Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Backblaze Drive Stats for 2024

backblaze.com

Post reply on HN
Print view View original

101–110 of 209 posts

Re: Backblaze Drive Stats for 2024

#101

Post by fc417fc802 »

echoangle wrote: post #79
userbinator wrote: post #67

Earlier quoted context omitted.

An amazing amount if you consider that 16EB is the amount of data a 64-bit quantity can address, and this is over a quarter of that.

Wow, that’s a cool stat. I wonder if people will ever seriously use 16EB of memory in a single system and will need to change to a more-than-64-bit architecture or if 64 bit is truly enough. This has „640k ought to be enough for anybody“ potential (and I know he didn’t say that).

> need to change to a more-than-64-bit architecture

Perhaps we don't need a single flat address space with byte-addressable granularity at those sizes?

I wonder how an 8 bit byte, 48 bit word system would have fared. 2*32 is easy to exhaust in routine tasks; 2*48 not so much.

Re: Backblaze Drive Stats for 2024

#102

Post by fc417fc802 »

remram wrote: post #98
malfist wrote: post #88

Earlier quoted context omitted.

Out of curiosity, what server cases can actually accommodate 50 drives?

Those are specialized NAS chassis. We have a number of them, 4U size, too heavy to move when the drives are in. edit: They look like this: https://knowledgebase.45drives.com/wp-content/uploads/2019/0... (image from ddg)

> too heavy to move when the drives are in

Careful not to drop it.

Re: Backblaze Drive Stats for 2024

#103

Post by manquer »

remram wrote: post #68
bhouston wrote: post #43

Based on the data, it seems they have 4.4 petabytes of storage under management. Neat. https://docs.google.com/spreadsheets/d/1E4MS84SbSwWILVPAgeIi...

Nowadays you can get a petabyte in a single machine (50 drives 20TB each).

Backblaze custom designed and uses what they call a pod holding 60units for a long time now

https://www.backblaze.com/cloud-storage/resources/storage-po...

Re: Backblaze Drive Stats for 2024

#104

Post by manquer »

malfist wrote: post #88
remram wrote: post #68

Earlier quoted context omitted.

Nowadays you can get a petabyte in a single machine (50 drives 20TB each).

Out of curiosity, what server cases can actually accommodate 50 drives?

https://www.backblaze.com/cloud-storage/resources/storage-po...

This is what backblaze use themselves currently able to hold 60 drives .

Re: Backblaze Drive Stats for 2024

#105

Post by jillyboel »

anotherhue wrote: post #92
jillyboel wrote: post #87

Earlier quoted context omitted.

you should still be able to saturate your bandwidth with poor latency

Not unless the protocol you use accounts for that. Smb for instance is tragic.

True, I was thinking of backup tools which tend to consider poor latency in their design.

Re: Backblaze Drive Stats for 2024

#106

Post by manquer »

crazygringo wrote: post #94
quintin wrote: post #49

It continues to surprise me why Backblaze still trades at a fraction of its peak COVID share price. A well-managed company with solid fundamentals, strong IP and growing.

I feel very confident that in 30 years AWS, Azure and Google Cloud will still be operating and profitable. I think there's a very small chance that Backblaze will be. Nothing against them, but it's virtually impossible to compete long-term with the economies of scale, bundling and network effects of the major cloud providers.

Cloud providers AWS in particular uses storage and transfer pricing as means of lock-in to other products , they can never be cost competitive to Backblaze , they has a thriving prosumer business .

I don’t think it is going anywhere.

Re: Backblaze Drive Stats for 2024

#107

Post by tharkun__ »

bragr wrote: post #96
CTDOCodebases wrote: post #72

Earlier quoted context omitted.

Take these stats with a grain of salt. I am becoming more and more convinced that hard drive reliability is linked to the batch more than to the individual drive models themselves. Often you will read online of people experiencing multiple failures from drives purchased from the same batch. I cannot prove this because I have no idea about Blackblazes procurement patterns but I bought one of the better drives in this…

>Often you will read online of people experiencing multiple failures from drives purchased from the same batch I'll toss in on that anecdata. This has happened to me a several times. In all these cases we were dealing with drives with more or less sequential serial numbers. In two instances they were just cache drives for our CDN nodes. Not a big deal, but I sure kept the remote hands busy those weeks trying to keep…

This has been the "conventional wisdom" for a very long time. Is this one of those things that get "lost with time" and every generation has to rediscover it?

Like, 25+ years ago I would've bought hard drives for just my personal usage in a software raid making sure I don't get consecutive serial numbers, but ones that are very different. I'd go to my local hardware shop and ask them specifically for that. They'd show me the drives / serial numbers before I ever even bought them for real.

I even used different manufacturers at some point when they didn't have non consecutive serials. I lost some storage because the drives weren't exactly the same size even though the advertized size matched, but better than having the RAID and extra cost be for nothing.

I can't fathom how anyone that is running drives in actual production wouldn't have been doing that.

Re: Backblaze Drive Stats for 2024

#108

Post by tredre3 »

Melatonic wrote: post #73

True enterprise drives ftw - even Seagate usually makes some very reliable ones. They also tend to be a little faster. Some people have complained about noise but I have never noticed. They are noticeable much heavier in hand (and supposedly most use dual bearings). Combined with selecting based on Backblazes statistics I have had no HDD failures in years

I'm not sure I follow you, are you really saying that your choice of Seagate was based on Backblaze's statistics? Maybe I'm missing something but aren't they the overall least reliable brand in their tables?

Re: Backblaze Drive Stats for 2024

#109

Post by RachelF »

loeg wrote: post #39

Hard to argue with those WDC/Toshiba numbers. Seagate's are just embarrassing in contrast. (HGST drives -- now WDC -- were great, but those are legacy drives. It's been part of WD for some time. The new models are WDC branded.)

...and many used Seagate drives have been resold as new in the last 3 years. They were used for crypto mining and then had their SMART parameters wiped back to "new" 0 hours usage.

https://www.heise.de/en/news/Hard-disk-fraud-Increasing-evid...

Re: Backblaze Drive Stats for 2024

#110

Post by Lammy »

walrus01 wrote: post #95
malfist wrote: post #88

Earlier quoted context omitted.

Out of curiosity, what server cases can actually accommodate 50 drives?

If you google "supermicro 72 drive server" it's definitely a thing that exists, but these use double-length drive trays where each tray contains two drives. Meaning that you need a "whole machine can go down" software architecture of redundancy at a very large scale to make these useful, since pulling one tray to replace a drive will take two drives offline. More realistically the normal version of the same supermicr…

> Meaning that you need a "whole machine can go down" software architecture of redundancy at a very large scale to make these useful

Also some serious cooling to avoid the drives in the front cooking the drives in the back (assuming front-to-back airflow).

Post reply on HN