Exploring Time Capsule: theoretical network speed vs practical throughput
Time Capsule, announced earlier this year, is a base station with an integrated hard drive and power supply. The previous segment of this series exploring Time Capsule in depth looked at the differences in members of the AirPort family. This segment, the second of six, compares the differences between the hypothetical maximum data transmission speed and typical real world performance of Time Capsule's SATA, USB 2.0, Ethernet networking, and WiFi Wireless networking interfaces.
In Theory, Theory and Practice Are the Same. In Practice, They Are Not
The theoretical speed rating of a given wired or wireless connection can be betrayed by a number of factors, from the overhead of the protocols involved to signal interference. Network data throughput is usually measured in megabits per second, which are an eighth of a megabyte per second. Disk speeds are typically cited in megabytes per second; here, I'll list both numbers to make it easier to compare disk and network throughput speeds.
SATA, or Serial ATA, has a theoretical maximum of 1200 Mbits/sec (150 MB/sec). However, existing hard drives can't even deliver data that fast; top disk output speeds are closer to 40 to 100 MB/sec, depending on whether the data is being read from the inside or outside of the disk platter, the disk spin speed, and other factors.
USB 2.0 has a theoretical maximum of 480 Mbits/sec (60 MB/sec). A USB hard drive is typically a standard ATA or SATA drive attached to a USB bridge chipset. The actual speed of the USB interface depends upon the performance of the chipset used as well as the performance of the computer the drive is attached to. That's because USB transfers most of the heavy lifting to the host computer's CPU.
USB has a faster theoretical maximum than Firewire 400 (400 Mbits/sec; 50 MB/sec), but Firewire 400 is actually much faster than USB because it uses smarter peer to peer interface hardware rather than pushing low level work onto the PC host's CPU as the simpler master to slave architecture of USB does.
On a Mac, Firewire is typically around twice as fast in real world transfer rates, with USB hitting around 18 MB/sec and Firewire reaching 35 MB/sec throughput. Windows' implementation of USB has historically been faster than Mac OS X's, with Windows' USB reaching throughput closer to 33MB/sec. That also explains why Firewire is more popular on the Mac than on the PC side; it's simply far more dramatically faster than USB on the Mac, while Firewire offers less of a noticeable boost in Windows. Macs also have Firewire Target Mode, which PC users lack. For more details on why USB is faster in Windows compared to the Mac, see the footnote: USB Performance in Windows vs Mac OS X at the end of this article.
Time Capsule doesn't use Firewire; it's USB only. There are two reasons for this. First, USB chipsets are cheaper than Firewire, because they do less (USB peripherals have less intelligence on board and transfer more work to the CPU). Second, Time Capsule and the AirPort Extreme are both designed as wireless network appliances, so the difference in performance between attached Firewire and USB drives typically wouldn't be noticeable. Test results presented in the next segment bear that out.
In reality, USB doesn't simply run at a given speed. The performance of a directly connected USB drive can be affected by a number of issues, from the performance of the host computer to interference caused by other USB devices on the same bus, to the overhead related to the drive's file system.
Ethernet Networking introduces even more complicating factors. There is the overhead of Internet Protocol addressing, as well as the file sharing protocols used, such as AFP on the Mac or SMB used by Windows, neither of which play into direct, non-networked protocols such as USB. There are also architectural issues such as the quality of the cables used and the performance of any switches (or old fashioned hubs) involved. All of these issues eat into the theoretical raw data transfer rate of Ethernet.
Fast Ethernet has a theoretical speed of 100 Mbits/sec (12 MB/sec), while Gigabit Ethernet has a theoretical speed of 1000 Mbits/sec (120 MB/sec). That suggests a double speed advantage of Gigabit Ethernet over USB (60 MB/sec), but neither protocol hits its maximum. In reality, a typical USB connected disk is roughly equal to or lesser than the throughput of a shared drive attached over a Gigabit Ethernet network.
Wireless Networking has all the complexity of traditional wired networking with the additional complications of signal strength issues such as radio interference and barriers, as well as additional overhead related to wireless transmission that commonly halves its real world throughput over the theoretical raw data rate.
- 802.11b has a theoretical speed maximum of 11 Mbits/sec with a typical transfer rate of around 4.5 Mbits/sec (0.5 MB/sec) with an ideal signal.
- 802.11g has a theoretical speed maximum of 54 Mbits/sec, with a typical transfer rate of around 23 Mbits/sec (2.5 MB/sec) with an ideal signal.
- 802.11n has a theoretical speed maximum of 300 Mbits/sec, with a typical transfer rate of around 74 Mbits/sec (9.25 MB/sec) with an ideal signal.
As the signal strength of a wireless network drops, the connection speed is automatically renegotiated and slower and slower rates until no connection is possible. The transfer rates of wireless networking make it ideal for browsing the web, as most US residents have a connection speed of around 1.5 Mbits/sec for DSL, or from 3 to 6 Mbit/sec with cable Internet service. Any version of WiFi is much faster than that.
However, very fastest wireless networking is required to perform intensive data transfers such as Time Machine backups, general file sharing, and media streaming, particularly if more than one client is using the network at once, or if one user is trying to do more than one thing with their wireless connection, such as backing up files while streaming audio to Apple TV, for example.
