8.2 Bridge- Based Bus Architectures From a logical viewpoint, all of the system components are connected directly to the system bus in the previous section. From an operational viewpoint, this approach is overly burdensome on the system bus because simultaneous transfers cannot take place among the various components. While every device at the same time, […]
Continue reading…
Principles of computer architecture
Input and output: simple bus architectures ( bus structure, protocol, and control, bus clocking, the synchronous bus, the asynchronous bus and bus arbitration—masters and slaves).
INPUT AND OUTPUT In the earlier chapters, we considered how the CPU interacts with data that is accessed internal to the CPU, or is accessed within the main memory, which may be extended to a hard magnetic disk through virtual memory. While the access speeds at the different levels of the memory hierarchy vary dramatically, […]
Continue reading…
SUMMARY OF MEMORY
■ SUMMARY Memory is organized in a hierarchy in which the densest memory offers the least performance, whereas the greatest performance is realized with the memory that has the least density. In order to bridge the gap between the two, the principle of locality is exploited in cache and virtual memories. A cache memory maintains […]
Continue reading…
Memory: case study: ramous memory and case study: the Intel Pentium memory system.
Case Study: Rambus Memory There was a time when computer technology would be pushed from the laboratory into the marketplace. As the consumer marketplace for computing devices exploded, the “technology push” was replaced by “market pull,” and consumer demand then dominated the preferences of technologists when it came to developing a new memory technology. High […]
Continue reading…
Memory: advanced topics (tree decoders, decoders for large rams and content-addressable (associative) memories).
Advanced Topics This section covers two topics that are of practical importance in designing memory systems: tree decoders and content-addressable memories. The former are required in large memories. The latter are required for associative caches, such as a TLB, or in other situations when data must be looked up at high speed based on its […]
Continue reading…
Memory: virtual memory (segmentation, fragmentation virtual memory vs, Cache memory and the translation look aside buffer).
SEGMENTATION Virtual memory as we have discussed it up to this point is one-dimensional in the sense that addresses grow either up or down. Segmentation divides the address space into segments, which may be of arbitrary size. Each segment is its own one-dimensional address space. This allows tables, stacks, and other data structures to be […]
Continue reading…
Memory: virtual memory (overlays and paging).
Virtual Memory Despite the enormous advancements in creating ever larger memories in smaller areas, computer memory is still like closet space, in the sense that we can never have enough of it. An economical method of extending the apparent size of the main memory is to augment it with disk space, which is one aspect […]
Continue reading…
Memory: cache memory (set associative mapped cache, cache performance, hit ratios and effective access times, multilevel caches and cache management).
7.6.3 SET ASSOCIATIVE MAPPED CACHE The set associative mapping scheme combines the simplicity of direct mapping with the flexibility of associative mapping. Set associative mapping is more practical than fully associative mapping because the associative portion is limited to just a few slots that make up a set, as illustrated in Figure 7-15. For […]
Continue reading…
Memory: cache memory (associative mapped cache and direct mapped cache).
Cache Memory When a program executes on a computer, most of the memory references are made to a small number of locations. Typically, 90% of the execution time of a program is spent in just 10% of the code. This property is known as the locality principle. When a program references a memory location, it […]
Continue reading…
Memory: chip organization (constructing large rams from small rams) and commercial memory modules.
7.1 Chip Organization A simplified pinout of a RAM chip is shown in Figure 7-3. An m-bit address, having lines numbered from 0 to m-1 is applied to pins A0-Am-1, while asserting CS (Chip Select), and either WR (for writing data to the chip) or WR (for reading data from the chip). The over bars […]
Continue reading…