(CS838: Topics in parallel computing, CS1221, Thu, Mar 4, 1999, 8:00-9:15 a.m.)

- Derive expression for the communication latency of an optimal
**noncombining one-to-all scatter**on a**wormhole all-port hypercube***Q*. Ignore the time overhead of packet combining and disassembling._{n} -
Derive expression for the communication latency of an optimal
**combining one-to-all scatter**on a**wormhole all-port mesh***M(z*if the source's address is_{1},z_{2},...,z_{n})*(a*._{1},...,a_{n})

Hint: Start with*n=2*or*n=3*and generalize to any*n*.

- the source has a packet of size
*M*for every other processor - the basic point-to-point latency of a packet of size
*m*over distance*d*ist(m,d)=t

where_{s}+dt_{d}+mt_{m}*t*._{m}\approx t_{d}<< t_{s}