 Welcome to the session on system attributes to performance. At the end of the session, students will be able to classify various attributes on which performance of system is measured. So when we talk about performance of computer system, we would describe how quickly a given system can execute a program or programs. So there we are looking the different things or different factors which help us to execute any system faster and is based on like hardware technology, architectural feature, efficient resource management, then again we are looking for the algorithm design, data structure, language efficiency, programming skills or programmer skill, again compiler technology. So performance of computer system we will describe how quickly a given system can execute. So here we are looking some performance indicators which are focusing on turn around time. So turn around time specifically it will depend on this can memory access, then input output things, then compilation time, operating system overhead, CPU time. So an ideal performance of computer system means a perfect match between the machine capability and program behavior. The machine capability can be improved by using better hardware technology and efficient resource management. But it is far as program behavior is concerned, it depends on the code used, compiler used and other run time conditions. So let us check these performance indicators one by one. The very first one is clock rate and the CPU. Since ION system overhead frequently overlaps processing by other programs. So it is fair to consider only the CPU time used by a program. And the user CPU time is the most important factor, CPU is driven by a clock with a constant cycle time. So always keep remember CPU is driven by a clock with a constant cycle time usually measured in nanoseconds which controls the rate of internal operations in the CPU. And the clock mostly has the constant cycle time t. So the inverse of the cycle time is clock rate. So the inverse of the clock, this cycle time is what? The clock rate. So that is f equals to 1 by tau that is 1 by t or we can call it some time tau. It is measured in megahertz. So about the average CPU, it is easy to determine the average number of cycles per instructions for a particular processor. I will repeat it is easy to determine the average number of cycles per instruction for a particular processor. If we know the frequency of occurrence of each instruction time of course, any estimate is valid only for a specific set of programs. And then only if there are sufficiently large number of instructions, so in general the term CPI is used with respect to a particular instruction set and a given program mix. So the time required to execute a program containing IC instructions is just total t. So here I will tell you I will just repeat again the keyword that is the time required to execute a program that is capital T. The time required to execute a program which contains IC instructions is just t equals to that is IC into CPI into tau fine. So here again I will elaborate these things. So the time required to execute a program containing number of instruction IC that is IC instruction is just capital T equals to IC into CPI into tau. So each instruction must be phased from memory, decoded and operands phased from memory. The instructions executed and the result will be get stored. So the time required to access memory is called memory cycle time. I will repeat the time required to access memory is called memory cycle time which is usually k times the processor cycle time tau. So the value of k depends on the memory technology and the processor memory interconnection scheme. So the processor cycle required for each instruction. The processor cycle required for each instruction CPI can be attributed. Cycle time needed or the cycles needed for instruction decoded execution p and cycle needed for memory reference that is m into k. So the total time required to execute a program can help to rewrite as t equals to IC into p plus m into k into tau. So here you have a question that is CPU is driven by a clock with constant cycle time t. This can be measured in different you have the units here so please choose the correct option. CPU is driven by a clock with a constant cycle time t. This can be measured in so your option is d that is nanosecond. So CPU is driven by clock with a constant cycle time this measured in nanoseconds. The next thing is about the system attributes. The five performance factors they are like number of instruction IC, processor cycle time that is p then memory access time that is m and the k then it's a clock cycle or we can say the clock cycle with constant time is tau are influenced by four system attributes one is instruction set architecture then compiler technology right then CPU implementation and control then cache and memory hierarchy. So these are what different attributes which are gets influenced by our five performance factors. So here the total CPU time can be used as a basis in estimating the execution rate of a processor. And again here we are moving for the next performance factor called MIPS rate if c is the total number of clock cycles needed to execute a given program then total CPU time can be estimated as c into t. So this t also we can write it as a 1 by f okay. So other relationship are easily observed that CPI equals to c by IC so t is nothing but IC into CPI into tau that we know then t equals to IC into CPI by f. So the processor speed is often measured in terms of millions of instructions per second and frequently called it as the MIPS rate of the processor okay. So MIPS rate is we are going to calculate as that is number of instructions upon total time t into 10 to the power 6 okay so that we can write with the different way that is f by CPI into 10 to the power 6 or f into IC upon c into 10 to the okay. So here the MIPS rate is directly proportional to the clock rate and inversely proportional to the CPI. So all four system attributes like instruction set, compiler, processor and memory technology affects MIPS rate that is million instruction per second. So which varies also from program to program. Now the very important part again we are moving here that is throughput rate and this throughput here we are going to calculate with the WP that you always keep in keep remember. So the number of programs a system can execute per unit time WS that is programs per second. So CPU throughput WP is defined as is equals to f upon number of instructions into CPI. So here a multiprogram system the system throughput is often less than the CPU throughput. Now I am coming to the one more important part that is programming environment. So the programming environment depends on the programming environment provided on the users and conventional computers are used in sequential programming environment with tools developed for uniprocessor computers and also parallel computers need parallel tools that allow specification or easy detection of parallelism and operating systems that can perform parallel scheduling of concurrent events. So shared memory allocation and shared peripheral and communication links. So in the programming environment it depends on the programming environment to the users for developing the parallel processing execution. So under the parallel computers we need parallel tools and advanced programming technologies and the operating systems which help us for parallel scheduling of concurrent events and the shared memory allocations and the shared peripherals with the communication links. So these are what my reference is thank you.