SRAM memory

When a CPU goes to get either its program instructions or data, it always has to get them from main memory. However, in some systems, there is a small amount of very fast SRAM memory, called cache memory, between the processor and main memory, and it is used to store the most frequently accessed information. Because it s faster than main memory and contains the most frequently used information, cache memory will increase the performance of any system. 

The register bank is mostly a sequential circuit, just like the program and data memory. Becattse of that, it is very sensitive to SEUs. The register bank can be implemented over a SRAM memory or by using flip-flops. In the first case, the same principles from the data program are applied. In the second case, hardware... 

The chosen embedded system was composed of a MIPS microprocessor hardened with HETA, an unhardened SRAM memory embedded in the FPGA, two SpW links (TARRILLO 2011) and the FPGA embedded Phase-Locked Loop (PLL) clock module. The system has some fault tolerant capabilities that are able to detect transient faults, but not necessary TID effects as radiation results will show. Figure 7.1 shows the architecture of the embedded system. 

To minimize the risk of data loss because of power failure or other reasons, the Mossbauer data are copied to a nonvolatile EEPROM (Electrically Erasable Programmable Read-Only Memory) every 9 minutes (software selectable). As the size of the EEPROM is smaller than the SRAM, the EEPROM can accumulate only up to ten Mossbauer spectra as a subset of the data from the SRAM. These spectra are obtained from the SRAM according to a pre-defined summation strategy. 

Hada H, Amanuma K, et al (2004) Capacitor-on-metal/via-stacked-plug (CMVP) memory cell technologies and application to a nonvolatile SRAM. Ferroelectric Random Access Memories Fundamentals and Applications 93, 215-232... 

Typical results obtained for the 256-kbit static random access memories (SRAMs), which were developed for space application by NASD A, are shown in Fig. 15. The SEU cross section increased with LET in a range above a threshold value and reached a constant... 

Fig. 3. Experimental demonstration of one-, two- and three-transistor logic circuits with CNTFETs [41]. Output voltages as a function of the input voltage are given for (A) an inverter, (B) a NOR gate, (C) a static random access memory cell (SRAM) and (D) a ring oscillator.

Since a cache should be fast it is constructed using SRAM. A typical solution is an LI cache on the same chip as the CPU and an external L2 cache made up of SRAM. Caches are organized in a number of equal sized slots known as cache lines. A cache line consists of several consecutive memory addresses. The line size varies from design to design, but is usually in the range of 64 to 512 bytes wide. Exactly how a lower level of memory is mapped to a higher and smaller level has a number of variations. At one extreme, the direct-mapped cache maps each memory element to exactly one of the cache lines. This can be quite restrictive. The opposite in flexibility is the fully associative cache which can map each memory address to any of the cache lines. The compromise between these extremes is called a set-associative cache. This means that a memory element can be mapped to a number, say 4, cache lines. Such a cache is then said to be 4-way set-associative. 

SRAM The S in SRAM stands for static. Static random access memory doesn t require the refresh signal that DRAM does. The chips are more complex and are thus more expensive. However, they are faster. DRAM access times come in at 80 nanoseconds (ns) or more SRAM has access times of 15 to 20 ns. SRAM is often used for cache memory. 

One type of memory is known as static random access memory (SRAM). It is called static because the information doesn t need a constant update (refresh). SRAM stores information as patterns of transistor ons and offs to represent binary digits. This type of memory is physically bulky and somewhat limited in its capacity. It can generally store only 256Kb (kilobits) per IC. The original PC and XT, as well as some notebook computer systems, use SRAM chips for their memory. 

Most new computers are moving away from SRAM, to the newer, more efficient type of memory known as DRAM. 

Dynamic random access memory (DRAM) was an improvement over SRAM. DRAM uses a different approach to storing the Is and Os. Instead of transistors, DRAM stores information as charges in very small capacitors. If a charge exists in a capacitor, it s interpreted as a 1. The absence of a charge will be interpreted as a 0. 

Because DRAM uses capacitors instead of switches, it needs to use a constant refresh signal to keep the information in memory. DRAM requires more power than SRAM for refresh signals and, therefore, is mostly found in desktop computers. 

There are two types of cache memory on-chip (also called internal or LI Cache) and off-chip (also called external or L2 Cache). Internal cache memory is found on Intel Pentium, Pentium Pro, and Pentium II processors, as well as on other manufacturer s chips. The original Pentium contains two 8KB-on-chip caches, one for program instructions and the other for data. External cache memory is typically either a SIMM of SRAM or a separate expansion board that installs in a special processor-direct bus. 

A, B, E, F. Since ROM stands for read-only memory and PROM stands for programmable ROM, they automatically are incorrect answers for the question. RAM and SRAM are both monikers of short-term memory (RAM), and therefore have to be erasable to be of any use. EPROM and EEPROMs are both special types of PROMs that are Erasable Programmable (EPROM) and Electronically Erasable and Programmable (EEPROM) and therefore have the ability to be erasable and re-usable. 

If SRAM and DRAM were the only different technologies to keep track of, purchasing memory would be a snap. Unfortunately, there are several (incompatible) types of DRAM from which you must choose ... 

A. Dynamic Random Access Memory (DRAM) is the type of memory that is expanded when you add memory. Static Random Access Memory (SRAM) is often used for cache memory. See Chapter 2 for more information. 

An alternative application of coupled inverters is to place an even number (say, 2) together and create a bistable logic reprogrammable memory circuit. An example SRAM formed in this way is shown in Fig. D.5 [145]. 

In this approach, entire wafers or segments of wafers containing ICs are stacked. Interconnections are formed from vias in the silicon. In vertically stacking memory chips or other IC chips, the chips are first thinned to several mils (4 mils or less), then adhesive bonded, and electrically interconnected directly from chip-to-chip, either from the edges or through vias in the silicon. In these approaches, both electrically conductive and electrically insulative adhesives are used. Thermally conductive preform adhesives or thin thermoplastic films are used to bond and isolate the chips within the stack and to dissipate heat. Several processes are available to vertically stack and interconnect chips of the same size and function such as SRAM, flash, and DRAM memory chips. Other processes have been developed to vertically stack chips of different sizes and functions, or to horizontally interconnect different chips in one layer, then vertically stack the layers. ... 

---