MacInTouch Amazon link...
Channels
Products, Questions

Ric Ford

MacInTouch
I stumbled into an embarrassing solution. Turns out I had a tiny (13 mm) USB flash drive attached, but unmounted, and I had forgotten it. When I removed it, I could connect again.
I knew USB 3 attachments could interfere, but I would expect the interference to block any network. I wondered if the interference blocks 802.11b/g/n networks differently than 802.11ac networks? I did a bit of research and found a number of sites that said the USB3 interference is in the 2.4 GHz range, but not 5 GHz. Maybe I need a new 802.11ac capable router to protect myself from myself.
I never thought of that, but it isn't at all surprising.

I can easily demonstrate severe interference between WiFi/Bluetooth and USB 3 devices - it's a very miserable problem (at least as bad as classic SCSI issues - worse, I think).

If I have a USB 3 storage device (USB 3 SSD) near the back of my 2015 MacBook Pro 15", it can make Bluetooth and WiFi quite flaky (to the point they don't work at all). Of course, it all depends on precise location, cable, device, etc., ad nauseum. I have to work around the problems - usually by repositioning the devices slightly.

802.11ac should be more robust than earlier wireless standards:
Wikipedia said:
IEEE 802.11ac
The specification has multi-station throughput of at least 1 gigabit per second and single-link throughput of at least 500 megabits per second (500 Mbit/s). This is accomplished by extending the air-interface concepts embraced by 802.11n: wider RF bandwidth (up to 160 MHz), more MIMO spatial streams (up to eight), downlink multi-user MIMO (up to four clients), and high-density modulation (up to 256-QAM).
...
In June 2013, Apple announced that the new MacBook Air features 802.11ac wireless networking capabilities, later announcing in October 2013 that the MacBook Pro and Mac Pro also featured 802.11ac.
Your note prompted me to check my WiFi connection (Option-click on the WiFi item in the menubar), and I found I'd stupidly been using 2.4GHz, though I have an 801.11ac AirPort Extreme router.

So, I opened up AirPort Utility on an iPhone and found that I could set a name specifically for the 5Ghz network, which I did. After making the change and waiting for the AirPort to restart, I manually connected to the new 5GHz network (new name) and then configured System Preferences > Network to put that 5GHz network at the top of my priority list.

Seems faster now...
 


Ric Ford

MacInTouch
Turns out I had a tiny (13 mm) USB flash drive attached, but unmounted, and I had forgotten it. When I removed it, I could connect again. I knew USB 3 attachments could interfere, but I would expect the interference to block any network
It’s not just USB, apparently...
Apple said:
Use Apple Pencil with your iPad or iPad Pro
... If you're charging your Apple Pencil (2nd generation) with your iPad Pro and your car's keyless entry device (key fob) is nearby, signal interference might prevent you from unlocking your car with your key fob.
 


The workaround is pretty simple. Buy your own Ethernet bridge and attach it to one of your router's LAN ports. Then plug all your devices into the bridge. Simple bridges with Gigabit speeds are not very expensive.
Which is apparently sort of what I've done (only the fast ethernet devices are on the managed D-Link DG5-2205 "bridge", as I will call it henceforth), and now I know why it turned the LAN LED on the RT2600ac green. Thank you, David, I always read your posts with interest and dawning comprehension.

But is the RT2600ac in fact "a... router that forces all of the Ethernet ports to run at the same speed"? As far as I know, its LED status lights are only meant to indicate the speed of the attached device. And why, I wonder, don't the LEDs on the D-Link bridge show the Ooma Telo and the printer as fast ethernet (orange), as the RT2600ac did? It has always indicated the correct speed of an attached device in the other applications I've used it in. Perhaps the Telo and printer are in fact gigabit? Perhaps I should turn the RT2600ac's LEDs off? :-)

Thanks again.
 


I've a new MacBook Pro, and have been really frustrated on frequent wireless disconnects. Ran Wireless Diagnostics without really much hope that it'd tell me anything - but lo! It told me the 2.4GHz network I'm using is getting too much congestion from the many surrounding 2.4GHz networks! Switched to 5GHz, and been rock solid ever since. This one little insight also helped me solve problems I'd been having on a gaming PC (solved unfortunately by a wifi adapter switch - existing one did not support 5GHz), and made me re-evaluate my opinion on my existing WiFi router (I'd been playing a bit of a shell game every couple years, getting frustrated with WiFi and switching it out).
 


