Re: Semote storage pool
2003-02-20 11:56:16
>It doesn't matter how you use it. Forcing TSM to work its I/O at network
>speeds when it is designed to work at bus speeds will show badly. You'll
>be disappointed.
Mark - I wonder if that traditional distinction will prevail in the near
future. Network evolution has been accelerating, largely driven
by such storage needs, and so I think the disparity between bus and wire
will tend to lessen, and to perhaps conceptually merge into a single amalgam
of "Device Access Pathing". Parallel pathing is becoming rather passe,
with serial pathing coming to the fore with the speed made possible by its
simplicity in not having to wait for the last horse to pass the finish line
before declaring unit data arrival. Generalized networking will always
entail more overhead in its packetizing, but I think we'll see an increasing
number of network-based storage units with very acceptable performance as
10GB and higher speed networking arrangements arrive.
Richard Sims, BU at home, under 2' of latest snow arrival
|
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- Re: Semote storage pool, (continued)
- Re: Semote storage pool, Jozef Zatko
- Re: Semote storage pool, Stapleton, Mark
- Re: Semote storage pool, GUILLAUMONT Etienne
- Re: Semote storage pool, Stapleton, Mark
- Re: Semote storage pool, Remeta, Mark
- Re: Semote storage pool, GUILLAUMONT Etienne
- Re: Semote storage pool, Jozef Zatko
- Re: Semote storage pool,
Richard Sims <=
- Re: Semote storage pool, Stapleton, Mark
|
|
|