Seagate FreeAgent GoFlex Desk 3 TB USB 3.0 External Hard Drive STAC3000101

The other day, I bought 6 TB of storage for under $300. This statement alone is startling to folks like me who have been following the storage and hard disk drive industry, but there is another fact that is more difficult to swallow: It will take days to copy data to these huge 3 TB drives, since the USB interface on each Seagate GoFlex Desk drive is good for just 30 MB/s on Apple Macintosh computers. Searching for a faster alternative led me to crack open the case and experiment with the drive inside.

Introducing the GoFlex Desk

You should read Lemons Into Lemonade: Seagate Repackages SATA As GoFlex before you continue…

I purchased two 3 TB Seagate GoFlex Desk drives from a local retailer for just $139 each. This is an amazingly cheap way to get 6 TB of storage!

My goal is to back up all of my Tech Field Day video to the two drives, using rsync to ensure that each contains a full exact copy of the video data folders. I’ll then store one off-site in a fireproof box for extra protection.

I selected the Seagate GoFlex Desk based on my good experience with their portable line of GoFlex drives. I liked the idea that the drives can be connected to a faster interface (FireWire 800, for example) for filling and then use a slower, cheaper one (USB 2.0) to read the data later or in another location.

The capacity of these drives is simply astonishing, but I question the design. The drive sits in a sealed plastic box with little ventilation, and it got hot to the touch during active use. The interchangeable docks are great, but I was disappointed that the FireWire dock has just a single port – I couldn’t daisy-chain FireWire off my iMac for data transfer, so I was stuck with USB 2.0.

Opening The Case

You might also want to read How To Add An eSATA Port To An Intel iMac

I decided to try connecting the drive to another interface for the copy operation. I had an eSATA dock handy, and my iMac has a DIY eSATA port, but this required removing the drive from its plastic container. Here’s how I accomplished that task.

Note that this likely voids the warranty on the drive, and I found that it did not function properly anyway. More on that later, though.

Step 1: Crack the Case

First, we must crack open the plastic case. The case splits in half along the seams, as one might assume. To locate the top, place the drive flat on a table with the GoFlex (SATA) port on the bottom. We will be removing the top of the case from this perspective.

Using a broad, flat spudger or putty knife, press firmly at the top of the seam in the case to release the clips inside. You have to press very firmly, but the clips will give way one by one.

The clips are more visible in the image above.  Repeat the process on the other side, and pry apart the ends.

Remove the Drive

Now that we have the plastic case open, we can remove the hard disk drive itself from the inner steel case.

Although the drive appears to be easy to remove, it is bolted into a three-sided steel case. Pull it free from the plastic case and we can begin to extract it.

Pry off the rubber bumpers or feet you see and you will discover a screw beneath each one. Unscrew all four and you can extract the hard disk drive itself.

Seagate Barracuda ST3000DM001

Turns out this was one of Seagate’s new Barracuda drives! See No More Green Drives from Seagate for more info!

Inside my 3 TB GoFlex Desk I found a Seagate Barracuda drive, presumably a 7200 rpm Barracuda XT. But the disk, model ST3000DM001, is not listed on Seagate’s web site. I presume it’s a special OEM drive not intended for consumer use apart from the GoFlex system.

Surprisingly, this apparently is not an Advanced Format (4K sector) drive. It reported 512 KB sectors. More interestingly, although I reformatted it with GPT, the drive itself appeared to be have MBR format, something that shouldn’t work with a 3 TB drive. Seagate is doing some special mojo here.

This meant that the drive did not function correctly when directly connected with SATA. Though I probably could have reformatted it fresh, it would probably not work with the GoFlex dock then. It also did not function with the portable GoFlex adapter, and just attempting this required a complex cabling setup between that adapter and the drive since it requires more power than USB can deliver.

Stephen’s Stance

Buying 3 TB of storage for less than $150 is a modern miracle, and I’m happy with these drives as purchased. But cracking them open isn’t all that worthwhile, since the format requires the GoFlex Dock adapter. I could wipe them entirely, of course, but that defeats my intended use. So I repacked the drive in its plastic box and will rely on the official connectivity method.

© sfoskett for Stephen Foskett, Pack Rat, 2011. | How To Open a Seagate GoFlex Desk Hard Disk Drive Case
This post was categorized as Apple, Everything, Personal, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Is Apple already abandoning the "blade" SSD in the MacBook Air?

Apple is once again on the cutting edge of PC design, reportedly ditching traditional storage controllers and turning to motherboard mounted flash storage for the new MacBook Air. Although most discussion so far has focused on the merits of this particular solution, the repercussions of such a move go far beyond Apple’s sub notebook and point to a new era when SCSI and ATA interfaces no longer dominate.

What Apple May Do

We will not know for sure until reviewers get their hands on the expected MacBook Air refresh later this month, but rumors suggest that Apple will do away with the SATA controller when they moved to Intel’s Sandy Bridge CPU architecture.

Readers of this blog may recall that Apple previously switched to an all SSD storage lineup in the last MacBook Air refresh. This system used “blade” SSDs, but these include a traditional SATA controller and interface to the motherboard which is a substantial performance bottleneck. Reviews show that the MacBook Air, while quick, does not reach the level of I/O performance theoretically possible from flash memory.

