What is RISC?

February 10, 2020

The Central Processing Unit Architecture operates using an instruction set architecture. RISC, or Reduced Instruction Set Computer, is a type of microprocessor architecture that uses a small, highly-optimized set of instructions. RISC offers some advantages over CISC and have even been used in the current world’s fastest supercomputer.

Origins of RISC

RISC architecture started in the late 1970s. According to Standford, the first RISC projects came from IBM, Standford, and UC-Berkeley. A lot of research went into the development, a big portion that lead to RISC was John Cocke of IBM Research who proved that about 20% of instructions in a computer did 80% of the work TechTarget.

How is RISC Different From the Current CISC

According to Gartner, RISC has five design principles:

  • Single-cycle execution
  • Hard-wired control, little or no microcode
  • Simple instructions, few addressing modes
  • Load and store, register-register design
  • Efficient, deep pipelining

This means that RISC emphasizes software and favors more transistors for storing memory. This is in contrast to CISC which emphasizes hardware and used many transistors for storing complex instructions. RISC also has less steps between machine instructions and execution. RISC only has machine instructions leading to instruction execution, while CISC has machine instructions to microcode conversion to microinstruction then microinstruction execution.

RISC in Today’s Computers

There are many computer systems that take advantage of a RISC processor. Some notable ones are the A13 Bionic, which powers all iPhone 11 models, and the A12X Bionic, which powers the iPad Pro. RISC processors are not only in smaller devices, or devices that draw little power. The supercomputer Astra, made by Hewlett Packard Enterprise, is the largest ARM-based supercomputer to be used by the Sandia National Laboratory for nuclear research (Engadget). Astra runs off of Cavium ThunderX2 ARM processors. According to Engadget, the ARM chipset offers 33 percent faster memory speeds than many CISC 64 bit processors.

Even the most powerful supercomputer in the world, as of November of 2019, is powered off of RISC based processors, running at an astounding 200 PFLOPS/s (200,000,000,000,000,000 FLOPS/s). The Summit Supercomputer located at Oak Ridge National Laboratory uses a total of 9,216 IBM POWER9 processors which use the Power Instruction Set Architecture, which is based upon RISC.

These processors are considered “ARM-based” processors which are generally less expensive than standard CISC processors. They are also considerably more (transistor) dense and more power efficient.

RISC has already made a big appearance in modern computing, being seen in computers from the iPhone to the fastest supercomputer Summit. It offers denser processors and more power efficient ones.

About the author

Gregory Manley

Gregory Manley is a sophomore at Colorado School of Mines where he is majoring in Computer Science with a minor in Mining Engineering. He is the owner of iTech News and a contributor for Section’s Engineering Education Content Program. His management of iTech News has led him to work with many brands on writing technology focus articles.

This article was contributed by a student member of Section's Engineering Education Program. Please report any errors or innaccuracies to enged@section.io.