I mean for one thing I carry a backup of that data with me anyway. It's on my laptop and my phone.
They generally have RAID boxes with either Ethernet or thunderbolt connection.
Most video editing/rendering is done on big storage arrays anyway. So 256GB storage is more than enough for most serious multimedia professionals.
Besides, perhaps you're the user that needs the multi-TB version. Others, e.g. video professionals wont have huge tens-of-TB source files in their main system drive.
If you add a high performance spinning disk its performance won’t be enough and it will be noisy even when idle.
In the end, it’s better to not add anything and let professionals bring their own storage into the mix.
At the end of the day this won’t be sold to new beginners with no files. The buyers will have their resilient storage and the internal drive won’t be used much anyway.
Unless it’s configured as a developer workstation, but it’s generally MacBook Pro’s role.
This is an extremely common setup for video production, so it’s not out of the target market at all.
But I do agree that SAN storage is the norm here.
It's done this way so people can connect to the storage that's appropriate for the task at hand.
Different projects can be stored on different SAN/NAS arrays , each specc'ed out according to the size/needs of the project.
eg a short run animation doesn't need the same storage capacity of a full length feature film. They may have similar throughput needs though. (Summarising here, but the general concept is ok)
So, lets say someone is a Flame editor (Flame generally has high bandwidth needs). They're working on project A for today. So they connect to the storage being used for that project from their workstation. The next day, they might be working on a different project, so will be connecting to a different set of storage.
Other people using different software will connect to the same storage for their tasks, but can have different latency/throughput needs.
Obviously, this isn't the approach taken by single person multimedia er... "houses". ;)
High end networking gear already has higher throughput:
http://www.mellanox.com/page/ethernet_cards_overview
https://www.servethehome.com/mellanox-connectx-6-brings-200g...
A SAN/NAS using the same PCIe 4.0 SSD's you mention could probably fill the pipes too.
... and it would probably need a bunch of network stack tuning. ;)
> High end networking gear already has higher throughput:
100GB/s > 200Gbps
You would need 4x 200Gbps ports to reach 100GB/s, so 2x MCX653105A-ECAT (each 2x 16-lanes) at > $700 each, and pay for 1/10th of a ~$30 000 switch, IOW 100GB/s would cost you ~ $4400, before paying for the storage.
Sure, it could be done, but it wouldn't be cheap, and you'll have used most of the PCIe lanes.
EPYC servers (128 PCIe lanes) would probably be the go too, not Xeons.
This is just Imagineering though. ;)
With the specifics, wouldn't it be 4 cards needed? Each card has 2x 100Gb/s ports, so 8 ports in total.
That being said, after re-reading the comment they're talking about adding multiple PCIe cards to a box to achieve 100GB/s of local total throughput.
That would be achievable over a network by adding multiple PCIe 200Gb/s network cards too. :)
As a sibling comment mentions though... the cost difference would be substantial. :(
It's good to see that local storage has finally returned to the reasonable state of being faster than network storage. SATA / SAS was a long, slow period ...
So, even with protocol overhead from all the stack layers chewing up maybe an order of magnitude, that'd still leave 10GBps.
So .. I guess it's still possible, if impractical, to outperform a good PCIe SSD with the latest network interface.
More 0's needed? :)
Maybe a slight typo there? Need to add a few zeros? :)
Mind you, you'll pay $15K for it, but if you're in that business you can well afford it even if you're not a top tier Hollywood production shop.
Given that your storage array costs that much, the fully loaded Mac Pro price (somewhere in the $20K range?) is not that outrageous. The people who use Red cameras and G-Tech storage arrays are the Mac Pro demographic Apple is going for here.
Disclaimer: I used to work with G-Tech but no longer there.
They are popular though. I see them a lot, but I have had very little success with them over the years.
That's for storage physically located nearby though. eg same building, no more than 1 or two network hops (roughly).
Higher end networking from the last decade or so has become really fast. :)
Looking quickly at the US Ebay just now, here are some older single port ConnectX-2 cards:
https://www.ebay.com/itm/Mellanox-MHQH19B-XTR-PCI-E-Single-P...
US $23.00 each, free shipping. Note - I don't know the seller at all, this is just from a quick look on Ebay.
There's a tonne of stuff on there. ConnectX-2 is no longer supported by Mellanox, though the cards themselves are generally pretty bullet proof.
Drivers for Linux and FreeBSD come with the OS. :)
Drivers for Windows are a bit more complicated. Mellanox has an archive section with the ConnectX-2 drivers, which work for most people.
Cabling directly between two cards - instead of going via a switch - is pretty common for people just trying out the tech. It lets them plug one card into (say) a FreeNAS server, with the other card in their desktop or workstation. Removes the ~120MB/s limit of 1GbE, assuming any kind of reasonable disks in the connected NAS. :)
If this kind of thing is of interest, probably the best place on the whole internet (not joking) for this stuff is Serve The Homes' Networking forum:
https://forums.servethehome.com/index.php?forums/networking....
Plenty of advice and stuff to read there, and a good place to ask questions from people that have a clue. :)
I can have a 100TB volume for under $10,000 that multiple editors can R/W to over 10GbE. We probably average about 400-600mb/s to a client on a simple setup in real world situations, which is fast enough for a few editors to work on multiple streams depending on the server. It's easier to backup/archive from an IT standpoint.
Most people working solo don't need to access hundreds of TBs though. They're fine with local or thunderbolt attached storage devices.