If Apple integrates the flash controller and NAND modules directly onto the motherboard (and logically attached to the PCI express bus), performance will improve dramatically. Users of enterprise PCIe SSDs have already seen the incredible performance that these solutions are capable of. It is not exaggeration to say that a PCIe SSD is as much of an upgrade over a SATA or SAS SSD as they are over a traditional rotational hard disk drive.

Industry Implications

Moving from an SATA SSD to a true integrated storage design is not a trivial task. Flash memory controllers are still required to manage the unique characteristics of the chips involved, and the operating system drivers must be modified to address memory directly or through these new controllers.

Apple is ideally suited to making this shift, since they control both hardware and software design. It would be much more difficult for HP, IBM, or Dell to make this move, since they would need to coordinate with Microsoft to produce a successful product. Conversely, Microsoft would be hard-pressed to demand that their hardware OEMs make such a change since it would require extensive engineering and testing on their part.

If Apple does indeed abandon a traditional SATA interface for storage on the MacBook Air, they may consider doing the same across their entire product portfolio. The iOS devices (the iPhone, iPod, iPad, and Apple TV) have already made this shift, another engineering advantage for Apple. One imagines a next-generation line of MacBook Pro and iMac computers with lightning fast SSD in addition to traditional SATA optical and hard disk drives.

These new computers from Apple will perform so much better than similarly specified Windows PCs that the entire industry will be forced to make a similar shift. But Apple already holds a dominant position in the NAND flash memory industry thanks to the purchasing might that comes from iOS. This puts every other computer maker at a disadvantage both financially and in terms of order fulfillment.

Stephen’s Stance

As techies moan about the lack of upgrade options presented by a soldered in SSD, they miss the bigger industry picture. For too long, computers have been held back by traditional SCSI and ATA controllers. These are both a performance bottleneck and an impediment to innovation. A shift to an integrated PCI storage model makes much sense tactically and strategically for Apple, and I expect that these rumors are true. Furthermore, this move will put even more stress on Windows PC makers. Once again, Apple is outmaneuvering the competition.

© sfoskett for Stephen Foskett, Pack Rat, 2011. | Implications of the 2011 MacBook Air’s Unconventional SSD
This post was categorized as Apple, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Would you run VMware ESX on SATA? You might in a lab environment! Here's what to look for.

The VMware ESX hardware compatibility list is awesome but it’s kind of hard to wade through. It’s super-detailed, but difficult to navigate if one is browsing for compatible hardware. Although SATA and especially PATA aren’t exactly mainstream in enterprise datacenters, they’re the most-likely storage attachment for labs and tinkerers like me.

So I decided to put together a “cheat sheet” listing the compatible SATA and PATA chipsets. In the spirit of openness, I’m presenting this data here for all to see, and I welcome corrections and updates. Indeed, I’ll try to keep this page up to date as new hardware is added!

PATA Drivers for ESX

Parallel ATA isn’t widely available anymore, but those desiring to run ESX on older hardware will want to make sure it uses one of the following controllers. Sadly, there aren’t many of them, and not many “hobbyist” motherboards use these specific chipsets. But that’s the fact of it.

• AMD SB700/SP5100 and AMD SB800 series
• Intel ICH7

That’s it. Three chipsets from two vendors. Both are server-oriented, too, so they’re harder to find in cheaper desktop motherboards.

You’re likely safe if you use an Intel ICH7 server board, but most use other-brand controllers that won’t work without some hacking. And you might not want to hack on your storage drivers…

SATA Drivers for ESX

The situation is a little brighter for the new SATA standard. Although lots of serial ATA controllers remain unsupported, there are enough here that an average shopper ought to be able to spot one of them on a motherboard.

Again, buying an Intel board is preferred, though the latest Sandy Bridge chipsets (P67/H67) are notably absent. I’ve heard that the controllers may function fine, however.

• AMD SB700/SP5100
• Broadcom BCM HT1000, HT1100 (aka ServerWorks)
• Intel ESB2 (ICH6), ICH7, ICH9, ICH10 (X58)
• nVidia MCP55

Stephen’s Stance

No enterprise will probably use non-RAID SATA for a production VMware ESX server, but it’s interesting to know what’s supported. Although most of these are fairly dated (the latest platforms are notably absent), they may be backwards-compatible with the items on this list. I’d love to hear from folks who have successfully run ESX 4 on Intel’s new PCH controller found in the P67, or H67 boards specifically!

© sfoskett for Stephen Foskett, Pack Rat, 2011. | VMware ESX SATA and PATA Compatibility Cheat Sheet
This post was categorized as Enterprise storage, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Light Peak doesn't really need all that optical technology, so why use light at all?

As I considered the possibilities of the new Apple/Intel interconnect technology known as Light Peak, an odd parallel with 10 Gb Ethernet popped into my head. Much of the confusion around Light Peak revolves around connectors, power conduction, and backward-compatibility. Then, like the Grinch, I thought of something I hadn’t before: Why use optical at all? 10 GBASE-T does just fine over twisted pair, and short interconnect distances would reduce power draw to reasonable levels. What if Light Peak was electrical rather than optical?