5GHz networks used to be less congested in my area, too. (That has since changed now that pretty much everyone has a multi-band access point). A few observations:
  • beam-steering access points (like Gen 6 AirPorts) with multiple MIMO antennas can help with interference regardless of band being used. I have found recent AirPort base stations to be phenomenally stable (hardware and software). Computers with MIMO antenna systems similarly benefit.
  • 5GHz signals tend to get attenuated faster than 2.4GHz signals, so less junk from the outside may interfere inside your home. Conversely, inside a reinforced concrete home, the 5GHz signals may have even more coverage problems than 2.4GHz ones.
  • Some older hardware can be upgraded - for example, I upgraded my (2012?) MacBook Air to 802.11ac by swapping in a more modern radio module.
  • Adding more hard-wired base stations throughout the house helps - both by potentially increasing the transmission speed as well as reducing the power needed to transmit reliably.
 


5GHz networks used to be less congested in my area, too. (That has since changed now that pretty much everyone has a multi-band access point). A few observations ...
Another important thing is that in the 5GHz spectrum, there are many more non-overlapping channels.

In the 2.4GHz spectrum, they all overlap, meaning there are only three non-overlapping channels (and only one if you use 40MHz carriers).

In the 5GHz spectrum, the channels are 10MHz apart, so half of them (skip every other channel) can be used with a 20MHz carrier without overlap.
 


Ric Ford

MacInTouch
Just a quick tip that might help other folks:
  1. On an AirPort Extreme 802.11ac (firmware 7.9.1), defined a separate 5GHz network name (Mac AirPort Utility.app > select device > Edit button > Wireless tab > Wireless Options button).
  2. Selected 5GHz network
  3. Looked at statistics via Option-click on WiFi menubar item to see Tx Rate, etc.
  4. Ran DSLReports Speed Test and noted results.
  5. Selected 2.4GHz network
  6. Looked at statistics via Option-click on WiFi menubar item to see Tx Rate, etc.
  7. Ran DSLReports Speed Test and noted results.
I found that the 2.4GHz network provided double the speed vs. the 5GHz network!

Your milage may well vary greatly, but that unexpected result suggests trying both options to see which works better.

(In this case, I see neighboring networks at both 2.4 and 5 GHz, and the computer is something like 10-15 meters from the AirPort with some walls and floors (wood/sheetrock) in between.)
 


15 meters (about 46 feet, with intervening walls) may favor the longer distance that 2.4 GHz networks provide good coverage, especially with the intervening obstacles
 


To me, wireless connections were always about convenience, not speed. For a reliably fast connection, nothing beats a wired connection... even something as slow as gigabit is usually faster for real-life applications (e.g. large backups) than any wireless system I've used... with the downside of being tethered.

Step up to 10Gbe copper/optical or 10NBase-T copper, and it's not even a contest. That trunk is so wide that most homeowners don't have the storage hardware on both ends to saturate it consistently. My 10Gbe connection does large transfers in seconds that my WiFi connection will do in minutes.

The speeds are so different, I will manually turn off Wifi when my Mac is on the 10Gbe network, because the service order preference in the Network control panel does not seem to be respected consistently or somehow the WiFi is interfering with connection speeds to the server.
 


I found that the 2.4GHz network provided double the speed vs. the 5GHz network!
I’ve found much the same thing and try to ensure I’m running on the 2.4GHz side (my MacBook Pro automagically switches to the 5GHz side on reboot, or sometimes just randomly on its own).

For context, we live on a rural farm, so there are no neighbors to compete with, and we’re in a 100-year-old four-square that’s had much of its plaster and lath replaced with modern drywall.
 


Ric Ford

MacInTouch
To me, wireless connections were always about convenience, not speed. For a reliably fast connection, nothing beats a wired connection... even something as slow as gigabit is usually faster for real-life applications (e.g. large backups) than any wireless system I've used...

Step up to 10Gbe copper/optical or 10NBase-T copper, and it's not even a contest. That trunk is so wide that most homeowners don't have the storage hardware on both ends to saturate it consistently.
That's a bit confusing, as 10GigE is about the same speed as 10Gbps USB 3.1 Gen 2 or Thunderbolt 1 and much slower than Thunderbolt 2 or 3. Since Macs can "saturate" a Thunderbolt connection, it doesn't seem that it would be difficult to saturate 10GigE. In other words, I expect (and have seen in GigE tests) that you could do a backup or file transfer far faster over Thunderbolt than over 10GigE, and 10GigE might even be the more expensive option of the two. For computer-to-computer transfers, as I've posted in the past, Thunderbolt Target Disk Mode has extremely poor performance for unknown reasons that Apple has never explained or justified.

