Benchmarking Microarchitecture Using Software Research Paper

Benchmarking Microarchitecture Using Software - Research Paper Example 327). These are defined by many parameters such as the instructions set, the instruction encoding and the rules for using these instructions. Some of the rules are mnemonics addressing modes and functionality. The details at this level are invisible to the user. This is known as abstraction. The implementation of MIPS is borne on the platform of Reduced Instruction Set Computers (RISC). Validation and Verification Validation in system design is concerned with the confirmation that the goals set are the right ones. This is to enable the proper implementation of the project. Validation is usually done during several different stages of development. It is essential at each level because it ensures that the preceding level is valid before proceeding to the next one. Validation is achieved in different ways. In the micro architecture, use of test inputs and checking on the result is the most common way of validation and verifying the system. So, before the development of the I/O functions , an algorithm is used to test various features if the microsystem. The benchmark program checks for the validity of many features of the micro architecture and verifies some of the features (Patterson, p. 47). Some of these features of the microsystem are illustrated below. The benchmark program used generates a quadratic residue from the integer inputs used by the algorithm. This program checks for functionality of different features mostly in computation. This is because it is the most important feature of the design. The instruction that will be fed into the system will need to be computed consistently. This makes the feature the basis of the design. This computation results in the output of the residue problem used. The program can be altered by using different quadratic residue problems to check for consistency in the system. This is also good for checking the range of values that the program can accept as input and compute without crashing. This is vital because then the desi gners can tell whether the system can accept the range of inputs specified in the requirements. They can also check whether it will need reprogramming. During this kind of testing, sometimes, wrong inputs are put to check how the system will react to them. The system should be able to handle wrong inputs without devastating consequences that will lead to system failure. The users are likely to enter wrong inputs from time to time and their handling should be factored in the design. The use of an 8-bit feedback register shows that the system has functionality for feedback. Feedback is described as the process where the system uses the output generated from a process as input for the same or a similar process. This is common in scenarios where complex computations are performed. These computations require the output of a process to be used as input. If this value has not been generated correctly, the system is likely to encounter an error. This is another feature, which is checked by the system. The application of this feedback is handled using a register. There are several other methods used for feedback systems such as logic circuits, but they are all used interchangeably. The program also has an increment functionality. This functionality is used to increase the value in the counter. The counter has many uses in the system, but is mostly utilized in checking the number of times a program performs a certain function. The increment functions increase the value