Optical: Perennial Underdog of Local Connectivity

In 1996, I worked at an R&D lab for famed connectivity company, US Robotics. At that time, the transition from 10 Mb Ethernet to 100 Mb “Fast Ethernet” was under way, and I recall one of the switching engineers (yes, USR produced Ethernet switches before 3COM) patiently explaining that a transition to optical interconnects was inevitable. Twisted pair wiring, he explained, just couldn’t handle the high frequencies needed at mind-bending speeds like that. And the voltage required to send such a signal 100 meters over 22 gauge wire would make port density unacceptable for data center use.

We all know how this turned out: Many facilities (including those where I worked) invested in optical cabling to “future proof” the in-wall infrastructure, only to see 100 BASE-TX Fast Ethernet over twisted pair stomp all competitors. The same story came with Gigabit Ethernet in the late 1990′s, and yet copper 1000 BASE-T is widely used. Cabling has improved (today’s Cat 6a can handle the 500 MHz of 10 GBASE-T) and electrical engineering has worked wonders to make 10 Gb copper practical.

In short, optical cable has always been exposed as an unnecessary luxury in local interconnects. Optical TOSLINK or S/PDIF cables are common in home theater applications (though purists actually prefer coaxial copper), and optical interconnects are used in storage area networking (SAN) and other high-performance networking applications. These are mainly historical anomalies, however: In both cases, the needed bandwidth pushed the capabilities of copper cable at the time, but improvements have rendered the use of optical interconnects moot.

Light Peak Over Copper

I wrote about Light Peak/USB 3.0 convergence yesterday

The same is true for Light Peak. While an optical interconnect seems like a sure-fire way to bring massive bandwidth and consolidate ports on a computer, it’s neither necessary nor really all that valuable. Light Peak could easily use twisted pair or coaxial cable, especially over the short runs that personal computers require. While it’s interesting to see 10 Gb carried over a 30 meter cable, no home or office user would need this. And they’re supposed to be Light Peak’s target audience!

A “Light Peak over Copper” spec for a maximum 10 meter run between repeaters would be sufficient and would reduce the cost and complexity of the whole system. Rather than cobble together an optical-plus-copper interconnect, Light Peak over copper would carry both high-speed multiplexed data and reasonable electrical current to power attached peripherals. It could even be made natively backward-compatible with some existing spec like USB 3.0 or HDMI!

Ever heard of HDBaseT?

Light Peak over Copper could still carry multiple protocols at a data rate of 10 Gb/s. It could still enable single-port laptop-to-dock (or monitor) connectivity. But it would also be useful as an internal connection (replacing SATA) and as a power transmission system. What if Apple’s next MagSafe connector included power as well as USB, FireWire, HDMI, and DisplayPort signals? This is absolutely feasible and could be delivered at low cost. Optical Light Peak requires components from a number of manufacturers and demands precise plug/receptacle tolerances, while copper could be delivered as cheaply as USB.

Stephen’s Stance

I love the concept of one-wire connections, especially for laptop computers. It would be awesome (and very Jobsian) if my next MacBook Pro had a single port for every type of connectivity, from power to networking to display to storage. But this doesn’t require a nifty new optical connector; it just requires high bandwidth. Light Peak over Copper would do all that, and is far more practical than the “science project” systems demonstrated by Intel so far. Let’s have it!

© sfoskett for Stephen Foskett, Pack Rat, 2010. | What If Light Peak Was Electrical Rather Than Optical?
This post was categorized as Apple, Computer History, Personal, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

The ioSafe SoloPRO protects your data from a house on fire. Seriously! That's really what it does!

It’s hard to stand out in the world of external storage devices, and doubly-hard to compete with the hard disk drive makers themselves. This hasn’t stopped folks like Iomega, Verbatim, and LaCie from trying to impress customers with flashy cases, software bundles, and clever functionality. But clever new twist on the external hard drive concept just rolled into the Pack Rat lair: The ioSafe SoloPRO is fireproof and waterproof. Cool!

A New Kind of Data Protection

I’m an enterprise storage specialist. I talk about backups, snapshots, mirroring, replication, and archiving all the time. I’ve also delved deeply into the physical reliability of hard disk drives versus flash, tape, and RAID systems. But pretty much every data protection conversation takes for granted that storage is not meant to be robust. We just assume that the disk itself will be lost in the event of fire or flood, so we had better figure some way to protect the data off-site.

But what id this wasn’t the case? What if the disk itself could survive a house fire, standing up to the heat, crushing weight of collapse, and a thorough soaking from the fire hose? That’s exactly what ioSafe is promising with their line of storage devices!

The ioSafe line has a few tricks up its sleeve:

1. The hard disk drive mechanism is wrapped in a “HydroSafe” water barrier
2. This is encased in a thick layer of “DataCast” endothermic insulation, releasing water which evaporates and cools the drive when exposed to extreme heat
3. The “FloSafe” cooling channels allow airflow during normal operation but seal shut in a fire
4. A tough steel case surrounds everything, offering some protection against physical damage and allowing the device to be bolted to the floor for theft protection
5. Every ioSafe drive also includes data recovery services in the event of a fire or drive failure

All these CamelCase trademarks appear to work just fine, thank you. Numerous tests have been performed by amused journalists and bloggers, including backyard fires, dousing the unit in a swimming pool, and running over it with a bulldozer. In each case, the housing is ruined but the data survives. My favorite is the following video from HomeServerReview, but a quick glance through YouTube is worth the time if you like watching people destroy perfectly-good technology!

