Intel or PowerPC for RAID controller

Dear ALL

Please discuss pro & conts on using Intel vs PowerPC based RAID
controlers ... (e.g. some vendors use PPC750 and some use IOP303 or
newer) ...

Other query: while Intel ships newer IO Procesors (recently IOP321 and
IOP331) many vendors of raid array still use IO303 outdated processor
(& chipset) and even i960 ... Why?

Please comment: performance, reliability, applicability etc ...

Thanks for your time!

Cheers,

AL

(albatros66) writes:

Please discuss pro & conts on using Intel vs PowerPC based RAID
controlers ... (e.g. some vendors use PPC750 and some use IOP303 or
newer) ...

Cost, form factor, familiarity of developers, features; take your pick.

(albatros66) wrote in message om...
Dear ALL

Please discuss pro & conts on using Intel vs PowerPC based RAID
controlers ... (e.g. some vendors use PPC750 and some use IOP303 or
newer) ...

Other query: while Intel ships newer IO Procesors (recently IOP321 and
IOP331) many vendors of raid array still use IO303 outdated processor
(& chipset) and even i960 ... Why?

Please comment: performance, reliability, applicability etc ...

I assume you are talking about some sort of IDE/ATA-SCSI/FC RAID.
Basic advice: Intel vs. PowerPC is going to be the last of your
concerns. The real issues come down to:

-Quality of chassis construction (vibration, cooling, etc),
electronics (do the people who designed the circuit boards have any
idea what they are doing?), and electrics (what kind of $15 power
supply did they use?)
-Firmware and, in particular, error and bad block handling by firmware
-Quality of drives (new WD Raptors SATA drives are supposed to be good
but unproven and not cheap)

In summary, the key issues are quality of hardware and software. If
you can find two boxes that don't loose your data, then you might
worry about performance.

[This e-mail address is not accepting e-mail]

Benjamin Goldsteen wrote:

Other query: while Intel ships newer IO Procesors (recently IOP321 and
IOP331) many vendors of raid array still use IO303 outdated processor
(& chipset) and even i960 ... Why?

In summary, the key issues are quality of hardware and software. If
you can find two boxes that don't loose your data, then you might
worry about performance.

Also look at what happens at power loss. A 512 MB write-back cache with
no battery backup (AXUS why can't you deliver that battery backup option
mentioned on your website?) will surely mean data loss if the power goes
out.


Thomas

On 27 Oct 2003 06:08:47 -0800, (albatros66) wrote:

Dear ALL

Please discuss pro & conts on using Intel vs PowerPC based RAID
controlers ... (e.g. some vendors use PPC750 and some use IOP303 or
newer) ...

Why do you care?

Other query: while Intel ships newer IO Procesors (recently IOP321 and
IOP331) many vendors of raid array still use IO303 outdated processor
(& chipset) and even i960 ... Why?

Why not?

Look, these are embedded controllers, not general purpose CPUs. Based
on the desired performance and behavior (and their own abilities), the
engineering designers select a processor for the job. It really
doesn't matter what it is, *provided* that it does the job.

In many cases, using as faster (or newer, which is not the same thing)
controller would have minimal benefits because e.g. the performance is
bottlenecked by something else (often an ASIC or FPGA).

AL

Malc.

In article ,
Malcolm Weir wrote:

In many cases, using as faster (or newer, which is not the same thing)
controller would have minimal benefits because e.g. the performance is
bottlenecked by something else (often an ASIC or FPGA).

Actually, the usual performance bottleneck is lousy firmware that does
massive numbers of unnecessary data copies and saturates the card's
internal memory bandwidth. A careful comparison of the three or four
*extremely* similar 80303-based PCI RAID cards on the market should
suffice to teach one about whose firmware is Nothing To Write Home About
(and, therefore, whose cards should probably be avoided no matter how much
hardware they throw at a given generation of controller).

--
Thor Lancelot Simon
But as he knew no bad language, he had called him all the names of common
objects that he could think of, and had screamed: "You lamp! You towel! You
plate!" and so on. --Sigmund Freud

Anton Rang wrote in message ...
(albatros66) writes:

Please discuss pro & conts on using Intel vs PowerPC based RAID
controlers ... (e.g. some vendors use PPC750 and some use IOP303 or
newer) ...

Cost, form factor, familiarity of developers, features; take your pick.

To take or not to take, that's the question ;-)

* Cost are not much different: varying from 12000 USD to 18000USD for
4TB subsystem ...

