On certain websites you can order laptops, and sometimes even they allow you to change certain parts of it.
Let's take the example of HDD, the website offered quite an amount of types of HDD's.

The largest HDD they can deliver to the laptop is 2TB, could I instead ask them to let the slot open, and then buy let's say 860EB and put it in.

  • Will Linux adapt these entire 860EB? Will it be able to write to this disk as it could to normal 2TB disk? Is there no limitation to amount of data understood? Entire 860EB will be absolutely usable? And what if I had 4 slots, and all of them had 860EB HDD, would it still be able to utilize all 4 entire HDD's? Would the same apply to SSDs?

Don't ask me why would I need 860EB or how, I won't be probably getting 2TB disk either. This is just a thing that I thought about, that I find interesting to know. And knowledge is the key.

  • When thinking about speed, is speed of SATA all I need to worry about? It turns out it's maximum speed (according to Wikipedia) is theoretically 16Gbps which is 2GBs. That means that every HDD and SSD connected to my SATA bays will technically never be able to read/write more than 2GB per second. Is there anything else I should be paying attention to?

By Linux, I mean regular user-grade distribution of OS (Ubuntu, Fedora, Arch), not a server/super-computer/scientific one (RHeL, SL).
By 860EB, I mean 860 exabytes, which is equal to 860000000TB (860 million terabytes).

Dani AI

Generated

Brief summary tied to the thread: the Linux kernel and common user distributions have no built‑in “2 TB” ceiling — the kernel understands 64‑bit device sizes — but the practical limits are almost always hardware, partition table format, drive firmware/bridge chips, and the filesystem chosen. ’s idea of throwing an absurdly large device at a laptop runs into those practical limits long before the OS refuses to talk to the device. (kernel.googlesource.com)

Partitioning and addressing are the first blockers. Legacy MBR uses 32‑bit LBA fields (so with 512‑byte sectors that effectively clips at ~2 TiB); GUID Partition Table (GPT) uses 64‑bit LBAs and removes that small‑disk limit. For any disk >2 TiB make sure the device is using a GPT label and that firmware/HBA/bridge present the full capacity. (en.wikipedia.org)

Controller/drive addressing comes next. Many SATA/ATA stacks historically used 48‑bit LBA (LBA48), which tops out at 128 PiB (≈144 PB decimal); USB/SATA bridge chips, some RAID cards, or older HBAs can impose even smaller limits. That means an “860 EB” single drive is beyond the addressing capability of most commodity controllers today. (en.wikipedia.org)

Filesystems and volume managers set the remaining, higher ceiling. ext4, XFS, Btrfs and ZFS all have very large theoretical limits but different practical constraints (single‑file limits, block size effects, metadata scaling). For very large pools ZFS or XFS are commonly chosen on Linux servers; ext4 is fine for normal desktop/server sizes. Pick a filesystem whose documented limits and behavior match the planned usage. (docs.kernel.org)

Throughput: SATA 3.0 (the common “SATA6G”) is 6 Gbit/s per link (about 600 MB/s raw); so a single SATA SSD is limited by that link unless it uses PCIe/NVMe or a SATA‑Express/PCIe bridge. Multiple independent SATA ports can give aggregate bandwidth, but the real bottlenecks become the controller, PCIe lanes, RAID/MD/LVM overhead, and CPU. To inspect and prepare large disks on Linux, tools like lsblk and parted show transport, sector size and partition label; use GPT on disks >2 TiB. (en.wikipedia.org)

Practical takeaway linked to earlier posts: was right to name ZFS as “very large” in ambition, and ’s point about typical SSD/HDD sizes is also on target — for a desktop the limiting factors are physical fit, controller support and cost, not the Linux kernel itself. (openzfs.github.io)

Recommended Answers

All 10 Replies

There are no discs of that size, but some file systems can support that size of array or multi-disc volume, such as zfs which can handle file systems of 256 zetabytes, and single files of 16 exabytes. Currently, the biggest single disc I know of is about 4TB, but I think that is in a format that won't fit into a laptop.

I know, that they don't exist... but question was, would support immensely large HDD storage, I found 8TB, it would be nice if it fit.

I don't think there is any current HDD over 2TB that will fit into a laptop. There are 3 and 4TB units, but they are 3 1/2" and thus too big to fit in a laptop.

Anyway, to paraphrase Star Trek - disc space, the final frontier!

Okay, third attempt, :D, If I had a desktop, and if I had SATA entrance that would be able to handle 3 1/2" drives, would it be able to handle "infinitely" much storage memory? Like 8TB one?

There's still the 2nd question.

  • No reason for it not to be able to work with 8TB one

  • what do u mean by transfer memory? Speed with which you can write to a hard drive?

  • And would it work with much larger one? If I had 500TB one? Could I also use it?

  • hard drives*, if I had 4xSSD with 600MBs, it at 2GBs is the only limit I need to worry about it? Or is there anything else?

!bump

commented: See my comment below. +13

My laptop has an eSata port. I can connect an array with 8+TB of space to it without problem. Unfortunately, a lot of current laptops are moving exclusively to USB 3.0 ports and removing the eSata ones. Pisses the heck out of me! I need to get a net laptop, but none of the ones I am interested in have eSata ports without a USB -> eSata adapter.

Thanks, what about the second question?

Gee, why the downvote? Topic was burrowed, 6 days old and no one answered, I still requested answer. Explain the logic.

Don't know who/why the downvote, but I gave you an up-vote with comment (so to speak), which will leave you with more positive than negative points. :-)

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.