I really can't imagine this world where "professionals" keep everything on a single drive and are primarily concerned with capacity of that single drive.
My gaming computer has three SSDs, including a 120GB for the OS, and a pair of magnetic drives.
I totally understand gamers need more space, people who collect movies and videos need space, graphic designers need more space. Maybe even Android OS developers need it due to the Android Open Source Project, AOSP, having like 100 repos you need to clone all at once. I don't need it as an app developer, though.
Apple has a history of offering small base storage/memory options, and then charging more for the upgrade versions than they physically cost to produce. It's how they make different price level products when the products are all fundamentally the same. This lets them cater to poorer customers than they could otherwise since the rich ones buy the overpriced upgrades and subsidize the poorer customers' hardware. So it isn't all bad.
- 256 SSD 1400 MB/s Write 2200 MB/S Read
- 512 SSD 1846 MB/s Write 2500 MB/S Read
- 1T SSD 2500 MB/s Write 2600 MB/S Read
- 2T SSD 2600 MB/s Write 2700 MB/S Read
So this is mostly with write, the iMac Pro I'm writing this on with 1TB is in line with ~2800 MB/s write speed.But of course this could be very different with the new iMac which might have a very fast write SSD in the 256 GB model.
Saying no one would buy it is an exaggeration.
I refuse to believe that meticulously emptying out your trash and downloads folder so that you can continue to squeeze by with 256GB is now acceptable for ultra high end, "I don't know a single person who can afford this" computing.
If you're still relying on local storage in 2019 for development, analytics, video/image work, etc. you are simply doing it wrong
Though it is a bit slim if you want to store lots of multimedia or host multiple VMs.
Unless these cases are manufactured in remote monasteries, the costs don't add up and price point wasn't a concern.
EDIT: I see now there are only two internal drive bays, so that would mean tossing the 256GB that comes with it if you wanted to max out the internal storage.
Or using some of the several PCIe slots to add lots more SSDs. It looks like the default configuration has enough spare PCIe slots and lanes to accommodate 8 M.2 SSDs in suitable risers. That can get you another 16TB for ~$1.8k (consumer-grade SSDs).
The encoding/decoding capacity of Ethernet-over-twisted-pair seems to be reaching a point of sharply diminishing returns and the hardware for 10GBASE-T is really expensive for what's been, in my experience, a less reliable experience than SFP+. I'd rather tell somebody who needs 10GBASE-T to use an RJ-45 adapter than to have to go grab yet another Mellanox card to stick in a computer this expensive.
Is that the wrong way around? I'm struggling to parse otherwise.
1) https://en.wikipedia.org/wiki/Small_form-factor_pluggable_tr...
1 x 10 GbE for your regular Ethernet/IP traffic and 1 x 10 GbE for your storage traffic. You do run your network storage over a physically separate network, right?
WiFi is not really an option, for performance but also often for security reasons.
So yes, you'll have two cables from wall-jacks to the machine for networking.
Of course. WiFi isn't going to work in shared user environments that chew bandwidth for breakfast.
Not having to worry about the security issues of WiFi too can be helpful (depending on your security requirements).
Link aggregation in general only provides performance with multiple connections to multiple machines, so using link aggregation between one client (Mac) and one NAS will like result in zero performance improvement (the packets will only use one of the cables). It only makes sense if you have two or more NAS that you want to access simulatenously (or two or more clients accessing the same NAS, but that wouldn't be a use case for 2x 10gbe port on a client).
Synology also supports balance slb bonding, which in theory goes around this single connection restriction. However I ran into some connection problems with some Windows clients. Never went to the bottom of them but they went away when I disabled the bonding.
In any case, it is hard to saturate a 10gbe connection with a single NAS, unless it is packed with SSDs, which I wouldn't assume for mass storage. So I am not sure there is much value in aggregating the links in the first place.
I have multiple LACP bonds on my Juniper EX2200 at home working without issue, though the single stream limits you mentioned are the one thing LACP can’t fix.
"enterprise" DLink switches aren't really a thing yet, regardless of what their marketing team wants to brand them as. :(
Cisco, HPE, etc have "enterprise" switches. DLink might be in a decade.
> it is hard to saturate a 10gbe connection with a single NAS, unless it is packed with SSDs
No, it's just a matter of having enough spindles behind it.
As a rough guide, with a (say) average spinning rust HDD able to push out 100MB/s when reading, you'd only need 10 such drives to push out 1000MB/s (raw).
In the real world, you need extra spindles as some of the data being pushed out is just internal checksum/redundancy, and doesn't go over the network.
But for reading back large files in mostly sequential access, you'll hit 1GB/s from about 10 drives onwards pretty easily. More drives, more throughput.
Went with FreeNAS instead, as I was already very familiar with building systems, it's based on FreeBSD (OSS), and it gives better tuning on the higher end.
Doesn't have as pretty a GUI though. ;)
From maintaining large image storage FC SANs with Apple xServe and IBM DS for some years this is no fun to administrate for peak performance and I would assume most smaller shops have NAS instead of SANs (no arguing larger shops have SANs as we did driving ten photo studios).
Even the old 5400 RPM HDDs cost an arm and a leg if you bought it pre-installed into your MacBook. This was back in the days where you could unscrew the lid on your MacBook Pro with a standard screwdriver and replace the internal memory.
It's a neat psychology trick. Customers are far more receptive to upsells when they're the one doing the upselling.
Right.