On Tue, 7 Nov 2000, Bruce Sterling Woodcock wrote:
Umm, I think you've been mislead by EMC then. The do *not* have the same number of CPUs. You see, the Celerra DMs all have one CPU
[...]
Anyway, if number of heads still bothers you, just remember this was the F760. With the F840, the number of heads needed would be reduced significantly.
[...]
I agree, but Netapp *is* the clear winnder in the strict ops/s game. The reason everyone doesn't buy Netapp is the other reasons you mention.
[...]
EMC has not mislead us, despite their efforts to the contrary (and, NetApp, too, to be equal to both vendors). ;) I have no doubt, and many others would agree as well, that NetApp's architecture is much more elegant than EMC's. What I was trying to emphasize were the "CPU" units that actually push data onto the network. For NetApp, these are filer heads; for EMC, these are datamovers. You are 100% correct that the disk storage is much more complicated, powerful, and expensive on the EMC side, but it is also more robust (built-in monitoring, battery backup, failover, ability to local attach, etc.).
Then why did you claim they were about the same, when the details show they aren't? The NTAP stuff is minor and only accounts for a few %
[...]
Netapp will eventually produce all the different protocol numbers, I'm sure.
Actually, I didn't view "all results" at the SPEC page; I see NetApp has published some TCP numbers. Hoo-ray!
Let's compare F840 TCP NFS v3 (7,783 ops/s) with Celerra 507 DataMover 2 CPU (they don't list a single node) at 15,723 ops/s. Giving NetApp the benefit of the doubt of perfect linearity, a cluster of F840s would then be 15,556 ops/s. Damn, that's close. And, that's one main point I was trying to make.
One issue we had to deal with with NetApp and EMC is scaling. For small systems, the Celerra is prohibitively expensive. You have to buy a gigantic Symmetrix and a Celerra frame, which is tons of money, and you haven't even put a single DataMover in it yet! Their y-intercept on an x-y graph of $ vs. performance is high. Yet, incremental additions (the slope) of datamovers can be much less expensive than a similarly licensed NetApp head.
With NetApp, to add a redundant (i.e., clustered) head (trying to keep the comparison half-way fair, since EMC has N+1 failover), you have to add them in pairs. Their intercept is zero (no money for no heads and no disk), yet their slope can be very high. Depending on discounts from each vendor (does anyone pay list price?!), you can imagine a situation in which the line for each vendor *crosses*, for some number of datamovers/filers and the same amount of disk, that is, the incremental cost in adding another pair of clustered F840s and some disk will become more expensive overall than just adding another datamover and the same amount of disk.
These "other" issues touch on the complexity of the architecture and the specific application, for us and for each customer. Performance is "about the same" so concentrate on the other issues. Another main point I was trying to make.
"Out of the box" configurations I would think are unlikely, since it would mean spending a lot of testing resources on them for very little reward, and could even be confusing.
I think, to most customers less savvy than those on this list, the "big" numbers are confusing and misleading. Can "Volvo" really out-slalom "BMW"? Volvo says they can in their commercials. That's what your average customer is going to hear from marketing, and what both EMC and NetApp are doing with the big numbers. However, it *is* misleading (and illegal in some European countries). Which Volvo? Which BMW? By how much? Internal combustion technology in most commercial cars is roughly the same. What differs is the number of cylinders, turbocharging, intercooling, the ECM chip, etc.
Providing more configurations would allow the customer to find the configuration closest to what they need, and hence get an accurate indication of how well either box would do for them.
And, would it not require less work to grab a filer, install ONTAP, keep the defaults, and run the SPEC program, rather than tweaking it to juice every last op out of the box? (More work, sure, in running the SPEC program a *second* time...)
EMC would certainly always quote the numbers that made NTAP look the worst,
And, NTAP has fired the last volley with the 16-node non-failover "cluster." All I'm saying is let's stop the dick waving to see whose is the biggest.
Let's hear some newsworthy news on dual-CPU filer heads (I see that extra plugged socket in there...), what's next in clustering, etc.
Until next time...
The Mathworks, Inc. 508-647-7000 x7792 3 Apple Hill Drive, Natick, MA 01760-2098 508-647-7001 FAX tmerrill@mathworks.com http://www.mathworks.com ---