| 1 |
On 11/13/14 15:50, J. Roeleveld wrote: |
| 2 |
> |
| 3 |
> James, |
| 4 |
> |
| 5 |
> Backblaze is not selling harddrives. They sell storage on their servers. |
| 6 |
> The data they collect is based on different drives from different manufacturers. |
| 7 |
> They are quite open on what they use, check out their website and blogposts. You should |
| 8 |
> find the answers to your questions easily there. |
| 9 |
> Joost |
| 10 |
|
| 11 |
Wow, an honorable vendor, that uses "scientific" standards so other can |
| 12 |
readily collect similar data and verify their results. They even put |
| 13 |
the scientific method ahead of quarterly profit statements. Impressive. |
| 14 |
It's refreshing to know we still have vendors like this. One in a |
| 15 |
million. I never suggested they did anything specific like sell HD. |
| 16 |
|
| 17 |
There are a multitude or reasons to publish data that vendor A is better |
| 18 |
than vendor B. Furthermore, if an organization at that |
| 19 |
size/capitalization is not fully redundant, HA and cost effective, (1) |
| 20 |
they wont be in business very long and (2) it really does not matter |
| 21 |
because that sort of data only guides your next (corporate) bulk |
| 22 |
purchase of hardware, because *price* of component redundant hardware |
| 23 |
*always* trumps mtbf specs, imho. |
| 24 |
|
| 25 |
The important thing is for others to be able to independently verify |
| 26 |
their results. So the methods, tools and datasets should be readily |
| 27 |
available as a first step. Then when IBM, Google and dozens of others, |
| 28 |
including but not limited to universities, government labs (Los Alamos) |
| 29 |
and private users all collect similar data and discern similar |
| 30 |
conclusions, then you know the data is valid and the conclusions are |
| 31 |
just. But by then each of the vendors will have new product offerings; |
| 32 |
so once again, price drives the market and perceptions adjust the prices |
| 33 |
moderately to none at all. |
| 34 |
|
| 35 |
There are a myriad of factors that can affect these results. Independent |
| 36 |
verification is the gold standard for data analysis, imho. Not just |
| 37 |
"re-crunching numbers" from their datasets, but collecting your own |
| 38 |
independent datasets for analysis. I must have missed those links on the |
| 39 |
vendor's website? Enhancing the common, open source Network Management |
| 40 |
Systems such as nagios or jffnms, where others can readily collect data |
| 41 |
on drives themselves, and independently verify their conclusions is |
| 42 |
paramount for their published conclusions to hold a "scientific" basis |
| 43 |
of validity, imho. Otherwise, it's rather akin to "benchmarks", imho. |
| 44 |
|
| 45 |
Useful ? :: yes ; interesting ? :: most definitely ; |
| 46 |
conclusive ? :: not even close! |
| 47 |
|
| 48 |
|
| 49 |
peace, |
| 50 |
James |
Archives