The Pack Rat Test: Performance

When ioSafe offered to send me a SoloPRO for evaluation, I was excited to burn, drown, and mutilate it. But considering just how many torture tests were already performed, I decided to give that a miss. Instead, I hooked the SoloPRO to my test rig to see how well it handled everyday storage tasks. After all, most owners will never experience the kind of damage the ioSafe line can sustain!

My tests used Intech Software’s ZoneBench and QuickBench tools on a late-2009 iMac. The ioSafe SoloPRO was connected to the iMac’s secondary internal SATA port using my iMac eSATA mod. This is a 3 Gb/s SATA connection and ought to sustain just about anything a spinning disk can currently sustain. Tests were performed on a freshly-booted Mac OS X 10.6.4 64-bit system with no other programs running.

32 MB transfers reveal the performance limits of the ioSafe controller and Hitachi hard disk drive. Performance was excellent - better than the 7200 rpm Seagate drive in the iMac!

Sustained throughput was excellent as well, delivering between 130 and 140 MB/s in my tests. Real-world transfers were just as quick!

The SoloPRO uses a 7200 rpm Hitachi DeskStar 7K1000.C (HDS721010CLA332) hard disk drive. This is a modern, quick mechanism with a 32 MB buffer and 3.0 Gb/s SATA interface. Kudos to ioSafe for picking such a solid performer for this unit – I expected a slower 5400 rpm “green” drive given the modest performance expected by most buyers.

Connected to my iMac with eSATA, the SoloPRO could outrun the internal hard disk drive, both in benchmarks and real-world use. Moving data back and forth was a joy, and backing up the internal disk using Time Machine was amazing: I averaged over 100 MB/s, with Activity Monitor showing frequent peaks over 160 MB/s during the operation!

Although I did perform these tests using the USB interface, the outcome is predictably disappointing. USB 2.0 just can’t go faster than about 35 MB/s in sequential throughput, so the graphs and comparisons look awfully predictable. That said, the ioSafe SoloPRO was able to hit that 35 MB/s mark with ease.

The Pack Rat Test: Usability

The SoloPRO was delivered in a large box with generous packing material. It’s amazingly large, really, considering that it contains a single 3.5″ hard disk drive mechanism. ioSafe includes both USB and eSATA cables, though the latter cable was the short-necked “type I” variety.

The SoloPRO is larger and heaver even than it looks, with solid build quality. The power supply and switch aren’t much to get excited about, however, but no cheaper than the power devices most other manufacturers use. I would have like a power connector that locked in more securely, however.

I appreciate the protruding steel flange with holes for floor-mounting, and finally located a Kensington-style lock hole next to the fan. I was surprised that the holes in the face do not light up when using eSATA, though they do glow while using USB. On the other hand, the flashing light show is pretty distracting for a desktop drive.

The SoloPRO is somewhat noisy, though not any more than a desktop PC. The built-in fan runs continually, though the drive itself will spin down if the computer allows it to. Due to the short range of the cables, an under-desk location is about all an eSATA user can hope for. USB users will likely move it a bit further away so they won’t have to listen to the fan noise. Of course, it would be just fine in a wiring closet attached to a small server.

As mentioned, the SoloPRO is a mid-range offering from ioSafe. They also offer a basic USB 2.0 model (the Sol0), versions of the SoloPRO with SSD or USB 3.0, and an internal drive. This last is particularly clever: It packs the fire- and water-protection technology into a standard 3.5″ SATA drive form factor for use inside a computer.

My biggest complaint about the ioSafe products is that they’re all single-drive only. Although I appreciate their design and construction, all of the protective features ignore the most-common causes of data loss: Logical corruption and the loss of a disk. No storage system can protect data from a “fat-finger” error or operating system fault, but many do include reliability features for the device itself. The ultimate data protection system would include more than one disk drive and would go beyond merely mirroring the data and instead use advanced math (erasure coding, perhaps?) to ensure the consistency of data.

Perhaps ioSafe will consider using two 2.5″ drives and erasure coding in a smarter unit in the future. They could call it the “DuoPRO” – in fact, maybe that’s why they chose the curious “Solo” moniker to begin with!

Stephen’s Stance

No one is suggesting using such a drive as one’s only copy of data, but it makes lots of sense to back up your home or small-office computer to an on-site ioSafe drive. In fact, I would go so far as saying that it makes no sense not to use an ioSafe drive for on-site backups! The drives are somewhat more expensive than basic alternatives but well worth the premium.

A USB-only 1 TB ioSafe Solo costs just over $200 on Amazon, about 50% more than a basic external drive. The eSATA SoloPRO I tested does not appear to have reached Amazon, Buy.com, or NewEgg yet, but ioSafe sells it online for $249. The PRO unit includes the faster eSATA or USB 3.0 port, making it easier to fill up, but not every computer has one of these ports.