A Visual Speed Comparison
This chart shows the relative difference in throughput of the interfaces described above, with theoretical raw data rates in blue, and typical real world throughput in red. Note that these real world numbers are ideal peak maximums, not the average throughput users will see at all times. As detailed above, there are lots of factors that can eat into the actual real world performance. Time Capsule has performance limitations of its own, which are related to its design to primarily serve wireless clients. An upcoming segment will detail what Time Capsule itself can do.
Direct connection interfaces, such as SATA and USB, commonly deliver closer to half their theoretical maximum raw data rate, but as interfaces and drive mechanisms improve, the real world data throughput will rise. Ethernet networking interfaces, such as Fast Ethernet or Gigabit Ethernet, can hit peak transmission rates close to their maximums, but suffer from greater overhead compared to a direct connection interface.
Wireless networking throughput depends more on external factors to reach its full potential. Ideal signal strength is critically important to reach anywhere near the high end of real world throughput numbers. There are other factors that make a huge difference in wireless performance; Time Capsule and AirPort Extreme both support new features unique to the new 802.11n wireless networking protocol, including the use of multiple antennas (a technology referred to as MIMO) and the use of the 5 GHz radio spectrum. The next segment will look at the pros and cons of using this alternative frequency, which depending on the circumstances can either decrease signal range or deliver a major boost in your wireless data rate.
Footnote: USB Performance in Windows vs Mac OS X
In addition to the cabling and protocol specifics, there are other reasons for Windows PCs to outperform Macs in USB transfers. The testing done by BareFeats in the article USB 2.0 versus FireWire compared 2004 PowerPC Macs against 3 GHz Pentium 4 PCs; since USB pushes much of its work to the CPU, the speed of the host made a big difference in how fast USB performed on the two platforms.
Their testing also revealed that the first generation of the PowerMac G5 delivered poor I/O across the board, scoring lower than even the mobile PowerBook and low cost eMac in both Firewire and USB. That indicates that the theoretical expectations for USB (or any protocol) are nearly meaningless when compared to the actual speed of the disk, processor, the implementation of the protocol itself, and other factors that might cause interference or otherwise eat up the expected maximum throughput speeds. In other words, USB does not ever run at its maximum theoretical speed rating.
Additionally, Windows file sharing and disk protocols are simpler than on the Mac, because Windows handles and presents less metadata. This lightness makes for faster disk operations at the expense of the sophistication of the Mac's higher quality file icons, richer file type and creator codes, and other features missing in Windows.
There are other factors that affect cross platform throughput as well; Mac OS X suffers some degree of overhead from new features such as Spotlight indexing, while Windows PCs are typically burdened with running anti-virus scanning software that peels away a significant edge in performance. Clearly, there are lots of factors to account for in making direct performance comparisons, and neatly presented numbers can easily hide those details in a misleading way.
Previous articles related to Time Capsule and its AirPort Extreme cousin:
Exploring Time Capsule: how it fits into Apple's AirPort family
An in-depth review of Apple's 802.11n AirPort Extreme Base Station
Apple Time Capsule unboxing and preview
A Look Inside Apple's New Time Capsule
Answers to Time Capsule reader questions
In Theory, Theory and Practice Are the Same. In Practice, They Are Not
The theoretical speed rating of a given wired or wireless connection can be betrayed by a number of factors, from the overhead of the protocols involved to signal interference. Network data throughput is usually measured in megabits per second, which are an eighth of a megabyte per second. Disk speeds are typically cited in megabytes per second; here, I'll list both numbers to make it easier to compare disk and network throughput speeds.
SATA, or Serial ATA, has a theoretical maximum of 1200 Mbits/sec (150 MB/sec). However, existing hard drives can't even deliver data that fast; top disk output speeds are closer to 40 to 100 MB/sec, depending on whether the data is being read from the inside or outside of the disk platter, the disk spin speed, and other factors.
USB 2.0 has a theoretical maximum of 480 Mbits/sec (60 MB/sec). A USB hard drive is typically a standard ATA or SATA drive attached to a USB bridge chipset. The actual speed of the USB interface depends upon the performance of the chipset used as well as the performance of the computer the drive is attached to. That's because USB transfers most of the heavy lifting to the host computer's CPU.
USB has a faster theoretical maximum than Firewire 400 (400 Mbits/sec; 50 MB/sec), but Firewire 400 is actually much faster than USB because it uses smarter peer to peer interface hardware rather than pushing low level work onto the PC host's CPU as the simpler master to slave architecture of USB does.
On a Mac, Firewire is typically around twice as fast in real world transfer rates, with USB hitting around 18 MB/sec and Firewire reaching 35 MB/sec throughput. Windows' implementation of USB has historically been faster than Mac OS X's, with Windows' USB reaching throughput closer to 33MB/sec. That also explains why Firewire is more popular on the Mac than on the PC side; it's simply far more dramatically faster than USB on the Mac, while Firewire offers less of a noticeable boost in Windows. Macs also have Firewire Target Mode, which PC users lack. For more details on why USB is faster in Windows compared to the Mac, see the footnote: USB Performance in Windows vs Mac OS X at the end of this article.
Time Capsule doesn't use Firewire; it's USB only. There are two reasons for this. First, USB chipsets are cheaper than Firewire, because they do less (USB peripherals have less intelligence on board and transfer more work to the CPU). Second, Time Capsule and the AirPort Extreme are both designed as wireless network appliances, so the difference in performance between attached Firewire and USB drives typically wouldn't be noticeable. Test results presented in the next segment bear that out.
