at least one secure mode being a mode in the secure domain, said processor being operable such that when executing a program in a secure mode said program has access to secure data which is not accessible when said processor is operating in a non-secure mode. Further, a memory is provided for storing data required by the processor, and consists of secure memory for storing secure data and non-secure memory for storing non-secure data. The memory further contains a non-secure table and a secure table, the non-secure table being within the non-secure memory and arranged to contain for each of a number of first memory regions an associated descriptor, and the secure table being within the secure memory and arranged to contain for each of a number of second memory regions an associated descriptor. When access to an item of data in the memory is required by the processor, the processor issues a memory access request, and a memory management unit is provided to perform one or more predetermined access control functions to control issuance of the memory access request to the memory. The memory management unit comprises an internal storage unit operable to store descriptors retrieved by the memory management unit from either the non-secure table or the secure table, and in accordance with the present invention the internal storage unit comprises a flag associated with each descriptor stored within the internal storage unit to identify whether that descriptor is from the non-secure table or the secure table. By this approach, when the processor is operating in a non-secure mode, the memory management unit is operable to perform the predetermined access control functions for the memory access request with reference to access control information derived from the descriptors in the internal storage unit retrieved from the non-secure table. In contrast, when the processor is operating in a secure mode, the memory management unit is operable to perform the predetermined access control functions for the memory access request with reference to access control information derived from the descriptors in the internal storage unit retrieved from the secure table. This approach enables different descriptors to be used for the control of accesses to memory in either the secure domain or the non-secure domain, whilst enabling such different descriptors to co-exist within the memory management unit's internal storage unit, thereby avoiding the requirement to flush the contents of such an internal storage unit when the operation of the processor changes from the secure domain to the non-secure domain, or vice versa.

87 citations

Book ChapterDOI
01 Jan 1990
TL;DR: This paper considers an implementation of a persistent store based on a large virtual memory and shows how stability is achieved, which is important for the store being resilient to failures.
Abstract: Persistent systems support mechanisms which allow programs to create and manipulate arbitrary data structures which outlive the execution of the program which created them. A persistent store supports mechanisms for the storage and retrieval of objects in a uniform manner regardless of their lifetime. Since all data of the system is in this repository it is important that it always be in a consistent state. This property is called integrity. The integrity of the persistent store depends in part on the store being resilient to failures. That is, when an error occurs the store can recover to a previously recorded consistent state. The mechanism for recording this state and performing recovery is called stability. This paper considers an implementation of a persistent store based on a large virtual memory and shows how stability is achieved.

87 citations

Journal ArticleDOI
TL;DR: This work addresses the problem of system power reduction through transition count minimization on the memory address bus when behavioral arrays are accessed from memory by exploiting regularity and spatial locality in the memory accesses and determining the mapping of behavioral array references to physical memory locations to minimize address bus transitions.
Abstract: Arrays in behavioral specifications that are too large to fit into on-chip registers are usually mapped to off-chip memories during behavioral synthesis. We address the problem of system power reduction through transition count minimization on the memory address bus when these arrays are accessed from memory. We exploit regularity and spatial locality in the memory accesses and determine the mapping of behavioral array references to physical memory locations to minimize address bus transitions. We describe array mapping strategies for two important memory configurations: all behavioral arrays mapped to a single off-chip memory and arrays mapped into multiple memory modules drawn from a library. For the single memory configuration, we describe a heuristic for selecting a memory mapping scheme to achieve low power for each behavioral array. For mapping into a library of multiple memory modules, we formulate the problem as three logical-to-physical memory mapping subtasks and present experiments demonstrating the transition count reductions based on our approach. Our experiments on several image processing benchmarks show power savings of up to 63% through reduced transition activity on the memory address bus in the single memory case. We also observe a further transition count reduction by a factor of 1.5-6.7 over a straightforward mapping scheme in the multiple memories configuration.

87 citations


Network Information
Related Topics (5)
Cache
59.1K papers, 976.6K citations
94% related
Scalability
50.9K papers, 931.6K citations
92% related
Server
79.5K papers, 1.4M citations
89% related
Virtual machine
43.9K papers, 718.3K citations
87% related
Scheduling (computing)
78.6K papers, 1.3M citations
86% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202333
202288
2021629
2020467
2019461
2018591