Update: The ioSafe SoloPRO is now available online! Amazon lists it for about $250.

For comparison, the only real competition, a SentrySafe Waterproof 160 GB Hard Drive, is $300 for 160 GB! Another option is the SentrySafe Data Storage Chest, which accepts your 2.5″ portable USB drive for a massively-discounted $55. The 250GB ioSafe Pilot internal drive is much more expensive ($250 for 250 GB) but might be just what the doctor ordered if you want to protect data inside a computer.

I am very impressed overall by the ioSafe product. It is solidly-built, and I believe their fire- and water-proof claims. I would not hesitate to recommend this type of drive to small-business owners or “pro-sumer” users concerned about data protection. I do advise keeping an off-site copy of critical data, but the ioSafe is the safest method yet to store an on-site backup.

Disclosure: ioSafe provided the SoloPRO free of charge for testing after I expressed interest in the product

Note: Some of these links include affiliate codes that help pay for this blog. For example, buying an Amazon Kindle with this link sends a few bucks my way! But I don't write this blog to make money, and am happy to link to sites and stores that don't pay anything. I like Amazon and buy tons from them, but you're free to buy whatever and wherever you want.

© sfoskett for Stephen Foskett, Pack Rat, 2010. | ioSafe SoloPRO Review: Is It The Safest Place For Your Data?
This post was categorized as Apple, Everything, Personal, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

It won't be long before other manufacturers adopt the new SATA SSD form factor introduced in the MacBook Air

More information about the unconventional SSD used in Apple’s new MacBook Air. As I discussed in my previous coverage of this new flash form factor, it resembles a PCI Express Mini Card but is much smaller. Toshiba has now proved my speculation that the device uses SATA signals rather than the PCI Express lane used by the similar AirPort card. We also know that the lauded performance of the device is due to its chips and controller rather than skipping SATA in favor of PCIe as some had speculated.

Toshiba Blade X-gale™ HG Series SSD

Toshiba is one of the world’s largest NAND flash manufacturers, and the company uses these chips to produce integrated solid state disks (SSDs) for OEMs. Toshiba’s SSD product offerings are divided into two lines: The mainstream SG series and the high-performance HG Series. The SG line includes standard SATA, slim SATA, and mSATA cards, the latter using the Mini-PCIe form factor. These products only promise 50 MB/s of write performance, while the HG Series can top 180 MB/s writing.

The new "Blade X-gale" SSD form factor used in the MacBook Air is part of the high-performance HG Series from Toshiba

Since the MacBook Air’s SSD is part of the third-generation HG series, its performance is much better than the SSDs used in typical netbook-class computers. NAND flash drives typically suffer during writes, but Toshiba promises sequential write performance for the HG Series of 180 MB/s, matching the read performance of their SG Series. This is almost as fast as the 3 Gbps SATA interface used!

Tiny SSDs like these can only use a few high-density flash chips, so eking out this kind of performance is doubly impressive. The “blade” form factor is about the size of a large USB flash drive and includes just four NAND chips on the top of the board. Toshiba includes both read and write cache in their controller, as well as encryption hardware which is apparently disabled by default. The device supports TRIM, even though Mac OS X does not (yet).

Toshiba offers Blade X-gale SSDs in 64 GB, 128 GB, and 256 GB models. Apple apparently uses all three, offering the smaller pair in the 11 inch MacBook Air and the larger two in the 13 inch model. These products are not available at retail yet, and Japanese reseller PhotoFast has apparently withdrawn their GM2 SFV1 Air Upgrade Kit which used similar modules, so there’s no telling when MacBook Air owners will be able to upgrade.

The 256 GB module is 3.7 mm thick, 1.5 mm more than the 64 and 128 GB siblings. One assumes that this reflects its use of 8 NAND chips rather than 4, and this might lead to better performance as well. Speaking of dimensions, the card is actually 24 mm wide and 108.9 mm long. My previous guesses were quite a bit off.

Stephen’s Stance

This new SSD form factor is certainly intriguing. Although no standard name has been coined at this point (Toshiba’s name, “Blade X-gale,” is a trademark), we will be watching with great interest to see if it catches on. I’ll be calling it a “blade SSD” until I hear a better name.

The compact dimensions of the connector and module itself should be welcomed by tablet and portable device manufacturers, and the fact that it can carry PCI Express as well as SATA signals makes it very appealing. The next-generation NAND chips should easily double capacity within the next year, too!

© sfoskett for Stephen Foskett, Pack Rat, 2010. | Toshiba Offers “Blade” SSDs (Like Apple’s MacBook Air)
This post was categorized as Apple, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

© sfoskett for Stephen Foskett, Pack Rat, 2010. | How Fast Is It? A Storage Infographic
This post was categorized as Apple, Computer History, Enterprise storage, Features, Personal, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

The Four Horsemen of Storage System Performance: These four ugly gentlemen stand between you and your data.

