Flynn's Taxonomy as a modern classification

  • Thread starter Thread starter joel amos
  • Start date Start date
  • Tags Tags
    Classification
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
4 replies · 2K views
joel amos
Messages
104
Reaction score
0
Flynn's Taxonomy was a classification of computer architectures published in 1966. Computing has changed a lot since then. Is this taxonomy still useful? What are some of its shortcomings for evaluating modern computing systems?
 
Physics news on Phys.org
The short answer is yes. Why must older concepts have shortcomings? Fujitsu makes very expensive Sparc SIMD chips for Solaris systems, right now.
PS: if this is homework, which it sounds like, we may want to put it in another forum section.

Taxonomy is pigeon-holing to allow someone to start working on analyzing what they have. It often is used to set expectations or provide parameters or be explanatory for something. And to show how it relates to other similar things - in this case cpu's.

Taxonomy effectively started with Carl von Linne - Linnaeus (1707-1778). It (taxonomy) is actively used in many different disciplines.
 
Yes, this is homework. How do I move it?

jim mcnamara said:
Why must older concepts have shortcomings?

I'd tend to agree, but the question implies that there is. You can see no weaknesses of the taxonomy?
 
Ok. Taxonomic systems are based on generalizations. Flynn's taxonomy works for the von Neumann model or Havard model cpu and extensions thereof. It does not include other system models, like quantum computing, IMO.

[rant]
Note the IMO - this is exactly the point with a taxonomic system. It is a generalization. So you get to work out how to apply it to some extent at least. However, I've been in the field for quite a few years - everything I've worked with, overseen the purchase of - fits perfectly. So, from a practical standpoint, today, Flynn works darned well. Practical == I can buy a tested, debugged, working box at a non-astronomical cost.
[/rant]

Seriously:
What do your lecture notes say? I would go with whatever slant your prof takes. As I said, the IMO factor counts for something in this endeavor.
 
joel amos said:
Yes, this is homework. How do I move it?
You can't, but you can ask a mentor to do so, by clicking the Report button. In this case, however, I think that since this is more of a conceptual sort of question, the best place for this thread is right here.
 
  • Like
Likes   Reactions: jim mcnamara