What is a system-on-chip (SoC)? What are the advantages and disadvantages of an
ID: 3692168 • Letter: W
Question
What is a system-on-chip (SoC)?
What are the advantages and disadvantages of an SoC vs. a discrete component computer system?
Why are the advantages of an SoC particularly well-suited to embedded systems? How do SoC’s provide support for debugging software?
What is bit-banding?
How do SoC’s provide support for saving power?
What are the 4 general classes of interfaces found on SoC’s and how are these related to PC
architectures?
What is a shared system bus multiprocessor architecture?
Describe the following alternatives to a shared system bus architecture: crossbar, ring, 2D torus ring, 2D mesh network. Some of these alternatives have names similar to those used in the discussion of parallel multiprocessor organization. Why is the discussion of these alternatives in the context of a multicore different than the discussion in the context of parallel multiprocessors?
What is NUMA and why is it different from the memory model used in traditional processors like those studied in prerequisite courses?
Modern multicore architectures have been influenced by the evolution of personal computer
(PC) architectures.
i. What are the PC’s front-side and back-side buses?
ii. What are the purposes of the PC’s Northbridge and Southbridge?
iii. Why did the evolution of these architectures move towards serial buses and away from parallel buses?
iv. How has AMD’s HyperTransport and Intel’s Quickpath interconnects evolved from PC
architecture?
v. How has ARM chosen to evolve buses similar to PC buses?
What is the difference between a heterogeneous and a homogeneous multicore?
Why is concurrency in a parallel organization visible to software, while the instruction-level parallelism is general invisible to software?
What are the classifications of parallel architectures in the Flynn Taxonomy? Briefly describe each.
What is hyper-threading? Why might some people call hyper-threading a single-processor
“trick” while others might say it is a valid parallel organization?
What problem with superscalar architectures do hyper-threads exploit?
What is the theoretical speedup that can be obtained by an M-processor computer? Why is this
“ideal” speedup unlikely to occur in practice?
What is a shared-memory multiprocessor? How is a shared-memory multiprocessor different from a distributed-memory multiprocessor?
What is the “cache coherence” problem associated with multiprocessors, and how does the problem relate to shared-memory and distributed-memory systems?
What is cache snooping and how is it related to cache coherence?
What is a massively-parallel multiprocessor (supercomputer)? How is a cluster different from a supercomputer?
What is a grid and why is it different from both a supercomputer and a cluster?
When considering the performance of an interconnection network, of what relevance are the number of connections, the degree, and the diameter?
Describe: a fully connected network, a ring, an M-dimensional hypercube.
Suppose we must connect N = 8 processors using a fully connected network, a ring network, or
a 3-dimensional hypercube network. Compare the three network topologies with respect to the total number of connections, the degree, and the diameter for this specific case.
When considering the performance of a multiprocessor network, what is the efficiency metric and how does the efficiency relate to scalability?
What is Amdahl’s Law and why is it relevant to the performance of a multiprocessor?
Explanation / Answer
What is a system-on-chip (SoC)?
Answer : A system on a chip or system on chip is an integrated circuit (IC) that integrates all components of
a computer or other electronic system into a single chip. It may contain digital, analog, mixed-signal, and often
radio-frequency functions—all on a single chip substrate. SoCs are very common in the mobile electronics market because of
their low power consumption.
What are the advantages and disadvantages of an SoC vs. a discrete component computer system?
Answer: An SoC is only a little bit larger than a CPU, and yet it contains
a lot more functionality.Using SoCs, we can put complete computers in smartphones and tablets, and still have plenty of space for batteries. If you use a CPU, it’s very hard to make a computer that’s smaller than 10cm (4 inches) squared,
purely because of the number of individual chips that you need to squeeze in.
Due to its very high level of integration and much shorter wiring, an SoC also uses considerably less power.It’s much
cheaper to build a computer using an SoC, too.
Why are the advantages of an SoC particularly well-suited to embedded systems? How do SoC’s provide support for
debugging software?
Answer :
SoC is a system on a VLSI chip that has all needed analog as well as digital
circuits, processors and software, for example, single-chip mobile phone.ItEmbeds
multiple processors,memories,multiple standard source solutions,logic and analog units.
In this way,an SoC is particularly well-suited to embedded systems.
A common criteria for the debugging software is to be able to debug from reset and
through partial power cycles. This requires careful design of power domains and reset signals.
Critically, reset of the debug control register should be separated from that of the functional
system. Power-down can be handled in different ways such as ignoring power-down
signals when debugging or maintaining the debug logic in different power domains which are not
powered down while debugging.
What is bit-banding?
Answer:Bit Banding is defined as a method of doing atomic bitwise modifictions to memory.
How do SoC’s provide support for saving power?
Answer:Power savings can be achieved by integrating the ADC that interfaces with the sensor.
Integrated ADCs feature minimal static power dissipation, resulting in minimization of the power
consumption as speed is reduced; several performance modes, where power is minimized as performance
settings are reduced; multiple power modes, with the ability to change between them quickly and
without losing accuracy and performance.
What are the classifications of parallel architectures in the Flynn Taxonomy? Briefly describe each.
Answer:The four classifications defined by Flynn are based upon the number of concurrent instruction
streams and data streams available in the architecture.
Single instruction stream, single data stream (SISD)
A sequential computer which exploits no parallelism in either the instruction or data streams. Single control unit (CU) fetches single instruction stream (IS) from memory. The CU then generates appropriate control signals to direct single processing element (PE) to operate on single data stream (DS) i.e., one operation at a time.
Examples of SISD architecture are the traditional uniprocessor machines like older personal computers and mainframe computers.
Single instruction stream, multiple data streams (SIMD)
A computer which exploits multiple data streams against a single instruction stream to perform operations which may be naturally parallelized. For example, an array processor or graphics processing unit (GPU)
Multiple instruction streams, single data stream (MISD)
Multiple instructions operate on one data stream. Uncommon architecture which is generally used for fault tolerance. Heterogeneous systems operate on the same data stream and must agree on the result. Examples include the Space Shuttle flight control computer.
Multiple instruction streams, multiple data streams (MIMD)
Multiple autonomous processors simultaneously executing different instructions on different data. MIMD architectures include multi-core superscalar processors, and distributed systems, using either one shared memory space or a distributed memory space.
What is hyper-threading? Why might some people call hyper-threading a single-processor
“trick” while others might say it is a valid parallel organization?
Answer:Hyper-threading is Intel's proprietary simultaneous multithreading implementation used to improve parallelization of computations performed on x86 microprocessors.
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.