It is a discrete industry, little known to the general public, that of microprocessors. However, we owe him, quite simply, the advent of the information society. Without the microprocessor and its exponential performance, nothing we use every day would exist. Computing would have remained an arid, expensive, centralized technique reserved for companies and institutions. Didn’t IBM President Thomas Watson say in the 1950s that a dozen computers would be enough to cover the world’s needs? This scenario was foiled by the invention of the microprocessor only fifty years ago. And the world has been profoundly changed. However, designing and producing microprocessors is not easy. It is, in fact, one of the most sophisticated industries ever imagined by man. In 2021, it is at the heart of all strategic issues.
What is the Computer and the Role of the Processor
A computer program is, in essence, a flow of instructions executed by a processor. Each instruction requires one or more clock cycles. The instruction is executed in as many steps as there are cycles required. Sequential microprocessors execute the next instruction when they have completed the current instruction. In the case of instruction parallelism, the microprocessor can process several instructions in the same clock cycle, provided that these different instructions do not simultaneously mobilize a single internal resource. In other words, the processor executes instructions that follow each other and are not dependent on each other at different stages of completion. This upcoming execution queue is called a pipeline. This mechanism was first implemented in the 1960s by IBM.
Present Day Microprocessors:
Today, Power PC processor designers are not simply looking to execute multiple independent instructions at the same time; they are looking to optimize the execution time of all instructions. For example, the processor can sort the instructions so that all its pipelines contain independent instructions. This mechanism is called out-of-order execution. This type of processor was essential for consumer machines from the 1980s until the 1990s.
To avoid wasting time waiting for new instructions, and especially the time it takes to reload the context between each change of threads, the founders have added optimization processes to their processors so that the lines can share the pipelines, caches, and registers.
The Development of PC8548E
The PC8548E has the Power Architecture® processor and the system logic used for networking, storage options, and embedded applications for general purposes.
The running of the e500 processor is up to 1.2GHz.
Teledyne e2v is currently offering this Space processor with the Ceramic Non-Hermetic Flip Chip Space solutions. Added to this, the PC8548E has the QML-Y qualification.
With the use of the CMOS 90nm SOI technology from the NXP, this device has positively passed the test of Total Ionizing Dose and the test of the Heavy Ions. It demonstrates its appropriateness for space applications in the Rad Tol category.
For a long time already, the idea of having several processors coexists within the same component, for example, System on Chip. This consisted, for instance, in adding to the processor, an arithmetic coprocessor, a DSP, even a memory cache, possibly even all of the components found on a motherboard. Therefore, processors using two or four cores have appeared, such as IBM’s POWER4, for example, released in 2001. They have the technologies mentioned above. Computers with this type of processors cost less than buying an equivalent number of processors. However, performance is not directly comparable, depending on the problem being addressed. Specialized APIs have been developed to take full advantage of these technologies, such as Intel’s Threading Building Blocks.