And then buy another external drive, 4TB or larger, and do it again. Preferably a different brand of drive inside, preferably a an enterprise storage drive. And keep it off-site.
Gopher was cool because it was way easier to follow references to other content on other sites than FTP, even if you often had to download the content and figure out which app was needed to display it. Xgopher was cooler yet because it could display images. Then Mosaic showed up, and had the ability to handle many more data types, and the fuse was lit...
In the early 90s, everyone who knew anything about the 'Net knew where the main gophers were. Opened up communications a lot, and was far more direct than Usenet, which had already descended into being mostly noise by then. Great way to share information, 'til there was a greater one.
A RAID0 of 3 SSDs will be faster than a single SSD drive for multiple reasons, the primary one being that the kernel reads from all three devices at the same time, and (secondarily) both the SSDs and the kernel are doing read-ahead, so that once hundreds (if not thousands) of sectors are in memory, you're only looking at the time to copy them into the destination buffer.
For more speed, set your read-ahead buffers up -/sys/devices/(device)/hostX/targetX:0:0/X:0:0:0/block/sde/queue/read_ahead_kb, in Linux...
Usually defaults to 128KB....
As a software engineer with 30 years under my belt, I'd answer "I find doing the work and solving the problems far more rewarding than managing a team". Being the software lead is fine; I've found that being management doesn't do it for me.
True enough.
Simplistically, if ETOH as a fuel is viable as a fuel (or additive), take away the government subsidy and let's see if it can make it on it's own.
I'm betting that it can't.
If ETOH were actually worth anything (i.e., didn't harm engines, was *really* energy balance positive, didn't put aldehydes into the atmosphere, cause food prices to go up, could be produced from cellulose, etc.) it could survive without a government subsidy. The only reason it's still lurching along, taking up 40% of the corn produced in the USA, is because the lobbyists, farmers and ETOH producers can continue to suck $$ from the US gummint.
I wouldn't mention the Samsungs in the same sentence as Dells running Sandisk SSDs - there's a world of difference between the Samsung and Sandisk SSDs. We found that the IDE interface on the Sandisks was being reset under moderate usage - which might account for the "hangs" you mention. Have yet to see any resets on the (IDE interface) Samsung drives.
From reading about the NAND flash technology, they've gone from the original handful of thousand writes to a million (expected) in a couple of years. With the write balancing, lifetimes of 10 years seem quite reasonable.
Been using arrays of 4 and 8 32GB SSDs as both RAID0 and RAID5, off hardware RAID controllers and as Linux softraid, to push seek time to near 0 and throughput as far as possible. Bottom line is, they're significantly faster than "real" disks. We've found MTrons to be faster than Samsungs, generally 20 to 40%, and the MTron seek times are significantly better (they probably don't write-balance check as often under heavy usage). Only reliability problems I had were with another brand (neither Samsung nor MTron).
Slashdot Mirror
And then buy another external drive, 4TB or larger, and do it again. Preferably a different brand of drive inside, preferably a an enterprise storage drive. And keep it off-site.
Gopher was cool because it was way easier to follow references to other content on other sites than FTP, even if you often had to download the content and figure out which app was needed to display it. Xgopher was cooler yet because it could display images. Then Mosaic showed up, and had the ability to handle many more data types, and the fuse was lit... In the early 90s, everyone who knew anything about the 'Net knew where the main gophers were. Opened up communications a lot, and was far more direct than Usenet, which had already descended into being mostly noise by then. Great way to share information, 'til there was a greater one.
A RAID0 of 3 SSDs will be faster than a single SSD drive for multiple reasons, the primary one being that the kernel reads from all three devices at the same time, and (secondarily) both the SSDs and the kernel are doing read-ahead, so that once hundreds (if not thousands) of sectors are in memory, you're only looking at the time to copy them into the destination buffer. For more speed, set your read-ahead buffers up - /sys/devices/(device)/hostX/targetX:0:0/X:0:0:0/block/sde/queue/read_ahead_kb, in Linux...
Usually defaults to 128KB....
As a software engineer with 30 years under my belt, I'd answer "I find doing the work and solving the problems far more rewarding than managing a team". Being the software lead is fine; I've found that being management doesn't do it for me.
Let's see - which might be more valuable: the car, or your life? Put the car in neutral and let the engine blow up (or sit at it's rev limit).
True enough. Simplistically, if ETOH as a fuel is viable as a fuel (or additive), take away the government subsidy and let's see if it can make it on it's own. I'm betting that it can't.
If ETOH were actually worth anything (i.e., didn't harm engines, was *really* energy balance positive, didn't put aldehydes into the atmosphere, cause food prices to go up, could be produced from cellulose, etc.) it could survive without a government subsidy. The only reason it's still lurching along, taking up 40% of the corn produced in the USA, is because the lobbyists, farmers and ETOH producers can continue to suck $$ from the US gummint.
I wouldn't mention the Samsungs in the same sentence as Dells running Sandisk SSDs - there's a world of difference between the Samsung and Sandisk SSDs. We found that the IDE interface on the Sandisks was being reset under moderate usage - which might account for the "hangs" you mention. Have yet to see any resets on the (IDE interface) Samsung drives.
The ones I've used have a "secure erase" SCSI command available. Probably not going to be used often by Joe User.
From reading about the NAND flash technology, they've gone from the original handful of thousand writes to a million (expected) in a couple of years. With the write balancing, lifetimes of 10 years seem quite reasonable.
Been using arrays of 4 and 8 32GB SSDs as both RAID0 and RAID5, off hardware RAID controllers and as Linux softraid, to push seek time to near 0 and throughput as far as possible. Bottom line is, they're significantly faster than "real" disks. We've found MTrons to be faster than Samsungs, generally 20 to 40%, and the MTron seek times are significantly better (they probably don't write-balance check as often under heavy usage). Only reliability problems I had were with another brand (neither Samsung nor MTron).