Re: Efficient Sequential Access to Arrays

[tmoran@acm.org]

> - The data structure and algorithm is optimal from a machine operation
> perspective (if we consider the relative speeds of the machine code operations)
   Taking into account all the fancy pipelining and multiple arithmetic units
of the machine you are targeting?  How fast does this run in hand optimized
asm code?

>  If we instead store the index of the next element instead of the link, it
>  is slower because we have to use a relative addressing mode instead of an
> absolute one.
  How about storing not only links to other nodes, but have each node also
store a copy of its own index.

> to address the neighbours does not make any sense as division is slower
> than multiplication.
  That depends.  If the divisor is a power of two you can shift.