I agree about wired vs. wireless speeds and reliability, though, and another factor is wireless radio interference from USB 3 devices (and maybe also from high-speed video cables).
 


That's a bit confusing, as 10GigE is about the same speed as 10Gbps USB 3.1 ...it doesn't seem that it would be difficult to saturate 10GigE.
My understanding is that Thunderbolt/USB-C data transfer is, relatively speaking, unmediated direct memory access transactions, while Ethernet chips have to handle packet collisions as well as broken and partial packets in addition to the flow of "normal" packets. Is that incorrect?
 


For computer-to-computer transfers, as I've posted in the past, Thunderbolt Target Disk Mode has extremely poor performance for unknown reasons that Apple has never explained or justified.
This probably is not the explanation you've sought from Apple, but I wonder if it could be the cables that proliferated when first we entered the "donglegate" era.

Apple's first Thunderbolt 3/USB-C 4-port laptops shipped with cables that were unlabeled and not designed to carry data, but which could do so, slowly, but with errors that would often terminate the transmissions. Could that be the root of this problem?
 


Ric Ford

MacInTouch
My understanding is that Thunderbolt/USB-C data transfer is, relatively speaking, unmediated direct memory access transactions, while Ethernet chips have to handle packet collisions as well as broken and partial packets in addition to the flow of "normal" packets. Is that incorrect?
That's my understanding, as well, and I just skipped over those issues, so, yes, Ethernet is likely to be even slower than what I described vs. USB, Thunderbolt et al.
 


Ric Ford

MacInTouch
Apple's first Thunderbolt 3/USB-C 4-port laptops shipped with cables that were unlabeled and not designed to carry data, but which could do so, slowly, but with errors that would often terminate the transmissions. Could that be the root of this problem?
That's absolutely not the case here. I never use any of those cables for anything but charging. The Thunderbolt cables I used were fully qualified for full Thunderbolt speed, and, in fact, delivered that speed with Thunderbolt storage devices, but not with Thunderbolt Target Disk Mode.
 


That's absolutely not the case here. I never use any of those cables for anything but charging. The Thunderbolt cables I used were fully qualified for full Thunderbolt speed, and, in fact, delivered that speed with Thunderbolt storage devices, but not with Thunderbolt Target Disk Mode.
Does this generalize to using the Migration Assistant to populate a new Thunderbolt 3 MacBook Pro from an older Thunderbolt 3 MacBook Pro? If I cannot couple the two computers together using a certified Thunderbolt 3 cable, what is the best choice for migrating the data?
 


Ric Ford

MacInTouch
Does this generalize to using the Migration Assistant to populate a new Thunderbolt 3 MacBook Pro from an older Thunderbolt 3 MacBook Pro? If I cannot couple the two computers together using a certified Thunderbolt 3 cable, what is the best choice for migrating the data?
It will work but not as quickly as expected, and migrating from an external SSD backup may be faster.
 


I'm in the process of selecting another Synology NAS for work, and the model that seems to best match our small office environment is the relatively new DS1019+. It's a five-disk NAS, with two slots for NVMe cache (read or write or read/write), but the reviews have lamented it is only 1-Gigabit Ethernet and offers no upgrade path to 10Gbit. It does come with two 1Gbit RJ45 ports, which can be used for link aggregation for more speed, though from what I've read, that isn't easy to set up and wouldn't much matter in the real world when the connected computers are themselves limited to 1Gbit.

Over my years of using computers, I've paid considerable amounts to "future-proof" for technology options that unfortunately never seem to distill from vapor. Or I find, as in my adoption of Thunderbolt 1, that my expensive technology is effectively (or even absolutely) obsoleted.

Given my experience with Synologies "limited" to 1Gbit, I've concluded that's adequate for the kind of files we store on the server and the kind of backups we do. If we were a video-editing shop, I doubt any speed but the fastest would be enough.
I expect (and have seen in GigE tests) that you could do a backup or file transfer far faster over Thunderbolt than over 10GigE
This YouTube Video from "NASCompares" supports what Ric posted, and explains it with some real-world examples.
https://www.youtube.com/watch?v=YSiKrT0ImPI​

(Do note the Thunderbolt advantage is for users able to "direct connect" to the storage system.)

