If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
"Backblaze Hard Drive Stats Q1 2019"
https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our cloud storage ecosystem spread across three data centers. Of that number, there were 1,913 boot drives and 104,325 data drives. This review looks at the Q1 2019 and lifetime hard drive failure rates of the data drive models currently in operation in our data centers and provides a handful of insights and observations along the way. In addition, we have a few questions for you to ponder near the end of the post. As always, we look forward to your comments." Uh oh, those 12 TB Seagates are trending up. I suspect that they are not helium drives. I wish that they had more WDC drives but apparently they are driven by economics and WDC drives are definitely more expensive than Seagate drives. I do not think that HGST drives are old WDC drives but I could be wrong. Lynn |
#2
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
Lynn McGuire wrote:
"Backblaze Hard Drive Stats Q1 2019" https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our cloud storage ecosystem spread across three data centers. Of that number, there were 1,913 boot drives and 104,325 data drives. This review looks at the Q1 2019 and lifetime hard drive failure rates of the data drive models currently in operation in our data centers and provides a handful of insights and observations along the way. In addition, we have a few questions for you to ponder near the end of the post. As always, we look forward to your comments." Uh oh, those 12 TB Seagates are trending up. I suspect that they are not helium drives. I wish that they had more WDC drives but apparently they are driven by economics and WDC drives are definitely more expensive than Seagate drives. I do not think that HGST drives are old WDC drives but I could be wrong. Lynn ST12000NM0007 Some have an unblemished cover. https://images-na.ssl-images-amazon....L._SL1500_.jpg Some have a circular addition, as if the cover had a hole in it for some process step, and needed a cover afterwards. Which seems absurd, considering how much trouble they go to, to seal the top of the drive. It would not be like them to make "decal decorations" for arbitrary reasons, so that has some sort of function. https://images-na.ssl-images-amazon....ML._SX425_.jpg Both units have the same part number. Some references are made to "Malaysia", as if more than one plant makes them. This is a picture of what is underneath the cover. There are two seating planes. The inner plane is for the adhesive seal, the outer plane is for the welded cover (so the user cannot put mechanical stress on the sealing surface). And you can see the black thing in the lower right corner of the picture, might line up with that circular thing above it. I count eight platters. On the upper left corner, you can see the filter pak for particulate. There are no gas-flow shaping barriers. https://microless.com/cdn/products/f...30fdde8-hi.jpg Based on design, it's a helium drive. If it was air-filled, there would be a breather hole. Those platters would also be "thinner that regular platters" The more platters = more trouble. Paul |
#3
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
On 4/30/2019 5:18 PM, Paul wrote:
Lynn McGuire wrote: "Backblaze Hard Drive Stats Q1 2019" Â*Â* https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our cloud storage ecosystem spread across three data centers. Of that number, there were 1,913 boot drives and 104,325 data drives. This review looks at the Q1 2019 and lifetime hard drive failure rates of the data drive models currently in operation in our data centers and provides a handful of insights and observations along the way. In addition, we have a few questions for you to ponder near the end of the post. As always, we look forward to your comments." Uh oh, those 12 TB Seagates are trending up.Â* I suspect that they are not helium drives. I wish that they had more WDC drives but apparently they are driven by economics and WDC drives are definitely more expensive than Seagate drives.