* familiarity of developers: I'm not developer. I'm "end user" ( not a
user at gray end 8-}). I don't need to care about

* features - most of the subsystem have almoust the same features -
the lists of features are long or very long: ten or more RAID modes,
flashilg lamps, SNMP, hot everything etc etc ...

Let's take your pick again: which SCSI/FIBRE-to-IDE/SATA array is best
from any point of view ??? Is anybody brave enough to answer the
question??? I have to answer to my boss because whe are just buying
few such devices (some 20TB) ...

AL

(Benjamin Goldsteen) wrote in message . com...
(albatros66) wrote in message om...
Dear ALL

Please discuss pro & conts on using Intel vs PowerPC based RAID
controlers ... (e.g. some vendors use PPC750 and some use IOP303 or
newer) ...

Other query: while Intel ships newer IO Procesors (recently IOP321 and
IOP331) many vendors of raid array still use IO303 outdated processor
(& chipset) and even i960 ... Why?

Please comment: performance, reliability, applicability etc ...

I assume you are talking about some sort of IDE/ATA-SCSI/FC RAID.

Are you a prophet or so ? Indeed I was thinking about such subsystems
but I can say, that also well know vendors of SCSI-to-SCSI or FC-to-FC
arrays relies on these chipsets ...

Basic advice: Intel vs. PowerPC is going to be the last of your
concerns. The real issues come down to:

-Quality of chassis construction (vibration, cooling, etc),
electronics (do the people who designed the circuit boards have any
idea what they are doing?), and electrics (what kind of $15 power
supply did they use?)
-Firmware and, in particular, error and bad block handling by firmware
-Quality of drives (new WD Raptors SATA drives are supposed to be good
but unproven and not cheap)

In summary, the key issues are quality of hardware and software. If
you can find two boxes that don't loose your data, then you might
worry about performance.

I agree with all above, but ...

* Do you think, that internal bus with frequency e.g. 33MHz (or 66MHz)
is OK? The newer chips use 133MHz bus (or more ). Note that with
enclosure containing 16 SATA drives it becomes feasible to get trasfer
rates upt to 300Mbytes/s or more (I know about boxes with RAID5 rates:
200MBytes/s write and 300MBytes/s read)
* newer chipsets have the appropriate asic on board which works much
faster
* newer chipsets suport more cache memory and aslo offer cache memory
battery backup ...


Cheers,

AL

A careful comparison of the three or four
*extremely* similar 80303-based PCI RAID cards on the market should
suffice to teach one about whose firmware is Nothing To Write Home About
(and, therefore, whose cards should probably be avoided no matter how much
hardware they throw at a given generation of controller).

Do you have any hits on this ? Which are good ? which are bad ???
How I (i.e. end user) can compare firmwares?

Cheers,

AL

In article ,
albatros66 wrote:
A careful comparison of the three or four
*extremely* similar 80303-based PCI RAID cards on the market should
suffice to teach one about whose firmware is Nothing To Write Home About
(and, therefore, whose cards should probably be avoided no matter how much
hardware they throw at a given generation of controller).

Do you have any hits on this ? Which are good ? which are bad ???
How I (i.e. end user) can compare firmwares?

I think you should do the research yourself and reach your own
conclusions -- but Adaptec, Mylex, ICP, and IIRC LSI make, or made, cards
with the 80303 and very similar SCSI bus configurations (a couple of these
even use the exact same SCSI interface chips). These cards are all $100 or
less on eBay now; you can see for yourself how each performs for your
application. Note that the LSI card will probably operate in both I2O and
"native" (a.k.a. "Quartz") mode and performance isn't the same in both cases;
it's practically got two different firmwares on it AFAICT.

Adaptec seems to be moving away from the I2O-based firmware they acquired
with DPT. From a driver writer's perspective, this is somewhat frustrating
since I2O is pretty efficient from a software point of view and it was nice
when multiple RAID vendors were using it; really increased bang for the
buck. But in my experience, the performance of that I2O firmware left a
*lot* to be desired; in a trivial benchmark reading the same sector over
and over again, for instance, all that beefy hardware on the 3210 could
clear only about 4,000 -- cached! -- IOPS; linear write performance was
quite unimpressive when compared to the nearly-identical Mylex AcceleRAID
352; the cache write policy was not tunable and seemed to not effectively
coalesce short writes from many filesystems. A shame; when DPT originally
rolled out the Milennium cards it seemed like great things were on the
way...

--
Thor Lancelot Simon
But as he knew no bad language, he had called him all the names of common
objects that he could think of, and had screamed: "You lamp! You towel! You
plate!" and so on. --Sigmund Freud