Why do some data storage solutions perform better than others? What tradeoffs are made for economy and how do they affect the system as a whole? These questions can be puzzling, but there are core truths that are difficult to avoid. Mechanical disk drives can only move a certain amount of data. RAM caching can improve performance, but only until it runs out. I/O channels can be overwhelmed with data. And above all, a system must be smart to maximize the potential of these components. These are the four horsemen of storage system performance, and they cannot be denied.

The Chain of Command

It is tempting to think of storage as a game of hard disk drives, and consider only The Rule of Spindles. But RAM cache can compensate for the mechanical limitations of hard disk drives, and Moore’s Law continues to allow for ever-greater RAM-based storage, including cache, DRAM, and flash. But storage does not exist in a vacuum. All that data must go somewhere, and this is the job of the I/O channel.

To be useful, storage capacity must connect to some sort of endpoint. This could be the CPU in a personal computer or an embedded processor in an industrial device. Indeed, there are endpoints and I/O channels throughout modern systems, with potential bottlenecks, caches, and smarts at each point. “Storage people” like me tend to think too small – imagining that the I/O channel ends at the disk drive, the “front end” of the array, or the storage network. But data must travel further, all the way to its final useful point in the core of the CPU.

Once we consider I/O as a long chain of interconnected endpoints, we begin to see the fact that I/O constraints at any point can strangle overall system performance. This is not merely an academic exercise: Optimizing the I/O channel is a consuming passion for most practitioners of enterprise IT, including architects, engineers, and system developers. And, like a good game of Whack-a-Mole, increasing the speed of one link causes another chokepoint to rear its head.

Parallel and Serial I/O

Imagine you had a warehouse full of boxes to move across the country as fast as possible. There are a few options available to you:

1. A fast truck can zip back and forth with just a few boxes
2. A train is slower, but its many cars can haul a huge quantity

But there are realistic limits to both capacity and speed: The train has to fit on the tracks, and the truck can’t move at the speed of light. Plus, one must consider the time taken to load and unload the chosen vehicle.

We continually shift between parallel and serial I/O paradigms

The same trade-offs are true of computer busses: Serial channels can be optimized to zip individual bits back and forth, or parallel busses can be designed to carry whole bytes (or more) at a time. The simplicity of serial communications is tempting, but designers continue to resort to parallelization for added throughput.

Note: Most serial protocols actually feature two links, making them “full duplex”: One for transmit and another for receive.

Serial storage interconnects are dominant currently, with fraternal twins SAS and SATA coming to dominate the disk interface landscape. SAS and SATA share the same 1.5, 3, and now 6 gigabit per second serial physical interconnect, offering more than enough throughput for conventional hard disk drives and edging out older serial (Fibre Channel, SSA) and parallel (ATA and SCSI) alternatives.

Networks (Ethernet, Fibre Channel, and InfiniBand) are predominately serial as well, as are lower-end interconnects like USB and FireWire. Serial communication also dominates in the system bus world, with serial PCI Express toppling parallel PCI.

But parallel variants are often offered for increased throughput: Multi-lane PCI Express and bonded multi-link InfiniBand make up a fair portion of the installed base, while load balancing MPIO drivers are common in Fibre Channel storage. And let’s not forget that the “X4″ variants of Ethernet use multiple bonded links as well.

The Definition of Bottle Neck

Most English speakers have encountered the French term, “cul de sac”, meaning “bottom of the bag” or dead end. But hard disk drives have plenty of “bottom end”, or storage capacity. When it comes to disks, the issue is usually at the neck of the bag: Data just can’t be pulled out of a hard disk drive fast enough.

Emptying a barrel of wine through a spigot takes hours, but pry the end off and the floor is covered in a moment!

The density of modern hard disk drives (the capacity of our barrel) has been growing much more rapidly than the I/O channels serving them (the spigot). Where once a hard disk drive could be filled or emptied in an hour or two, modern drives take days or weeks!

I once called this “flush time“, but I think the wine metaphor is much more appetizing!

This “bottle neck” has serious implications beyond basic storage performance. Data protection is impacted, since ever-larger storage systems can no longer be backed up by dumping their content; system reliability is reduced, since week-long RAID rebuilds increase the risk of multiple drive failures; and cost containment efforts are also impacted, since adding spindles drives up prices.

Nowhere is this bottleneck more evident than in portable devices. Modern drives (like the 1 TB Seagate USB drive I recently reviewed) have massive capacity and pathetic performance. The USB 2.0 interface just can’t keep up, and this creates a limit to the expansion of capacity. It would take half a day to fill that drive under perfect conditions at 25 MB/s, reducing its value as a massive data movement peripheral. The emerging USB 3.0 standard promises to alleviate this performance issue for now, as illustrated with Iomega’s new external SSD.

Cache and solid state storage can help, but they have their own bottlenecks. Storage arrays typically use Fibre Channel or SAS SSDs, and their front-end interface remains the same. The best-performing SSDs use the PCI Express bus directly rather than emulating hard disk drives over SCSI interfaces. And even PCI Express might not be enough to handle the massive I/O of NAND flash or DRAM. In each case, the bottleneck moves down the chain.

