Current memory management policy is, on most sys-tems, global and in exible. Current systems provide poor mechanisms to allow the programmer to share knowledge of access patterns that may generate im-proved page swapping and caching. In this paper we modify a popular operating system's memory manage-ment policy to take into account hints given by the pro-grammer. The system consists of love and hate hints given to the kernel via system calls. It gives the pro-grammer the option of moving a page to the front of the swap line or informing the operating system that the page should be kept on the active list. The paper demonstrates the improved performance that a system can gain through the use of the love and hate hints
textContemporary DRAM systems have maintained impressive scaling by managing a careful balance betwe...
textThe trends in virtualization as well as multi-core, multiprocessor environments have translated...
As processors continue to exploit more instruction level parallelism, greater demands are placed on ...
The swap mechanism allows operating systems to manage more memory than the available RAM space, by t...
Abstract— Memory and swap management is an important issue in operating systems, owing to the large...
In a virtual memory manager, life is easy when you have a lot of free memory. A page fault occurs, y...
Recent byte-addressable Non-Volatile Memory (NVM) technologies enable hybrid memory systems comprisi...
Abstract. The swap mechanism allows an operating system to work with more memory than available RAM ...
Increased performance demand of modern applications has resulted in large memory modules and higher ...
Out-of-core applications consume physical resources at a rapid rate, causing interactive application...
Managing a fixed amount of memory (RAM) optimally is a long-solved problem in the Linux kernel. Man-...
Abstract. Memory traces record the addresses touched by a program during its execution, enabling man...
The memory system is often the weakest link in the performance of today’s computers. Cache design ha...
Phase-Change Memory (PCM) technology has received substantial attention recently. Because PCM is byt...
This paper describes a new technique to reduce page-faults in multiprocessing systems by supplying c...
textContemporary DRAM systems have maintained impressive scaling by managing a careful balance betwe...
textThe trends in virtualization as well as multi-core, multiprocessor environments have translated...
As processors continue to exploit more instruction level parallelism, greater demands are placed on ...
The swap mechanism allows operating systems to manage more memory than the available RAM space, by t...
Abstract— Memory and swap management is an important issue in operating systems, owing to the large...
In a virtual memory manager, life is easy when you have a lot of free memory. A page fault occurs, y...
Recent byte-addressable Non-Volatile Memory (NVM) technologies enable hybrid memory systems comprisi...
Abstract. The swap mechanism allows an operating system to work with more memory than available RAM ...
Increased performance demand of modern applications has resulted in large memory modules and higher ...
Out-of-core applications consume physical resources at a rapid rate, causing interactive application...
Managing a fixed amount of memory (RAM) optimally is a long-solved problem in the Linux kernel. Man-...
Abstract. Memory traces record the addresses touched by a program during its execution, enabling man...
The memory system is often the weakest link in the performance of today’s computers. Cache design ha...
Phase-Change Memory (PCM) technology has received substantial attention recently. Because PCM is byt...
This paper describes a new technique to reduce page-faults in multiprocessing systems by supplying c...
textContemporary DRAM systems have maintained impressive scaling by managing a careful balance betwe...
textThe trends in virtualization as well as multi-core, multiprocessor environments have translated...
As processors continue to exploit more instruction level parallelism, greater demands are placed on ...