Ask HN: Cheapest way to store 6TB online?
5 comments
Seems like Hetzner might be a great fit for you: http://www.hetzner.de/en/hosting/produkte_rootserver/ex4s
60 euros buys you 2x3tb of storage. It's raid 1 but I guess you could ask to have it raid 0.
60 euros buys you 2x3tb of storage. It's raid 1 but I guess you could ask to have it raid 0.
Raid 0 on 6TB is a terrifying prospect.
How big are your objects? 6TB in AWS S3 is about $720/month ...
If you are near a larger city, I would look into colocation hosting. Buy the hardware up front, and get some big disks. RAID or ZFS for redundancy.
Thanks all for the leads. Hetzner are seriously low priced, impressive.
In the end we've gone for co-location, we tried to avoid this due to the hassles we've had in the past around physical access.
I guess this workload isn't a great fit for the cloud just yet. Maybe some day!
In the end we've gone for co-location, we tried to avoid this due to the hassles we've had in the past around physical access.
I guess this workload isn't a great fit for the cloud just yet. Maybe some day!
I haven't checked, but I would think Amazon Glacier, assuming you only need to long-term archive the data.
Did you even read his post?
This is a key-value DB that he needs 1/2 sec response time to. Glacier is primarily a 'vault' type service where your access time is measured in hours.
This is a key-value DB that he needs 1/2 sec response time to. Glacier is primarily a 'vault' type service where your access time is measured in hours.
No! This crappy iOS app has a glitch and only showed the “cheapest way to store 6TB online” part!
That’s amused me somewhat chuckles
That’s amused me somewhat chuckles
Perhaps you missed the "< 500ms request time and 200 concurrent req/s" part of it??
The data is accessed via random lookups and needs to maintain <0.5s service times per lookup and can expect around 200 concurrent accesses at any time.
It's currently hosted in a small Cassandra cluster and that gives lookup times of 15ms-20ms, well within acceptable limits. However there's no reason it couldn't be moved to another indexing system (it could even be moved to a suitably tuned filesystem with little effort).