A Chain of Bottlenecks

Let’s follow a typical I/O operation from the disk to the CPU core and count the I/O channels:

1. A read head senses the state of a bit of magnetic material on the surface of a disk
2. The head transmits this signal to a buffer on the disk controller board
3. The data is picked up by the disk controller CPU and transmitted over a SATA or SAS connection
4. The storage array or RAID controller receives the data and moves it over an internal bus to another buffer or cache
5. The data is picked up by another CPU in the array controller and sent out another interface using Fibre Channel or Ethernet
6. The data is buffered and retransmitted by one or more switches in the storage network
7. The host bus adapter (HBA) on the server side receives the data and buffers it again before sending it over a local PCI Express bus to system memory
8. The server memory controller pulls the data out of system memory and sends it via a local bus to the CPU core

There are actually many more steps than this, but the picture should be clear by now. There are many, many I/O channels to consider when it comes to storage, and the drive interface is just one potential bottleneck.

We constantly move bottlenecks around - as one link is improved, another choke-point appears

Optimizing Storage I/O

Tactical steps to improve storage performance typically focus at one link in the chain: Drive vendors move from 1.5 Gb to 3 Gb SATA, or SAN buyers upgrade from 4 Gb to 8 Gb Fibre Channel. But the basic architecture of enterprise storage has remained constant for over a decade, and the reliance on block SCSI commands endures. This is all about to change.

One critical bit of I/O optimization exists at the point of connection between the various chipsets inside the server. AMD pulled the memory controller off of the “northbridge” with their Athlon line. Intel did the same with their Nehalem and is eliminating the northbridge entirely with the Lynnfield/Jasper Forest CPU lines. This gives serious bandwidth to the crucial PCI Express-to-CPU-core link, moving the bottleneck downstream.

We are in the midst of a massive upgrade of the storage network as well. Between 8 Gb Fibre Channel and iSCSI and Fibre Channel over 10 Gb Ethernet, not to mention persistent interest in InfiniBand, storage network throughput is rapidly expanding. As with the internal PC connections, the expansion of network bandwidth has pushed the bottleneck to the storage array interface for the time being.

Microsoft and Intel recently pushed over a gigabyte per second over 10 GbE using iSCSI, but they needed multiple storage targets to feed that connection. It isn’t that modern storage systems couldn’t push that kind of I/O (indeed, arrays are tens to hundreds of times faster internally thanks to their spindles and cache), but that the conventional storage protocols are tightly linked to a single “front-end” interface. The current state of the art for storage array design is moving to distributed models, exemplified by pNFS and scale-out NAS concepts like MaxiScale (now acquired by Overland).

Once the array interfaces can pump out massive I/O, attention will turn once again to the disk interfaces themselves. Although 6 Gb/s SAS and SATA is now a reality, this interface is inappropriate for future high-performance SSDs. Arrays designed around flash or DRAM are likely to switch to PCI Express as their internal connection of choice for performance and to optimize data placement on these new devices. Companies like Nimbus and NetApp are already moving in this direction.

Time To Get Smart

Hard disk drive spindles make up the bulk of storage capacity, but small amounts of cache make them far more effective. But both of these horsemen must operate within the constraints of the I/O channels they pass through. This brings us to the final horseman of performance: Smarts. Clever designers have created clever controlling mechanisms to overcome the limits of spindles, cache, and I/O channels.

© sfoskett for Stephen Foskett, Pack Rat, 2010. | The Four Horsemen of Storage System Performance: I/O As a Chain of Bottlenecks
This post was categorized as Computer History, Enterprise storage, Personal, Terabyte home, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

The new MacBook Air includes tiny SSD and AirPort cards

Apple updated the ultra-slim don’t-call-it-a-netbook MacBook Air this week. Along with a wimpy out-of-date CPU, the new Air features all-SSD storage of an entirely new and apparently proprietary type. Let’s take a look and see what we can see.

Just The Facts

You might want to read my piece about Unconventional SSDs: PCI Express Mini Card (Mini PCI-E) or the updated information in Toshiba Offers “Blade” SSDs (Like Apple’s MacBook Air)

The MacBook Air isn’t supposed to be upgradeable, and Apple went so far as to use a special 5-point Torx-ish safety screw to keep us out. As revealed by iFixit, just about everything inside is soldered in except for the Wi-Fi and SSD cards. Apple, shockingly, included an inside shot of the 13″ Air on their official Design page, showing that both machines use the same SSD connector.

Is this new form factor called a “blade SSD”?