Thunderbolt does seem to have real world limits and issues.
For computer-to-computer transfers, as I've posted in the past, Thunderbolt Target Disk Mode has extremely poor performance for unknown reasons that Apple has never explained or justified.
Apologies for linking to YouTube videos instead of more useful text explanations, but whichever search tool I use, I'm directed to these, and not text posts. This one, from a UK company, "SPANdotCom," that has a very good series of NAS video tutorials and explanations, delves into differences between Thunderbolt and 10GbitE and describes at a semi-technical level where and why 10GbitE may be the better choice than Thunderbolt, though it is very much a "network-centric" discussion.
https://youtu.be/PKeZD9QjM74?t=85​

One consideration. News out of Computex (May 28 - June 1, 2019) foretells major improvements in the PCIe interface on which Thunderbolt data transfer is based, with possible implications for current investment in Thunderbolt 3 - and 10GbitE:
Engadget said:
Gigabyte's next-gen SSD shows the incredible potential of PCIe 4.0
It can hit 5 GB/s, over 50 percent faster than the best NVMe drives right now.
Broadcom said:
BCM57508 - 200GbE
The BCM57508 200G PCIe 4.0 Ethernet Controller builds upon the success of the widely-deployed NetXtreme E-Series architecture by combining a high-bandwidth Ethernet controller with a unique set of highly-optimized hardware acceleration engines to enhance network performance and improve server efficiency.
Even with PCIe 4 just entering the supply chain, PCIe seems to be following along at a faster than historical pace.
Techspot said:
PCIe 5.0 specification announced, will bring 128GB/s transfer speeds
Taking 18 months to develop, PCIe 5.0 doubles the peak bandwidth of PCIe 4.0 from 64GB/second to 128GB/s via x16 configuration.

The reason PCIe 5.0 has been ratified before PCIe 4.0’s consumer launch is the latter’s delay; having stuck with a four-year release cycle for PCIe 1.0, 2.0, and 3.0, there were seven years between the arrival of PCIe 3.0 in 2010 and the completion of PCIe 4.0 in 2017.
 


Ric Ford

MacInTouch
This one, from a UK company, "SPANdotCom," that has a very good series of NAS video tutorials and explanations, delves into differences between Thunderbolt and 10GbitE and describes at a semi-technical level where and why 10GbitE may be the better choice than Thunderbolt, though it is very much a "network-centric" discussion.
https://youtu.be/PKeZD9QjM74?t=85
I took a brief look at that, and the argument for 10GbE seemed really weak: that hard drive NAS can't take advantage of faster speeds. Well, of course, hard drives are slow, but that's not at all the case for SSDs. 10GbE should be fast enough for SD Card readers, too.
 


My understanding is that Thunderbolt/USB-C data transfer is, relatively speaking, unmediated direct memory access transactions, while Ethernet chips have to handle packet collisions as well as broken and partial packets in addition to the flow of "normal" packets. Is that incorrect?
This is going to depend heavily on the NIC [Network Interface Controller] you're using. A high performance NIC typically includes a dedicated packet processing chip (or an FPGA programmed for packet forwarding) and should support a DMA interface to system memory. When coupled with high-performance packet forwarding software (e.g. something based on DPDK instead of your OS kernel's generic device drivers), a modern computer can definitely keep up with these high bit rates - up to the limits of the PCIe bus itself. (Of course, the Ethernet ports built-in to the motherboard on a typical PC or Mac will almost certainly not be a high performance NIC.)

That being said, you probably won't achieve those speeds just running typical applications (web browsers, video streamers, etc.) on a consumer network. Most applications use the OS kernel for packet forwarding, and OS kernels are optimized for overall system performance, not just network packet forwarding, and therefore will probably not be able to keep up with a 10G (or 40G or 100G or 400G) network interface.

As for packet collisions and the like, that is not likely to be the case. Although shared buses (as is the case with old-style 10/100M Ethernet with dumb repeater-hubs) are theoretically possible with modern Ethernet (CSMA/CD is still part of the specifications), anyone running a high-speed Ethernet network will have each host connected directly to a port on a switch-type hub (a bridge or a router). In such a topology, collisions are impossible, and full-duplex (where a host can transmit and receive at the same time) is typical. With a network like this, your performance may be limited by the capabilities of the switches you are using, congestion (from other hosts on the network), and other related things, but not from collisions.