In reality, USB doesn't simply run at a given speed. The performance of a directly connected USB drive can be affected by a number of issues, from the performance of the host computer to interference caused by other USB devices on the same bus, to the overhead related to the drive's file system.
Ethernet Networking introduces even more complicating factors. There is the overhead of Internet Protocol addressing, as well as the file sharing protocols used, such as AFP on the Mac or SMB used by Windows, neither of which play into direct, non-networked protocols such as USB. There are also architectural issues such as the quality of the cables used and the performance of any switches (or old fashioned hubs) involved. All of these issues eat into the theoretical raw data transfer rate of Ethernet.
Fast Ethernet has a theoretical speed of 100 Mbits/sec (12 MB/sec), while Gigabit Ethernet has a theoretical speed of 1000 Mbits/sec (120 MB/sec). That suggests a double speed advantage of Gigabit Ethernet over USB (60 MB/sec), but neither protocol hits its maximum. In reality, a typical USB connected disk is roughly equal to or lesser than the throughput of a shared drive attached over a Gigabit Ethernet network.
Wireless Networking has all the complexity of traditional wired networking with the additional complications of signal strength issues such as radio interference and barriers, as well as additional overhead related to wireless transmission that commonly halves its real world throughput over the theoretical raw data rate.
- 802.11b has a theoretical speed maximum of 11 Mbits/sec with a typical transfer rate of around 4.5 Mbits/sec (0.5 MB/sec) with an ideal signal.
- 802.11g has a theoretical speed maximum of 54 Mbits/sec, with a typical transfer rate of around 23 Mbits/sec (2.5 MB/sec) with an ideal signal.
- 802.11n has a theoretical speed maximum of 300 Mbits/sec, with a typical transfer rate of around 74 Mbits/sec (9.25 MB/sec) with an ideal signal.
As the signal strength of a wireless network drops, the connection speed is automatically renegotiated and slower and slower rates until no connection is possible. The transfer rates of wireless networking make it ideal for browsing the web, as most US residents have a connection speed of around 1.5 Mbits/sec for DSL, or from 3 to 6 Mbit/sec with cable Internet service. Any version of WiFi is much faster than that.
However, very fastest wireless networking is required to perform intensive data transfers such as Time Machine backups, general file sharing, and media streaming, particularly if more than one client is using the network at once, or if one user is trying to do more than one thing with their wireless connection, such as backing up files while streaming audio to Apple TV, for example.
A Visual Speed Comparison
This chart shows the relative difference in throughput of the interfaces described above, with theoretical raw data rates in blue, and typical real world throughput in red. Note that these real world numbers are ideal peak maximums, not the average throughput users will see at all times. As detailed above, there are lots of factors that can eat into the actual real world performance. Time Capsule has performance limitations of its own, which are related to its design to primarily serve wireless clients. An upcoming segment will detail what Time Capsule itself can do.
Direct connection interfaces, such as SATA and USB, commonly deliver closer to half their theoretical maximum raw data rate, but as interfaces and drive mechanisms improve, the real world data throughput will rise. Ethernet networking interfaces, such as Fast Ethernet or Gigabit Ethernet, can hit peak transmission rates close to their maximums, but suffer from greater overhead compared to a direct connection interface.
Wireless networking throughput depends more on external factors to reach its full potential. Ideal signal strength is critically important to reach anywhere near the high end of real world throughput numbers. There are other factors that make a huge difference in wireless performance; Time Capsule and AirPort Extreme both support new features unique to the new 802.11n wireless networking protocol, including the use of multiple antennas (a technology referred to as MIMO) and the use of the 5 GHz radio spectrum. The next segment will look at the pros and cons of using this alternative frequency, which depending on the circumstances can either decrease signal range or deliver a major boost in your wireless data rate.
Footnote: USB Performance in Windows vs Mac OS X
In addition to the cabling and protocol specifics, there are other reasons for Windows PCs to outperform Macs in USB transfers. The testing done by BareFeats in the article USB 2.0 versus FireWire compared 2004 PowerPC Macs against 3 GHz Pentium 4 PCs; since USB pushes much of its work to the CPU, the speed of the host made a big difference in how fast USB performed on the two platforms.
Their testing also revealed that the first generation of the PowerMac G5 delivered poor I/O across the board, scoring lower than even the mobile PowerBook and low cost eMac in both Firewire and USB. That indicates that the theoretical expectations for USB (or any protocol) are nearly meaningless when compared to the actual speed of the disk, processor, the implementation of the protocol itself, and other factors that might cause interference or otherwise eat up the expected maximum throughput speeds. In other words, USB does not ever run at its maximum theoretical speed rating.
Additionally, Windows file sharing and disk protocols are simpler than on the Mac, because Windows handles and presents less metadata. This lightness makes for faster disk operations at the expense of the sophistication of the Mac's higher quality file icons, richer file type and creator codes, and other features missing in Windows.
There are other factors that affect cross platform throughput as well; Mac OS X suffers some degree of overhead from new features such as Spotlight indexing, while Windows PCs are typically burdened with running anti-virus scanning software that peels away a significant edge in performance. Clearly, there are lots of factors to account for in making direct performance comparisons, and neatly presented numbers can easily hide those details in a misleading way.
Previous articles related to Time Capsule and its AirPort Extreme cousin:
Exploring Time Capsule: how it fits into Apple's AirPort family
An in-depth review of Apple's 802.11n AirPort Extreme Base Station
Apple Time Capsule unboxing and preview
A Look Inside Apple's New Time Capsule
Answers to Time Capsule reader questions
Comments
On the paper, Time Capsule looked like just what I was looking for. Like 499 very well invested Euros.
