¡GoConqr estrena diseño! Cuéntanos qué te parece.

Mapa mental

CPU Performance

Harry Taylor Inglés

Contenido del mapa

CPU Performance

  1. Clock Speed

    • Clock speed is the number of operations that a CPU can execute per second. Most modern processors can achieve 4,000,000,000 instructions per second
      • You can increase the Clock speed by building a processor that has more transistors inside of it. However this becomes more challenging as engineers try to fit more transistors into the same area.
  2. Cache Memory

    • RAM, even though it is fast compared to other storage devices, is still quiet slow when compared to the speeds that processors can operate at. To avoid this problem most modern computers will have a small amount of faster memory call 'Cache Memory'.
      • With the cache being built directly into the processor to reduce distance travelled by data. By anticipating the instructions and data that will most likely be used, this data is can be kept in the cache , therefore increasing the overall speed of the processor.
        • However cache has some disadvantages. Increasing the the cache will slow down its overall speed. Therefore most modern computer will contain several small layers of cache memory. The first and smallest cache (meaning that it is the fastest) will be checked. This is followed by the next largest until the RAM is checked
  3. Cores

    • In a computer each core is a distinct processing unit in the CPU. A core will have its own cache as well as sharing a higher-level cache.
      • When a computer is tasked with multi-tasking, different cores will run different application or problems or will work on the same problem. This odes not mean that the computer will work three or four times fast though.
  4. Pipelining

    • Pipeling is used in any modern processor. It is where while one instruction if being executed the next instruction in line will be being decoded and the one after that will be fetched
    • Pipeling does have its limitations though. Sometimes it is unable to accurately predict what instruction needs to fetched and decoded in the right order. In cases like this we need to 'flush' the pipes of the existing instructions. The more we have to do this, the less of the benefits we get from the process.

Información

Descripción
Mapa Mental sobre CPU Performance, creado por Harry Taylor el 13/09/2015.
Creado por
Harry Taylor
Visitas
0
Valoración

Escríbenos

¿Sobre qué es tu mensaje?

¿Qué te parece el nuevo diseño?

¿Te gusta?