Â* I do not think that HGST drives are old WDC drives but I could be wrong. Lynn ST12000NM0007 Some have an unblemished cover. https://images-na.ssl-images-amazon....L._SL1500_.jpg Some have a circular addition, as if the cover had a hole in it for some process step, and needed a cover afterwards. Which seems absurd, considering how much trouble they go to, to seal the top of the drive. It would not be like them to make "decal decorations" for arbitrary reasons, so that has some sort of function. https://images-na.ssl-images-amazon....ML._SX425_.jpg Both units have the same part number. Some references are made to "Malaysia", as if more than one plant makes them. This is a picture of what is underneath the cover. There are two seating planes. The inner plane is for the adhesive seal, the outer plane is for the welded cover (so the user cannot put mechanical stress on the sealing surface). And you can see the black thing in the lower right corner of the picture, might line up with that circular thing above it. I count eight platters. On the upper left corner, you can see the filter pak for particulate. There are no gas-flow shaping barriers. https://microless.com/cdn/products/f...30fdde8-hi.jpg Based on design, it's a helium drive. If it was air-filled, there would be a breather hole. Those platters would also be "thinner that regular platters" The more platters = more trouble. Â*Â* Paul Thanks ! Is that 8 platters ? wow ! Lynn |
#4
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
Lynn McGuire wrote:
On 4/30/2019 5:18 PM, Paul wrote: Lynn McGuire wrote: "Backblaze Hard Drive Stats Q1 2019" https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our cloud storage ecosystem spread across three data centers. Of that number, there were 1,913 boot drives and 104,325 data drives. This review looks at the Q1 2019 and lifetime hard drive failure rates of the data drive models currently in operation in our data centers and provides a handful of insights and observations along the way. In addition, we have a few questions for you to ponder near the end of the post. As always, we look forward to your comments." Uh oh, those 12 TB Seagates are trending up. I suspect that they are not helium drives. I wish that they had more WDC drives but apparently they are driven by economics and WDC drives are definitely more expensive than Seagate drives. I do not think that HGST drives are old WDC drives but I could be wrong. Lynn ST12000NM0007 Some have an unblemished cover. https://images-na.ssl-images-amazon....L._SL1500_.jpg Some have a circular addition, as if the cover had a hole in it for some process step, and needed a cover afterwards. Which seems absurd, considering how much trouble they go to, to seal the top of the drive. It would not be like them to make "decal decorations" for arbitrary reasons, so that has some sort of function. https://images-na.ssl-images-amazon....ML._SX425_.jpg Both units have the same part number. Some references are made to "Malaysia", as if more than one plant makes them. This is a picture of what is underneath the cover. There are two seating planes. The inner plane is for the adhesive seal, the outer plane is for the welded cover (so the user cannot put mechanical stress on the sealing surface). And you can see the black thing in the lower right corner of the picture, might line up with that circular thing above it. I count eight platters. On the upper left corner, you can see the filter pak for particulate. There are no gas-flow shaping barriers. https://microless.com/cdn/products/f...30fdde8-hi.jpg Based on design, it's a helium drive. If it was air-filled, there would be a breather hole. Those platters would also be "thinner that regular platters" The more platters = more trouble. Paul Thanks ! Is that 8 platters ? wow ! Lynn 8 by 1.5TB would give you a 12TB drive. At a guess, 1.5TB tech would be PMR. Whereas 6 by 2TB could build a 12TB drive as well. I think 2TB is achieved by SMR for writes. And then the write performance wouldn't be as steady and dependable. They don't always give platter info, so finding a picture is an instant way to get proof. Paul |
#5
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
On Tuesday, April 30, 2019 at 6:18:47 PM UTC-4, Paul wrote:
Lynn McGuire wrote: "Backblaze Hard Drive Stats Q1 2019" https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our ://microless.com/cdn/products/f18e4964c96110fb07f12102f30fdde8-hi.jpg Based on design, it's a helium drive. If it was air-filled, there would be a breather hole. Those platters would also be "thinner that regular platters" The more platters = more trouble. Paul What is the advantage of helium, other than less air resistance and I guess less heat transfer? RL |
#6
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
RayLopez99 wrote:
On Tuesday, April 30, 2019 at 6:18:47 PM UTC-4, Paul wrote: Lynn McGuire wrote: "Backblaze Hard Drive Stats Q1 2019" https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our ://microless.com/cdn/products/f18e4964c96110fb07f12102f30fdde8-hi.jpg Based on design, it's a helium drive. If it was air-filled, there would be a breather hole. Those platters would also be "thinner that regular platters" The more platters = more trouble. Paul What is the advantage of helium, other than less air resistance and I guess less heat transfer? RL It has reduced air resistance. And heat transfer is improved. "The thermal conductivity of gases. The value of thermal conductivity for most gases and vapors range between 0.01 and 0.03 W/mK at room temperature. Notable exceptions are Helium (0.15) and Hydrogen (0.18)" I don't think there is any plan to switch to Hydrogen. If you get a datasheet for drives, you can kinda tell which drives have Helium, just from the operating power numbers. Reads MTBF 12TB 7.0W 2,500,000 10TB 7.1W 2,500,000 8TB 7.1W 2,500,000 ---- ---- --------- 6TB 9.3W 2,000,000 4TB 9.0W 2,000,000 2TB 7.4W 2,000,000 1TB 7.4W 2,000,000 I've not read any descriptions of what happens when the drive runs out of Helium. It's guaranteed to have Helium for five years. I presume the flying height changes. Paul |
#7
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
On 5/3/2019 11:46 PM, Paul wrote:
RayLopez99 wrote: On Tuesday, April 30, 2019 at 6:18:47 PM UTC-4, Paul wrote: Lynn McGuire wrote: "Backblaze Hard Drive Stats Q1 2019" Â*Â* https://www.backblaze.com/blog/backb...stats-q1-2019/ "As of March 31, 2019, Backblaze had 106,238 spinning hard drives in our ://microless.com/cdn/products/f18e4964c96110fb07f12102f30fdde8-hi.jpg Based on design, it's a helium drive. If it was air-filled, there would be a breather hole. Those platters would also be "thinner that regular platters" The more platters = more trouble. Â*Â*Â* Paul What is the advantage of helium, other than less air resistance and I guess less heat transfer? RL It has reduced air resistance. And heat transfer is improved. Â*Â* "The thermal conductivity of gases. Â*Â*Â* The value of thermal conductivity for most gases and vapors Â*Â*Â* range between 0.01 and 0.03 W/mK at room temperature. Notable Â*Â*Â* exceptions are Helium (0.15) and Hydrogen (0.18)" I don't think there is any plan to switch to Hydrogen. If you get a datasheet for drives, you can kinda tell which drives have Helium, just from the operating power numbers. Â*Â*Â*Â*Â*Â*Â* ReadsÂ* MTBF Â* 12TBÂ* 7.0WÂ*Â* 2,500,000 Â* 10TBÂ* 7.1WÂ*Â* 2,500,000 Â*Â* 8TBÂ* 7.1WÂ*Â* 2,500,000 Â* ----Â* ----Â*Â* --------- Â*Â* 6TBÂ* 9.3WÂ*Â* 2,000,000 Â*Â* 4TBÂ* 9.0WÂ*Â* 2,000,000 Â*Â* 2TBÂ* 7.4WÂ*Â* 2,000,000 Â*Â* 1TBÂ* 7.4WÂ*Â* 2,000,000 I've not read any descriptions of what happens when the drive runs out of Helium. It's guaranteed to have Helium for five years. I presume the flying height changes. Â*Â* Paul Sure, we could use hydrogen for hard drive cooling. Older power plant generators use pure hydrogen gas for cooling. With a very involved purging and oil based sealing process. Of course, hydrogen filled hard drives could bring back excitement to computing when they have a containment breach. Lynn |
#8
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
Lynn McGuire wrote:
Sure, we could use hydrogen for hard drive cooling. Older power plant generators use pure hydrogen gas for cooling. With a very involved purging and oil based sealing process. Of course, hydrogen filled hard drives could bring back excitement to computing when they have a containment breach. Lynn I've been burned by hydrogen. I have "experience" :-( Unfortunately, the Hydrogen reacts with some of the metals, whereas the Helium is nicely inert. Hydrogen is also exothermic upon expansion, so if a hole develops in a high pressure cylinder, the stream of gas should self ignite from the heat. Many years ago, they experimented with zeolite filled cylinders, to reduce the rate of gas escape from high pressure hydrogen cylinders. You could shoot a cylinder with a high power rifle, and the gas stream was gentle enough (because of the zeolite rate limiting) to not catch fire. Until someone smoking a cigarette sets it off, of course. Then if you're in a contained area, look out. If you order a cylinder of hydrogen from Liquid Air, of course that won't have any rate limiting. Just whatever your gas regulator might apply. The zeolite experiment was intended as a solution for carrying the gas in a moving vehicle, as a fuel store. Paul |
#9
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