But now after I got my hands on one it turned out to be a real bummer. I attached it to my network. Usually I achieve connection speeds up to 35MB per sec from one Mac to another. A pretty good speed, I guess.
But the fastest connection to my TC I have achieved yet is 14MB per sec. That's less than halve of what I get to other clients in my network. I tried different cables, different switches, but nothing. The average speed is even worse, it's around 5 MB per sec.
And as you can imagine, that is no speed to live with.
An other problem I had was that the TC started to drop connections as soon as more than one Mac connected to the share. It even rendered itself almost inaccessible when all my four Macs were connected.
The end of the story is that I will send it back next week. TimeCapsule is now officially, after almost five years of absolute pleasure, my very first Apple product that made me regret the purchase. - Happily we have the two weeks return policy in Europe...
When data is traveling in one direction, such as copying a file, the single-direction speed is half that.
So the "theoretical" speed of 802.11n is 150Mbs which is about 18MBs.
This contrasts with a real-life speed of 9MBs.
C.
I got my 1TB TimeCapsule a few days ago. And what can I say, I am really disappointed. I planned to use it as a NAS, TimeMachine storrage, printer server, as a WiFi-Acces Point and LAN-Switch and as connection to my DSL modem. The whole nine yards, just what it was built for.
On the paper, Time Capsule looked like just what I was looking for. Like 499 very well invested Euros.
But now after I got my hands on one it turned out to be a real bummer. I attached it to my network. Usually I achieve connection speeds up to 35MB per sec from one Mac to another. A pretty good speed, I guess.
But the fastest connection to my TC I have achieved yet is 14MB per sec. That's less than halve of what I get to other clients in my network. I tried different cables, different switches, but nothing. The average speed is even worse, it's around 5 MB per sec.
And as you can imagine, that is no speed to live with.
An other problem I had was that the TC started to drop connections as soon as more than one Mac connected to the share. It even rendered itself almost inaccessible when all my four Macs were connected.
The end of the story is that I will send it back next week. TimeCapsule is now officially, after almost five years of absolute pleasure, my very first Apple product that made me regret the purchase. - Happily we have the two weeks return policy in Europe...
I would have guessed this much to be true if all Macs are using wireless, but I'm curious if this is your scenario or if you are using any directly connected through the Gig ports.
Time Capsule isn't necessarily the problem, people's expectations of technology in general is far above realistic values because we dream far larger than is possible to achieve. Apple has some responsibility in this to make this more clear in their products. However, most people's unrealistic expectations are not quenched by an explanation of physical limits, they just look to the next over promising company & buy that product. As a result all companies have to play up features & performance a bit to get people to actually buy their products.
The other side of this is the "backup everything" mentality that would be fine for Gigabit connection but is completely irrational over wireless. People love to complain that this isn't properly explained to them but realistically it is all in their documentation that they choose not to read. It's like a communications e-mail to all staff in a company, the # of people that actually read it is probably less than 25%, even when it is marked critical. On top of this people then complain about not getting the message so companies also send all user voicemail, which people still manage to ignore. You really can't win.
Time Capsule is an incredible deal if you are using a direct Gigabit ethernet connection, it blows USB out of the water! This combined with comparisons on price for the equivalent # of products doing the same services is an insane deal! Problem is most people are to narrow minded & non-technical to deal with the technology that is given to them, & they are completely unwilling to take a little time to learn about it in the documentation that is freely offered.
All this being said, as technology becomes a larger part of our lives we should all make an effort understand a little bit more about how it works & what it's limits are. I'll give you that 14MB & 5MB seems pathetic next to 35MB but that performance is more likely because of the # of Macs connected at once, each one is taking part of that 1 wireless networks throughput. Apple will likely tweak performance as reports come in on the TimeCapsule from the real world but just taking it back because it falls short of unrealistic expectations is not a wise decision in my opinion. It's your money & your choice though.
I would have guessed this much to be true if all Macs are using wireless, but I'm curious if this is your scenario or if you are using any directly connected through the Gig ports.
Time Capsule isn't necessarily the problem, people's expectations of technology in general is far above realistic values because we dream far larger than is possible to achieve. Apple has some responsibility in this to make this more clear in their products. However, most people's unrealistic expectations are not quenched by an explanation of physical limits, they just look to the next over promising company & buy that product. As a result all companies have to play up features & performance a bit to get people to actually buy their products.
The other side of this is the "backup everything" mentality that would be fine for Gigabit connection but is completely irrational over wireless. People love to complain that this isn't properly explained to them but realistically it is all in their documentation that they choose not to read. It's like a communications e-mail to all staff in a company, the # of people that actually read it is probably less than 25%, even when it is marked critical. On top of this people then complain about not getting the message so companies also send all user voicemail, which people still manage to ignore. You really can't win.
Time Capsule is an incredible deal if you are using a direct Gigabit ethernet connection, it blows USB out of the water! This combined with comparisons on price for the equivalent # of products doing the same services is an insane deal! Problem is most people are to narrow minded & non-technical to deal with the technology that is given to them, & they are completely unwilling to take a little time to learn about it in the documentation that is freely offered.
