| 1 |
On Fri, 28 Apr 2006, kashani wrote: |
| 2 |
|
| 3 |
> This infatuation with the Google stuff that very few people (ie none of |
| 4 |
> us) have the in house infrastructure to handle or the available cash is |
| 5 |
> useless. Unless someone has actually built their own mini Google and |
| 6 |
> wants to tell us all about it with nice numbers like total cost, source |
| 7 |
> code, transactions per second, cost per GB of user data, throughput, and |
| 8 |
> other data points. |
| 9 |
|
| 10 |
The Director of Engineering at Google came and gave a talk about this at |
| 11 |
the NY Linux User Group about a year ago. My comments were based on his |
| 12 |
presentation. I certainly am not "infatuated" with Google but its |
| 13 |
interesting to hear about approaches to handling very large data sets. |
| 14 |
|
| 15 |
|
| 16 |
|
| 17 |
-- |
| 18 |
A |
| 19 |
-- |
| 20 |
gentoo-server@g.o mailing list |
Archives