Packet corruption is always possible, but it should be very rare if your cabling infrastructure is properly built, according to EIA/TIA specifications. If you're seeing a lot of packet corruption, then you need to inspect your cables and premises wiring, because something is not up to spec for the Ethernet standard you're deploying.
 


I took a brief look at that, and the argument for 10GbE seemed really weak: that hard drive NAS can't take advantage of faster speeds.
The point I thought the discussion made in favor of 10GbE over Thunderbolt 3 was 10GbE's ability to support many more connections and longer "range." I've found no NAS with Thunderbolt support for more than two simultaneous users. The 6.6 foot maximum length of Thunderbolt 3 cables means those two users would have to be close (or daisy-chained in through intermediate powered Thunderbolt docks). I don't think anyone doubts that Thunderbolt 3 is faster than 10GbE.

This QNAP NAS line supports both Thunderbolt 3 and 10GbE. Fill it with SSDs in RAID 0, and it should be blazing fast on Thunderbolt, while offering the opportunity to connect to a 10GbE network.

In my own view, it would be better to provide those who need massive fast local storage their own devices. Then back them up to storage configured for durability and reliability.
Amazon said:
QNAP TVS-472XT-PT-4G-US 4 Bay Thunderbolt 3 NAS with 4GB RAM, 10GbE, M.2 PCIe NVMe SSD slots
Current Amazon price, diskless, is either $1,149 or $1,139, depending on which of Amazon's variable price tabs open!
That being said, you probably won't achieve those speeds just running typical applications (web browsers, video streamers, etc.) on a consumer network.
And that's another really good point, though it would favor Thunderbolt if really fast external storage is crucial. "My" networks (home and office) are built with "consumer grade" gear. No 10GbE in sight - Netgear Nighthawks in two locations together with unmanaged switches, an Airport Extreme in another.

We have one system and one MacBook Pro used for graphic design, and both of those rely on local attached storage for both creation and backup of large files.

The kind of "work files," spreadsheets, small databases, word processing documents, PDFs, in our daily workflow seem plenty fast from a slow 6-year-old ARM Synology over its 1GbE RJ-45.
 


Ric Ford

MacInTouch
The point I thought the discussion made in favor of 10GbE over Thunderbolt 3 was 10GbE's ability to support many more connections and longer "range." I've found no NAS with Thunderbolt support for more than two simultaneous users. The maximum length limits of Thunderbolt 3 cables means those two users would have to be close (or daisy-chained through intermediate, powered Thunderbolt docks). I don't think anyone doubts that Thunderbolt 3 is faster than 10GbE.
I understand that Ethernet is probably easier/better for connecting a whole bunch of users to one NAS, but the irony is that 10GbE connections will likely require a Thunderbolt adapter, since only special-order 2018 Mac Minis have the option of built-in 10GbE ports.

2-meter Thunderbolt 3 cables are readily available:
Amazon said:
And long Thunderbolt 2 cables are available, too:
Corning said:
Thunderbolt™ Optical Cables
Thunderbolt™ Optical Cables by Corning connect computers and devices at incredible speed and over longer distances. They’re thin, light and remarkably tough — Optical Cables by Corning can be bent, squeezed, and tangled.
Available in 5.5m, 10m, 30m, and 60m lengths.
(Thunderbolt 2 optical cables can be used with Thunderbolt 3 Macs via Apple's Thunderbolt 3-Thunderbolt 2 adapter.)
 


Ric Ford

MacInTouch
A few more notes on AirPort devices and related issues: First of all, AirPort Utility has useful built-in help (via a Help menu). Apple web pages have some overlapping and helpful material, including:
 


I understand that Ethernet is probably easier/better for connecting a whole bunch of users to one NAS, but the irony is that 10GbE connections will likely require a Thunderbolt adapter, since only special-order 2018 Mac Minis have the option of built-in 10GbE ports.
If money is no object, the iMac Pro comes with 10 Gbit Ethernet built in (via the normal RJ-45 port).
 



Since Macs can "saturate" a Thunderbolt connection, it doesn't seem that it would be difficult to saturate 10GigE.
In my experience, it actually is difficult to saturate a 10Gbe connection, though it depends on the hardware. Perhaps if you're running an all-flash storage system on both ends and have favorable conditions re: the data you're transferring (large files typically allow faster transfers on a MB/s basis), the network protocols in use (i.e. NFS vs. AFP vs. SMB, etc.), and the transfer protocol in use (for example, Rsync works on a file-by-file basis).