All this being said, as technology becomes a larger part of our lives we should all make an effort understand a little bit more about how it works & what it's limits are. I'll give you that 14MB & 5MB seems pathetic next to 35MB but that performance is more likely because of the # of Macs connected at once, each one is taking part of that 1 wireless networks throughput. Apple will likely tweak performance as reports come in on the TimeCapsule from the real world but just taking it back because it falls short of unrealistic expectations is not a wise decision in my opinion. It's your money & your choice though.
No offense, but my "35MB per sec to other Macs" should have been a hint that I'm talking about a Gigabit-LAN connection. WiFi speeds are OK with Time Capsule, at least with my old PowerBook that doesn't even have the new WiFi standard.
I am no amateur with network technologies. I am well aware of what I can expect and what not. And a peak performance of 15MB per sec with an average performance of 5 MB per sec in a gigabit LAN environment is not OK, from any point of view. No less since I am able to reach up to 35 MB a sec with 20 MB a sec average performance in the very same network.
I know that I can't expect the same NAS performance that a dedicated server can bring, but a SINGLE LAN connection should at least serve up to 20 MB per sec with 15 MB average performance in a single direction.
I am the very last person talking negatively about an Apple product. But TimeCapsule is a sad waste of money. Maybe it's just my unit, but that's the way it is. If you want to use it just wirelessly it might be a fine deal, though!
Time Capsule, announced earlier this year, is a base station with an integrated hard drive and power supply. The previous segment of this series exploring Time Capsule in depth looked at the differences in members of the AirPort family. This segment, the second of six, compares the differences between the hypothetical maximum data transmission speed and typical real world performance of Time Capsule's SATA, USB 2.0, Ethernet networking, and WiFi Wireless networking interfaces.
In Theory, Theory and Practice Are the Same. In Practice, They Are Not
The theoretical speed rating of a given wired or wireless connection can be betrayed by a number of factors, from the overhead of the protocols involved to signal interference. Network data throughput is usually measured in megabits per second, which are an eighth of a megabyte per second. Disk speeds are typically cited in megabytes per second; here, I'll list both numbers to make it easier to compare disk and network throughput speeds.
SATA, or Serial ATA, has a theoretical maximum of 1200 Mbits/sec (150 MB/sec). However, existing hard drives can't even deliver data that fast; top disk output speeds are closer to 40 to 100 MB/sec, depending on whether the data is being read from the inside or outside of the disk platter, the disk spin speed, and other factors.
USB 2.0 has a theoretical maximum of 480 Mbits/sec (60 MB/sec). A USB hard drive is typically a standard ATA or SATA drive attached to a USB bridge chipset. The actual speed of the USB interface depends upon the performance of the chipset used as well as the performance of the computer the drive is attached to. That's because USB transfers most of the heavy lifting to the host computer's CPU.
USB has a faster theoretical maximum than Firewire 400 (400 Mbits/sec; 50 MB/sec), but Firewire 400 is actually much faster than USB because it uses smarter peer to peer interface hardware rather than pushing low level work onto the PC host's CPU as the simpler master to slave architecture of USB does.
On a Mac, Firewire is typically around twice as fast in real world transfer rates, with USB hitting around 18 MB/sec and Firewire reaching 35 MB/sec throughput. Windows' implementation of USB has historically been faster than Mac OS X's, with Windows' USB reaching throughput closer to 33MB/sec. That also explains why Firewire is more popular on the Mac than on the PC side; it's simply far more dramatically faster than USB on the Mac, while Firewire offers less of a noticeable boost in Windows. Macs also have Firewire Target Mode, which PC users lack. For more details on why USB is faster in Windows compared to the Mac, see the footnote: USB Performance in Windows vs Mac OS X at the end of this article.
Time Capsule doesn't use Firewire; it's USB only. There are two reasons for this. First, USB chipsets are cheaper than Firewire, because they do less (USB peripherals have less intelligence on board and transfer more work to the CPU). Second, Time Capsule and the AirPort Extreme are both designed as wireless network appliances, so the difference in performance between attached Firewire and USB drives typically wouldn't be noticeable. Test results presented in the next segment bear that out.
In reality, USB doesn't simply run at a given speed. The performance of a directly connected USB drive can be affected by a number of issues, from the performance of the host computer to interference caused by other USB devices on the same bus, to the overhead related to the drive's file system.
Ethernet Networking introduces even more complicating factors. There is the overhead of Internet Protocol addressing, as well as the file sharing protocols used, such as AFP on the Mac or SMB used by Windows, neither of which play into direct, non-networked protocols such as USB. There are also architectural issues such as the quality of the cables used and the performance of any switches (or old fashioned hubs) involved. All of these issues eat into the theoretical raw data transfer rate of Ethernet.
Fast Ethernet has a theoretical speed of 100 Mbits/sec (12 MB/sec), while Gigabit Ethernet has a theoretical speed of 1000 Mbits/sec (120 MB/sec). That suggests a double speed advantage of Gigabit Ethernet over USB (60 MB/sec), but neither protocol hits its maximum. In reality, a typical USB connected disk is roughly equal to or lesser than the throughput of a shared drive attached over a Gigabit Ethernet network.
Wireless Networking has all the complexity of traditional wired networking with the additional complications of signal strength issues such as radio interference and barriers, as well as additional overhead related to wireless transmission that commonly halves its real world throughput over the theoretical raw data rate.
- 802.11b has a theoretical speed maximum of 11 Mbits/sec with a typical transfer rate of around 4.5 Mbits/sec (0.5 MB/sec) with an ideal signal.