On 5/7/2019 2:51 AM, Paul wrote:
Lynn McGuire wrote: Sure, we could use hydrogen for hard drive cooling.Â* Older power plant generators use pure hydrogen gas for cooling.Â* With a very involved purging and oil based sealing process.Â* Of course, hydrogen filled hard drives could bring back excitement to computing when they have a containment breach. Lynn I've been burned by hydrogen. I have "experience" :-( Unfortunately, the Hydrogen reacts with some of the metals, whereas the Helium is nicely inert. Hydrogen is also exothermic upon expansion, so if a hole develops in a high pressure cylinder, the stream of gas should self ignite from the heat. Many years ago, they experimented with zeolite filled cylinders, to reduce the rate of gas escape from high pressure hydrogen cylinders. You could shoot a cylinder with a high power rifle, and the gas stream was gentle enough (because of the zeolite rate limiting) to not catch fire. Until someone smoking a cigarette sets it off, of course. Then if you're in a contained area, look out. If you order a cylinder of hydrogen from Liquid Air, of course that won't have any rate limiting. Just whatever your gas regulator might apply. The zeolite experiment was intended as a solution for carrying the gas in a moving vehicle, as a fuel store. Â*Â* Paul Hydrogen just wants to be free ! I used to buy 30 to 50 high pressure cylinders of hydrogen from Air Liquide a month. You've never lived until your cylinder farm rots through the piping and vents to atmosphere one day. Lynn |
#10
|
|||
|
|||
"Backblaze Hard Drive Stats Q1 2019"
On 05/07/2019 3:34 PM, Lynn McGuire wrote:
On 5/7/2019 2:51 AM, Paul wrote: Lynn McGuire wrote: Sure, we could use hydrogen for hard drive cooling.Â* Older power plant generators use pure hydrogen gas for cooling.Â* With a very involved purging and oil based sealing process.Â* Of course, hydrogen filled hard drives could bring back excitement to computing when they have a containment breach. Lynn I've been burned by hydrogen. I have "experience" :-( Unfortunately, the Hydrogen reacts with some of the metals, whereas the Helium is nicely inert. Hydrogen is also exothermic upon expansion, so if a hole develops in a high pressure cylinder, the stream of gas should self ignite from the heat. Many years ago, they experimented with zeolite filled cylinders, to reduce the rate of gas escape from high pressure hydrogen cylinders. You could shoot a cylinder with a high power rifle, and the gas stream was gentle enough (because of the zeolite rate limiting) to not catch fire. Until someone smoking a cigarette sets it off, of course. Then if you're in a contained area, look out. If you order a cylinder of hydrogen from Liquid Air, of course that won't have any rate limiting. Just whatever your gas regulator might apply. The zeolite experiment was intended as a solution for carrying the gas in a moving vehicle, as a fuel store. Â*Â*Â* Paul Hydrogen just wants to be free ! I used to buy 30 to 50 high pressure cylinders of hydrogen from Air Liquide a month.Â* You've never lived until your cylinder farm rots through the piping and vents to atmosphere one day. Lynn On a small scale, back when I was an automotive mechanic in the dead of winter I had put a fast charger on a cars dead battery outside near the fence where the ac plugs where, after about an hour I went out to shut down and remove the charger in the dark, Inadvertently I must have hit the metal hold down and the plus+ terminal with the charger clip and caused a spark. Well all that freshly produced hydrogen in the empty space in the battery blew and threw parts of the battery top and a good batch of sulfuric acid in my face, I ran like hell back into the garage and grabbed the water hose and doused my face and front with plenty of cold water. For the first time ever I was lucky I wore glasses and did not get any in my Eyes. Since then I have had a great respect for hydrogen gas. Rene |
|
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
"Backblaze Hard Drive Stats for 2017" | Lynn McGuire[_3_] | Storage (alternative) | 1 | February 21st 18 03:41 AM |
Backblaze, "Hard Drive Stats for Q3 2017" | Lynn McGuire[_3_] | Storage (alternative) | 3 | October 27th 17 02:13 AM |
"Backblaze Hard Drive Stats for 2016" | Lynn McGuire[_3_] | Storage (alternative) | 4 | February 3rd 17 09:12 PM |
"One Billion Drive Hours and Counting: Q1 2016 Hard Drive Stats" | Lynn McGuire[_2_] | Storage (alternative) | 7 | May 22nd 16 07:30 AM |