摘要 :
Double Data Rate Synchronously Dynamic RAM(DDR-SDRAM)is a kind of new high-capacity and high-speed ram chip. The article describes its working features. Implementation of the DDR SDRAM controller is analyzed and some details are also given.
摘要 :
In the arsenal of resources for improving computer memory system performance, predictors have gained an increasing role in the past few years. They can suppress the latencies when accessing cache or main memory. In our previous wo...
展开
In the arsenal of resources for improving computer memory system performance, predictors have gained an increasing role in the past few years. They can suppress the latencies when accessing cache or main memory. In our previous work we proposed predictors that not only close the opened DRAM row but also predict the next row to be opened, hence the name 'Complete Predictor'. It requires less than 10 kB of SRAM for a 2 GB SDRAM system. In this paper we evaluate how much additional hardware is needed and whether the activations of the predictors will slow down the DRAM controller.
收起
摘要 :
In the arsenal of resources for improving computer memory system performance, predictors have gained an increasing role in the past few years. They enable hiding the latencies when accessing cache or main memory. In our previous w...
展开
In the arsenal of resources for improving computer memory system performance, predictors have gained an increasing role in the past few years. They enable hiding the latencies when accessing cache or main memory. In our previous work we proposed a DDR SDRAM controller with predictors that not only close the opened DRAM row but also predict the next row to be opened. In this paper we explore the possibilities of trying the same techniques on the latest type of DRAM memory, DDR3 SDRAM, with further improvements of the predictors.
收起
摘要 :
There are occasions, such as the present, when developments in the computer industry create an opportunity to upgrade or purchase new systems at prices that offer much better value for money than at other times. The usual scenario...
展开
There are occasions, such as the present, when developments in the computer industry create an opportunity to upgrade or purchase new systems at prices that offer much better value for money than at other times. The usual scenario for this type of opportunity is the imminent release of a new technology that will suddenly make a range of existing computers and components obsolete. However, it is important to remember these 'obsolete' systems and parts were the latest technology at the time and will still perform adequately for many years for most common uses and, therefore, offer excellent value for money. For example, a well-known mail-order computer company was offering a budget desktop computer system during October 2000 for 1699 dollar - that system had a 733 Megahertz (MHz) processor, 32 Megabyte (MB) random access memory (RAM), 7.5 Gigabyte (GB) hard drive, 48-speed compact disk (CD) drive, 15-inch monitor and software package, but no video accelerator card.
收起
摘要 :
In the arsenal of resources for computer memory system performance improvement, predictors have gained an increasing role in the past years. They can suppress latencies when accessing cache or main memory. In our previous work we ...
展开
In the arsenal of resources for computer memory system performance improvement, predictors have gained an increasing role in the past years. They can suppress latencies when accessing cache or main memory. In our previous work we considered using predictors to decrease Synchronized Dynamic Random Access Memory (SDRAM) latency. If one wants to apply several predictors at the same time, there has to exist proper synchronization of the predictors. This stems from the fact that SDRAM controller has to activate and update each of the predictors at proper times, while at the same time executing standard SDRAM commands, such as opening/closing rows, accessing columns, etc. In this paper we propose a universal synchronization algorithm for predictors for SDRAM memories, which can be used with practically any predictor, regardless of its specific method of predicting, and with very little or no modifications.
收起