- 802.11g has a theoretical speed maximum of 54 Mbits/sec, with a typical transfer rate of around 23 Mbits/sec (2.5 MB/sec) with an ideal signal.
- 802.11n has a theoretical speed maximum of 300 Mbits/sec, with a typical transfer rate of around 74 Mbits/sec (9.25 MB/sec) with an ideal signal.
As the signal strength of a wireless network drops, the connection speed is automatically renegotiated and slower and slower rates until no connection is possible. The transfer rates of wireless networking make it ideal for browsing the web, as most US residents have a connection speed of around 1.5 Mbits/sec for DSL, or from 3 to 6 Mbit/sec with cable Internet service. Any version of WiFi is much faster than that.
However, very fastest wireless networking is required to perform intensive data transfers such as Time Machine backups, general file sharing, and media streaming, particularly if more than one client is using the network at once, or if one user is trying to do more than one thing with their wireless connection, such as backing up files while streaming audio to Apple TV, for example.
A Visual Speed Comparison
This chart shows the relative difference in throughput of the interfaces described above, with theoretical raw data rates in blue, and typical real world throughput in red. Note that these real world numbers are ideal peek maximums, not the average throughput users will see at all times. As detailed above, there are lots of factors that can eat into the actual real world performance. Time Capsule has performance limitations of its own, which are related to its design to primarily serve wireless clients. An upcoming segment will detail what Time Capsule itself can do.
Direct connection interfaces, such as SATA and USB, commonly deliver closer to half their theoretical maximum raw data rate, but as interfaces and drive mechanisms improve, the real world data throughput will rise. Ethernet networking interfaces, such as Fast Ethernet or Gigabit Ethernet, can hit peak transmission rates close to their maximums, but suffer from greater overhead compared to a direct connection interface.
Wireless networking throughput depends more on external factors to reach its full potential. Ideal signal strength is critically important to reach anywhere near the high end of real world throughput numbers. There are other factors that make a huge difference in wireless performance; Time Capsule and AirPort Extreme both support new features unique to the new 802.11n wireless networking protocol, including the use of multiple antennas (a technology referred to as MIMO) and the use of the 5 GHz radio spectrum. The next segment will look at the pros and cons of using this alternative frequency, which depending on the circumstances can either decrease signal range or deliver a major boost in your wireless data rate.
Footnote: USB Performance in Windows vs Mac OS X
In addition to the cabling and protocol specifics, there are other reasons for Windows PCs to outperform Macs in USB transfers. The testing done by BareFeats in the article USB 2.0 versus FireWire compared 2004 PowerPC Macs against 3 GHz Pentium 4 PCs; since USB pushes much of its work to the CPU, the speed of the host made a big difference in how fast USB performed on the two platforms.
Their testing also revealed that the first generation of the PowerMac G5 delivered poor I/O across the board, scoring lower than even the mobile PowerBook and low cost eMac in both Firewire and USB. That indicates that the theoretical expectations for USB (or any protocol) are nearly meaningless when compared to the actual speed of the disk, processor, the implementation of the protocol itself, and other factors that might cause interference or otherwise eat up the expected maximum throughput speeds. In other words, USB does not ever run at its maximum theoretical speed rating.
Additionally, Windows file sharing and disk protocols are simpler than on the Mac, because Windows handles and presents less metadata. This lightness makes for faster disk operations at the expense of the sophistication of the Mac's higher quality file icons, richer file type and creator codes, and other features missing in Windows.
There are other factors that affect cross platform throughput as well; Mac OS X suffers some degree of overhead from new features such as Spotlight indexing, while Windows PCs are typically burdened with running anti-virus scanning software that peels away a significant edge in performance. Clearly, there are lots of factors to account for in making direct performance comparisons, and neatly presented numbers can easily hide those details in a misleading way.
Previous articles related to Time Capsule and its AirPort Extreme cousin:
Exploring Time Capsule: how it fits into Apple's AirPort family
An in-depth review of Apple's 802.11n AirPort Extreme Base Station
Apple Time Capsule unboxing and preview
A Look Inside Apple's New Time Capsule
Answers to Time Capsule reader questions
[ View this article at AppleInsider.com ]
Hi, my question relates to the speed of backups with Time Capsule vs. an AirPort Extreme Base Station. An earlier article about Time Capsule included the statement that: "Time Capsule uses a direct SATA connection to its internal hard drive, suggesting the potential for faster drive access compared to an externally connected USB disk, the only disk sharing option available to AirPort Extreme base stations."
I thought the AirPort Extreme Base Station allowed the connection of a NAS device such as the Drobo/Droboshare to one of the Base Station's Gigabit Ethernet LAN ports, which would make that NAS storage accessible to all Macs on the AEBS Gigabit network. Canon advised me its networkable printers would be accessible to networked computers in this manner, so I assumed a NAS device would work the same way.
Is that idea incorrect?
Thanks.
As I understand it, the performance is not limited by USB / SATA or Gigabit. The limit is caused by the relatively weak processor in the Airport Extreme or Time Capsule. It struggles to transfer data and organize it into a file-system.
In a typical wireless installation, 9MB/s is going to give users wireless backup that is about half as fast as a local noisy USB drive. Which makes it an attractive product.
As a NAS on a gigabit network, I don't think it's nearly so competitive.
C.
You need less of a pixel offset and more of a blur radius on those drop shadows for the numbers in your graph. That will make them much easier to read.
