The When first released, the Dhrystone benchmark fulfilled a useful function – at least it gavean alternative indicator to vendors’ literal MIPS ratings. However, almost twenty yearslater, there are undoubtedly better benchmarks available for measuring processorperformance.Code profiling demonstrates that Dhrystone is not a generalparticularly good predictor of real-world performance benchmark used , especially for complex embedded systems. The benchmark’ssusceptibility to being optimized, yielding large variations in results, is a considerableweakness if the intent is to evaluate processor execution timeuse the numbers for comparative purposes. It doesn't accurately measure ARM’s ownmeasurements, presented here, demonstrate that each method of optimization canincrease the speed of apparent DMIPS/MHz performance, and when combined can improve the cpuDhrystone score by up to three times.ARM’s customers still occasionally ask for Dhrystone MIPS data, and so ARM willcontinue to provide these numbers. What makes dhrystone useful is it gives a more meaningful overall performance However, ARM does not recommend that Dhrystoneresults be used as part of the system because it uses a comparison machineany embedded processor evaluation exercise, due to its manynoted deficiencies. When benchmarking Where ARM provides a systemDhrystone figure, always make sure what is it you're benchmarking forwill be based on itsstandard tool suite under the conditions outlined above, and will therefore be 100%publicly and independently reproducible. Dhrystone's popularity back then still lingers in this age; although there are In general, more representative data will be obtained by using a lot suite of benchmark software out there that surpasses data.First, this enables benchmarks to reflect the requirements of specific application domains.Second, the already aged Dhrystone; There is no doubt that Dhrystone is effects of any one of optimization will be mitigated by the many useful benchmark software out therehaving awide spread of algorithms and coding styles. Results based on such an approach are morelikely to be representative of processor performance than compiler or tool chainperformance.