Jack Battiste
Forum Replies Created
-
The Cinebench 15 database (cbscores link posted below) isn’t a perfect fit for everybody, as we have a wide variety of creative professionals here, but it might be useful for some. Workstations are benchmarked in terms of CPU and GPU performance, then the digits are stored online for comparison.
Worth a mention that the top render score from all workstations running OSX is listed as a Xeon X5680. If I am not mistaken, dual X5680 CPUs in a Mac … is probably from a common upgrade kit circa 2009-2012. There were (and still are) vendors offering to upgrade an 8 core “Nehalem” into a 12 Core “Westmere” at 333-3.46 GHz.
Jack Battiste
http://www.BattisteCreative.com
Retouching, Illustration & CGI -
^^ Sounds good. I actually use both RAID-5 and RAID-10. My workstation houses A four drive RAID-10 internally which store my current projects. In the event of a drive failure I want the quickest re-build possible with minimal deadline busting downtime. I have two 4 drive RAID-5 arrays for hourly back ups (OWC QX2 connected via PCIe SSD port). One BackUp is always running, the other I update once per week and store off-studio as theft and fire insurance. So that’s three copies of everything. Perhaps a bit over the top by many standards but works for me!
Jack Battiste
http://www.BattisteCreative.com
Retouching, Illustration & CGI -
Hi Tangier, First time caller, long time listener.
1. I don’t buy into the mindset that one type of RAID level is better than the other, because the “best” way to configure a RAID array, always depends upon specific needs. More importantly, I have noticed that folks tend to fall into two camps, those that are picturing worst case scenarios, and those who are looking to squeeze out the very last drop of performance, or capacity, or available budget. Thus both creative professionals and IT staff alike are confronted with two uncomfortable options, more money now or a new resume later.
2. Somebody who is looking to handle a worst case scenario, and evaluating risk management will likely favor mirrored arrays over parity arrays. Parity fans are quick to offer up Raid 6 as a quick counter punch, but let’s examine that more closely. RAID-6 offers insurance on a 2 drive failure, while Raid 10 offers up to 50% drive failures. With 4 drives this not significant. As the number of drives increases though, the differences here become quickly apparent. A 12 drive RAID-6 offers insurance for a maximum of two drive failures, while keeping the volume of 10 drives. A 12 spindle RAID-10 will offer insurance on a maximum of 6 dead drives, but is doing so by taking a 50% hit on capacity. Vocal advocates for RAID-6 will be looking at the 50% capacity hit and will recoil in horror. Might even point out that having “six spare tires” only works as long as you get flats on the right six. Meanwhile the President and CEO of the RAID-10 Fan Club is quite happy to Mad Maxing it with 6 spare tires, and will point to the mathematical probabilities of unrecoverable read errors when rebuilding parity arrays with when the drives are 3TB or larger.
3. For RAID 5 & 6, a reasonable approach is favor enterprise grade drives with URE (unrecoverable read error) rates of 1 read error for every 10^15, or 10^16. RAID-10 is more forgiving when using lower end drives because URE’s do not effect RAID-10 during a rebuild, you just end up with a bad block, not a destroyed array. For this reason, cheaper drives will often be used to offset the cost impact of the lower space utilization. Matt Simmons has summarized this point quite succinctly :
Quote : “ We started off with no one knowing what RAID was, then we went to everyone knowing what RAID was (or at least, being familiar enough with it to mistake it for being a backup). RAID-5 was sweet, because it had a parity stripe, and if you had a bunch of drives shoved together, it could tolerate the loss of a drive. How awesome was that? Then Robin Harris went and wrote Why RAID-5 Stops Working in 2009. It was an excellent article, and it was exactly right for the purposes that Robin meant it. It’s been a good way to point out to people that maybe they shouldn’t have RAID-5 configurations with 40TB worth of the cheapest SATA drives known to man. (continued)
RAID-5 has a parity, so that any individual pieces of data can be lost, and instead of recovering the data by copying it, it’s recalculated by examining the remaining data bits. So when we encounter a URE during normal RAID operations, the array calculates what the missing data was, the data is re-written so we’ll have it next time, and the array carries on business as usual. But when a drive dies, we have to replace it, and that’s when things get hairy. In order to rebuild the array, the new drive needs to be populated, and in order to do that, the entire contents of the remaining drives need to be read, in order to calculate the parity information. Assuming we have a RAID-5 array that has 3 3-TB disks, we’re now reading 6 terabytes of information. What is the statistical likelihood of encountering a URE? 1 in 2. A coin-flip. Have a RAID-5 array with 4 3-TB disks? That’s 1 in 1, almost certainly a failure. You can see how quickly this goes downhill. Now, a lot of people see this, freak out, and say “oh my god, I’m never using RAID-5 again! RAID-5 is the devil! It’s EVIL!”, but remember what is driving the numbers…it’s the URE rate.”
—Matt Simmons, I come not to Praise RAID-5 … But Not To Bury It Either
https://www.standalone-sysadmin.com/blog/2012/08/i-come-not-to-praise-raid-5/Hope this is useful!
Jack Battiste