Thanks,
Cubert
Whimsical,
As I understand it, the performance is not limited by USB / SATA or Gigabit. The limit is caused by the relatively weak processor in the Airport Extreme or Time Capsule. It struggles to transfer data and organize it into a file-system.
In a typical wireless installation, 9MB/s is going to give users wireless backup that is about half as fast as a local noisy USB drive. Which makes it an attractive product.
As a NAS on a gigabit network, I don't think it's nearly so competitive.
C.
Absolutely my point. This thing costs 499? including tax. That is about 750$. IMHO enough money to expect a better controller that is able to handle the data at a reasonable rate.
Whimsical,
As I understand it, the performance is not limited by USB / SATA or Gigabit. The limit is caused by the relatively weak processor in the Airport Extreme or Time Capsule. It struggles to transfer data and organize it into a file-system.
In a typical wireless installation, 9MB/s is going to give users wireless backup that is about half as fast as a local noisy USB drive. Which makes it an attractive product.
As a NAS on a gigabit network, I don't think it's nearly so competitive.
C.
I agree that it seems Time Capsule's (and the Extreme's) processor is the weak link in all of this. And related to that is this quote from the story:
"Time Capsule and the AirPort Extreme are both designed as wireless network appliances, so the difference in performance between attached Firewire and USB drives typically wouldn't be noticeable"
I disagree on two points: 1) it has LAN ports so it's not strictly a "wireless network appliance", and 2) that there would be no difference between USB and Firewire even if it was just a wireless appliance.
As the article stated several times, the big difference between USB and Firewire is that a USB chip offloads much of the work to the CPU while the FW chip handles the workload itself. This makes the power of TC's processor all the more critical. If TC had a more powerful processor, there would maybe be less of a performance difference between USB and FW. But since they put an underpowered processor in it, I believe a FW connection would have allowed much better throughput to the disk.
To test this, I had my MBP connected wirelessly (N) to an Airport Extreme (all the latest firmware) and tested transfer times from the MBP to and from: 1) a disk connected to the Extreme's USB port, and 2) the same disk connected to a PowerBook via FW400 with the PB connected to the Extreme via Gb Ethernet.
Despite the data having to travel over the extra Ethernet leg of the network and being handled by a 4-year old PB before being sent to the FW drive, the throughput was between 2-4 times faster than to the Extreme's USB port. So clearly, the wireless network is not the only bottleneck. The USB controller/CPU combo is also a bottleneck.
Note: wireless is a bottleneck, just not the only one. Connecting the MBP via Gb Ethernet roughly doubles the performance of the USB port. But it also doubles, and in some case triples, the performance of the FW drive connected to the PB.
The real reason Apple probably chose USB was to enable network printing capability. If they had included FW for AirDisk, they still would have had to include USB for printing, increasing the cost and complexity of the device.
Edit: If anyone cares, my informal tests included two folders of files. One folder had about 170 MB of 1400 small files which would be a typical backup scenario (I'm a little unclear how TM does network backups. It may combine files into larger chucks of data in the sparse disk image it creates). A second folder was about 850 MB cosisting of 110 photos, each about 8 MB, representing backing up my Aperture library.
Absolutely my point. This thing costs 499? including tax. That is about 750$. IMHO enough money to expect a better controller that is able to handle the data at a reasonable rate.
Well, then take it back. FWIW, I have a 500 gig TC and I'm fine with its performance for what it is made for. Contrary to the notion that this thing is supposed to be some sort of NAS, I view it for what it is - a network attached backup disk. For that (where you don't care about speed as long as it gets done and gets done right), it works perfectly with next to no setup.
You can pick up a ReadyNAS 1TB for about $1100 I guess. That's not much more and would probably match what you're looking for.
Absolutely my point. This thing costs 499? including tax. That is about 750$. IMHO enough money to expect a better controller that is able to handle the data at a reasonable rate.
Yes, at roughtly $750, I've got a linux PC with a 5 disk raid setup next to my furnace. It does applebabble (AFP), samba (SMB), and nfsmount. It runs two different media servers (mt-daapd, and slimserver) as well as doing all the usual linux server stuff: mail server, web server, time server, dns etc. I backup to it using timemachine and it's been working flawlessly as a timemachine backup since November 8th. In addition, it's got lots of server type features baked in, like emailing me in when a drive is failing, or the raid set has an issue.
My performance on my linux box is way better than my AEBS with disk, it's not even close.
The advantage in the time capsule is that it's small, quiet, and easy. Those are big things for many people. But it's hardly state of the art when compared to a larger box on price and performance alone. small light and easy cost money.
Sheldon
Wireless speeds (300Mbs / 54Mbs ) refer to the bi-directional speed of the network.
When data is traveling in one direction, such as copying a file, the single-direction speed is half that.
So the "theoretical" speed of 802.11n is 150Mbs which is about 18MBs.
This contrasts with a real-life speed of 9MBs.
It has nothing to do with bidirectional, it's just the signalling frequency. It's a half-duplex medium, if the bulk of the transfers is just one direction, it should be better than if transfers are both ways. Take away payload, interference, distance penalties, anti-collision measure penalties, acknowledgements, among other things, it goes down a lot.
No offense, but my "35MB per sec to other Macs" should have been a hint that I'm talking about a Gigabit-LAN connection. WiFi speeds are OK with Time Capsule, at least with my old PowerBook that doesn't even have the new WiFi standard.
