Did I say mirroring was parity-protected
LOL. You're punch drunk and can't even write straight anymore. This is what you said. Now you know somebody is at the losing end of an argument when he starts to argue endlessly on exactly how one should interpret what he wrote.
This is important, because all customers that buy data storage arrays, NAS or SANs, want parity-protection.
Either mirroring or some form of other RAID'd parity protection."
Why does EMC submit SFS numbers with RAID *OFF*?
The RAID Advisory Board considers striping as RAID 0 so how can it be RAID *OFF*? RAID levels are typically physically partitioned during the final assembly process so exactly how do you turn it off? Only a solitary voice of reason like you considers it RAID OFF. Can everybody else be wrong? LOL.
The benchmarks also measure availability Gus, if the submitting company actually submits that way.
Wait a minute. Let's use some common sense here. How in the world can a damn speed benchmark measure availability when it measures results in milliseconds?
Let's go to RAB for the authoritative word on this matter otherwise you're going to try to win this losing argument forever.
....RAID Levels basically describe how data and redundant data are mapped across the disks of an array; information which is less important than an understanding of just how much EDAP (Extended Data Availability & Protection) is provided by a storage system.
How Much EDAP?
The determination of the amount of EDAP capability to purchase is a function of application need. A few applications may require no EDAP capability. Others may require the highest level of EDAP capability. To select a disk system with an EDAP capability higher than needed for the intended applications is a waste of money. To select a disk system with an EDAP capability inadequate for the intended applications puts the user at risk of either losing vital data or of not being able to gain timely access to vital data.
raid-advisory.com
You can shout about NTAP's superior performance all you want and insist that everybody buy NAS on that basis alone despite the fact that any NAS product depends heavily on the availability of a congested network for its REAL-WORLD performance. Actually, the preference is that you do your shameless cheerleading on the NTAP board instead of the EMC board. In case you haven't noticed, the Celerra has been the fastest NAS device for 3 years straight; although, Celerra is high-end NAS which connects to high-end Symmetrix while NTAP's flagship clusters are still mid-range products.
In terms of performance, the IP4700 is the only product that directly competes with NTAP's F840s. It was just introduced last December. EMC tends to establish the ultra-reliability of its products before rapidly improving performance, typically with feedback from its large and diverse customer base so pre-judge at your own risk. In this case, the mid-range market is a fairly new one for EMC.
Again, you need to be more diligent and think for yourself by checking the history of RAID vendors whose RAID controller technology couldn't keep up with the rapid increases in capacity and functionality.
It requires a remarkably high level of self-deception to insist that a chassis which puts up to 168 disk drives behind one processor (F840) will always outperform a chassis which puts up to 100 disk drives behind four processors (IP4700).
By the way, I take it your silence on the matter is a tacit admission on your part that NTAP is way behind EMC in terms of integrating SAN and NAS?
Bye, bye NTAP parrot. My downloads are done. |
