Re: [Observium] SSD for better performance
The maths and specs suggest a modern SSD should survive observium workload for quite a while, but I'm still worried about the random failure rate and the likelyhood that individual cells get worn out.
In general SSDs are such a strange tech, there's almost no way to really improve them except with fancy controller hacks, the flash itself is a dead end tech.
Good luck :-)
Adam
Cushard Michael mcc@us.axway.com wrote:
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Well I see stable failure(i know sound weird) rate between 1 year and 3 years so far.
On 20.09.2012 0:33, Adam Armstrong wrote:
The maths and specs suggest a modern SSD should survive observium workload for quite a while, but I'm still worried about the random failure rate and the likelyhood that individual cells get worn out.
In general SSDs are such a strange tech, there's almost no way to really improve them except with fancy controller hacks, the flash itself is a dead end tech.
Good luck :-)
Adam
Cushard Michael mcc@us.axway.com wrote:
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Why is your silly mail client breaking reply headers?
OT: Curious to see how the SSD will hold up. :-)
On 19/09/2012 22:33, Adam Armstrong wrote:
The maths and specs suggest a modern SSD should survive observium workload for quite a while, but I'm still worried about the random failure rate and the likelyhood that individual cells get worn out.
In general SSDs are such a strange tech, there's almost no way to really improve them except with fancy controller hacks, the flash itself is a dead end tech.
Good luck :-)
Adam
Cushard Michael mcc@us.axway.com wrote:
Hi,
Data is dumped to spindle nightly for backups.
As with everything, the question isn't "if" it fails, it's "when". I'll report back when it does.
Regards,
Michael
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
participants (3)
-
Adam Armstrong
-
Nikolay Shopik
-
Tom Laermans