I am no amateur with network technologies. I am well aware of what I can expect and what not. And a peak performance of 15MB per sec with an average performance of 5 MB per sec in a gigabit LAN environment is not OK, from any point of view. No less since I am able to reach up to 35 MB a sec with 20 MB a sec average performance in the very same network.
I know that I can't expect the same NAS performance that a dedicated server can bring, but a SINGLE LAN connection should at least serve up to 20 MB per sec with 15 MB average performance in a single direction.
I am the very last person talking negatively about an Apple product. But TimeCapsule is a sad waste of money. Maybe it's just my unit, but that's the way it is. If you want to use it just wirelessly it might be a fine deal, though!
I stand corrected & man enough to say when I am wrong. Though I referenced the 35MB I glazed it over thinking of 15 & 5 not sounding to terribly off from a wireless n connection.
I still wish to hold to some of my points though as a fellow network knowledgeable I do apologize for assuming you a simple user. Hope you can forgive me.
Seeing the real world speeds for this has moved me personally in the direction of an external drive attached to my iMac & then shared to the rest of my Mac world. Time Capsule has some cool features in is a cool all in one product but if the Gigabit connection does not achieve full speeds I am also out.
I would say though that if I had no wireless router at all & only had laptops in my house this would be a very reasonable expense to me given the full feature set I'd get with it.
The article about USB on Macs vs Windows was an old one & based on PPC. USB on Intel Macs is much faster & I find it disturbing this was so easily misrepresented. Keep facts straight people.
I have a 17" MacbookPro 2.16GHz and I can tell you that I've done some simple tests transferring 2 to 3 gb video files over USB2 and the Windows computer is ALWAYS faster....significantly so. I've also done the same tests on a Macbook 2.0GHz with the same results.
In my opinion, except for video editing, USB2 works just fine on my Mac. I've got 500gb of video that I use daily. Works perfectly. By the way, I recommend the Seagate Free Agent Pro external hard drives. I've got a 750gb unit that works amazingly well. (Love the power sensing, only spins-up when it needs to)
However, when Time Machine is doing an incremental backup, I get nowhere near those transmission speeds. First it takes ages for the backup to commence, sometimes more than 20 minutes while it just says "preparing backup". And the transmission rates during the backup are simply terrible. Somtimes a 1.5 MB incremental backup takes more than an hour!
So for some reason there seems to be a huge difference between copying files manually to the TC and doing so via the automatic TM backup.
Does anybody have an idea as to why that is?
BTW, I have a 12" G4 PB, which of course doesn't use the n-standard. Signal strength is excellent though.
PS: I manually started an incremental backup before writing this post. It still says "preparing backup" - since more than seven minutes now. Something is clearly wrong here...
I just did some testing. The transmission speed for copying files onto the TC wirelessly is not great but okay for a usual incremental backup. In theory. I just copied a folder with a couple of movie files totalling 114.5 MB in 87 seconds which translates to 1.31 MB/sec.
However, when Time Machine is doing an incremental backup, I get nowhere near those transmission speeds. First it takes ages for the backup to commence, sometimes more than 20 minutes while it just says "preparing backup". And the transmission rates during the backup are simply terrible. Somtimes a 1.5 MB incremental backup takes more than an hour!
So for some reason there seems to be a huge difference between copying files manually to the TC and doing so via the automatic TM backup.
Does anybody have an idea as to why that is?
BTW, I have a 12" G4 PB, which of course doesn't use the n-standard. Signal strength is excellent though.
PS: I manually started an incremental backup before writing this post. It still says "preparing backup" - since more than seven minutes now. Something is clearly wrong here...
Take a look at the activity monitor/Network activity when doing the backups. There is a lot of low-level activity which i interpret as file matching. I read somewhere (and I can't find it) where the FSEvents, which tracks what files have changed since the last backup, compromises speed and completeness by tracking folders which have changed vs every single file. This would mean that TM would have to do some level of comparison, at the folder level, to determine just what files were changed vs. not. This would require significant network activity to get the in formation from the remote and compare it. This though is consistent with what I observe in my TM on my macbook pro over 802.11n. If I'm at home and doing and turn TM on so its doing backups every hour, backups are fairly quick and reasonable small (10's of MB). If I've been away for a while the first backup takes a lot longer, even if its only 100 MB or so, and its almost the same amount of time if I've had large file changes (like installing iPhone SDK which resulted in a 4.7GB incremental). I've come to think the amount of time is more related to the number of files that need to be backed up rather than the size.
Overall I can say I am very very pleased with TM over 802.11n. The sometimes long times for TM 'preparing backup' while maybe puzzling are more than acceptable IMO because
1) It all just works
2) The priority seems to be set such that I don't seen any interference with ongoing use of the system. I do video compression and viewing at the same time and don't see changes in timing or hesitation when TM is running. This may also account for the length of time it sometimes takes.
Note, I was also doing very well over 802.11g (before I upgraded my original MBP to 802.11n) but I did do the original, first, backup over a wired connection.
But the fastest connection to my TC I have achieved yet is 14MB per sec. That's less than halve of what I get to other clients in my network. I tried different cables, different switches, but nothing. The average speed is even worse, it's around 5 MB per sec.
And as you can imagine, that is no speed to live with.
I don't think I've ever gotten more than 5 MB per sec for anything on my network. But there have been no new Apple products in my household since the Intel transition; though I wish there were