Even though SSDs have allegedly amazing performance figures, it pays to place close attention to specifications. Some cheaper consumer-grade stuff uses a fast buffer up front and slower flash memory for actual storage. That works well for most operations (the buffer doesn't hit its limits much and then empties to flash) but may not work at all for long sustained transfers (the buffer is always hitting its limits, slowing the transfer to the speed of the Flash in the back).

No doubt, a DAS using Thunderbolt is theoretically faster than a "similar speed" ethernet-based solution due to lower overhead. Be careful however, regarding what is considered a DAS... (see my other post) I doubt anyone will ship a "real" Mac-compatible DAS / NAS combination ever, simply due to ownership / permissions issues.

When it comes to spinning media, you either stripe a lot of disks (which is risky) or you will need a whole lot of disks to get to 1000MB/s transfers, even under favorable conditions. A FreeNAS server running a common Z2 Pool with 6 drives per VDEV would likely need over 30 spinning drives, for example.

SSDs theoretically can run a lot faster than their spinning ancestors but it's usually pretty cost-prohibitive as SSDs optimized for fast writes (Intel Optane P4801x series, etc.) offer only 1GB for every $2.50, well above the usual price points for regular SSDs. There is no free lunch.

I can't speak to the issues observed re: Target disk mode. However, if the BlackMagic test files are small, I'd be content with the numbers you observed. Here is an output for the 100GB P4801x that I ran on my FreeNAS server. Note the huge discrepancy re: small file performance vs. large file performance - over 50x difference!
Code:
Synchronous random writes:
         0.5 kbytes:     28.0 usec/IO =     17.4 Mbytes/s
           1 kbytes:     27.0 usec/IO =     36.1 Mbytes/s
           2 kbytes:     27.5 usec/IO =     71.1 Mbytes/s
           4 kbytes:     23.4 usec/IO =    167.1 Mbytes/s
           8 kbytes:     30.4 usec/IO =    257.0 Mbytes/s
          16 kbytes:     44.1 usec/IO =    354.3 Mbytes/s
          32 kbytes:     64.6 usec/IO =    483.5 Mbytes/s
          64 kbytes:    103.7 usec/IO =    602.7 Mbytes/s
         128 kbytes:    161.1 usec/IO =    776.1 Mbytes/s
         256 kbytes:    285.8 usec/IO =    874.6 Mbytes/s
         512 kbytes:    527.5 usec/IO =    947.9 Mbytes/s
        1024 kbytes:    988.2 usec/IO =   1012.0 Mbytes/s
        2048 kbytes:   1905.6 usec/IO =   1049.5 Mbytes/s
        4096 kbytes:   3730.2 usec/IO =   1072.3 Mbytes/s
        8192 kbytes:   7398.6 usec/IO =   1081.3 Mbytes/s
 


This QNAP NAS line supports both Thunderbolt 3 and 10GbE. Fill it with SSDs in RAID 0, and it should be blazing fast on Thunderbolt, while offering the opportunity to connect to a 10GbE network.
Yes and no. QNAP claims this is a DAS when it's not: QNAP uses Ethernet over Thunderbolt, and whatever Thunderbolt bus you connect the QNAP to will take over all devices on that bus. So the experience is very much like a NAS (including setting up IP addresses and so on), except that you use Thunderbolt cables instead of 10Gbe copper or optical gear.

If I remember correctly, the QNAP offers two Thunderbolt connectors, i.e. up to two users can connect at the same time. I thought it would be pretty ideal for me, except I owned a MacBook Air at the time (which only has one Thunderbolt connector). Once the QNAP was attached, I lost my dock, all attached drives, the monitor, etc. They all now "belonged" to the QNAP. QNAP confirmed this is by design. Not so much a problem for computers with multiple Thunderbolt ports (dedicate one to the QNAP, attach all other gear to the other ports), but a deal breaker for me.

I am very content with my current FreeNAS server, which uses a combination of spinning media and SSDs to offer good performance. An Intel P4801x 100GB NVME stick works as a SLOG (for writes), while another SSD speeds up metadata read operations (for file browsing, rsync and so on). Even with rsync, backup performance has been fantastic.
 



Amazon disclaimer:
As an Amazon Associate I earn from qualifying purchases.

Latest posts