• The Air uses a new type of connector that looks a lot like a Mini PCI-Express slot but definitely isn’t.
  • The connector is exactly the same width as the USB shield: That’s about 7 mm, not the 24 mm of a Mini PCIe slot.
  • This connector is used for both the AirPort (Wi-Fi) board and the SSD, so it’s not some kind of shrunken SATA port.
  • The SSD has contacts on only the “bottom” side with what looks like a ground plate on the “top”, while the AirPort seems to have contacts on both sides (or at least on “top”).
  • The connector is split with six pins on one card edge and 12 on the other. A one-sided version thus has 18 pins (plus ground) while a two-sided variant has 36 pins. For comparison, Mini PCIe is a two-sided card edge with 8 pins and then 18 pins, for a total of 52 pins.
• System Profiler reports that the SSD and AirPort connect to the NVidia MCP89 chipset in different ways:
  • The SSD is a SATA device on the AHCI lines. While it’s possible that Apple could have designed an internal SATA controller and be presenting it on a PCIe lane, it’s much more likely that it’s really using SATA over the new connector described above. Since SATA needs just 5 or 6 pins plus ground (4x data and 3.3 volts and 5 volts of power), there are plenty of connectors for it.
  • The AirPort (802.11n Wi-Fi) card, on the other hand, appears to be using a PCI Express lane. Since PCI Express also needs just 6 to 8 pins, this fit, too.
• The cards are much smaller than conventional Mini-PCIe cards as well:
  • The SSD is 24×108.9 mm, much smaller than the 30×51 mm typical of Mini-PCIe SSDs. This compact size makes the Toshiba NAND chips appear to be giants, but they’re really the same chips used on other SSDs.
  • The AirPort card is really compact at about 22×30 mm.

It appears that Apple (or their supplier) has developed this new form factor to be even thinner and narrower than the already-small PCI Express Mini Card format. One wonders if other vendors will adopt this new smaller “Air Card” format as well.

Apple's "Air Card" connector is single- or double-sided with up to 36 pins

Stephen’s Stance

If the MacBook Air had used a PCI Express-based SSD like the Fusion-IO or OCZ models, it really would have been a revolutionary move. But booting a computer from a device like this would have been challenging, requiring revisions to the EFI firmware, drivers, and Mac OS X itself. Therefore, it is not surprising to see a SATA connection used instead. So no, Robin, this isn’t the end of SATA (yet).

Yup, it’s SATA. More information is available in my article, Toshiba Offers “Blade” SSDs (Like Apple’s MacBook Air)

Some will undoubtedly complain about Apple’s proprietary format, but this is neither a new development nor a particularly damning one. Most modern Macs, including the MacBook, MacBook Pro, and Mac Mini, already use proprietary AirPort daughter cards and no one howled in protest when they switched from standard PCI Express Mini Cards. Having the storage on the same type of connector is good from a bill of materials standpoint, saving Apple a few dollars. And this is not an upgradeable machine. Moaning about the low-spec Core 2 Duo CPU and soldered-on 2 GB of RAM is much more sensible than whining about a proprietary SSD.

Images from Apple, Inc. and iFixit.com

© sfoskett for Stephen Foskett, Pack Rat, 2010. | Apple’s Unconventional New MacBook Air SSD
This post was categorized as Apple, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Unconventional SSDs

Most conventional SSDs use conventional disk interfaces, normally serial ATA (SATA). These plug right into just about any modern computer using standard internal SATA power and data connectors. But small form factor PCs like the MacBook Air need a more-compact SSD, leading to some clever (and confusing) engineering.

Most modern computers use the PCI Express and USB busses for internal peripheral connections other than storage. Most folks are familiar with the full-size USB ports and PCI Express slots found in desktop machines, but portable computers (and a few others, including my iMac) need something more compact.

Introducing PCI Express Mini Card

This led to the development of the PCI Express Mini Card standard, commonly called Mini PCI-E. Like the similar ExpressCard format, Mini PCI Express includes both USB and a single PCI lane, along with a host of other pinouts for flexible use in portable machines:

• A single PCI Express lane
• USB 2.0
• System Management Bus (SMBus) – a 2-wire lightweight communication bus derived from I2C for power-on/off commands
• Pins reserved for GPS use
• Pins for diagnostic LEDs showing Wi-Fi activity
• Pins to communicate with a GSM SIM card slot located elsewhere
• Reserved pins for an additional PCI Express lane
• 1.5 and 3.3 volt power

This is one full-featured slot, as you can see, packing many useful features into a compact connector.

Nearly every modern laptop has at least one Mini PCI-E slot, and they are frequently used for Wi-Fi and 3G radio devices. Apple often uses them for their “AirPort” Wi-Fi adapters, so machines as diverse as the desktop iMac include Mini PCI-E slots.

Mini PCI-E SSDs

mSATA SSDs like this Toshiba model reuse reserved Mini-PCIe pins for SATA connections

As SSD performance increases, conventional disk interfaces like SATA and SAS are becoming increasingly-saturated. And block storage protocols like ATA and SCSI make little sense when communicating with NAND flash devices like SSDs. So PCI Express SSDs have become increasingly popular lately. Companies like Fusion IO and SandForce are aggressively pushing these devices, and the performance numbers are impressive.

Unsurprisingly, the availability of a compact PCI Express Mini Card slot in portable machines has led to the development of tiny PCI Express SSDs for these applications. A wide variety of options are available but, as with ExpressCards, caution is in order since not all are equal. Manufacturers have developed a variety of Mini PCI-E SSDs using the various pins in this tiny slot:

• Native PCI Express SSDs can use the single included PCI Express lane and are compatible with most computers
• Cheaper USB SSDs (essentially glorified thumb drives) can use the included USB pins and are also compatible with most computers, though much slower
• A de-facto standard for SATA over Mini PCI-E called mSATA is seeing wider adoption but is incompatible with standard Mini PCI-E slots since it uses (reserved) pins 5 and 7 for SATA communication
• Various alternative methods of sending SATA signals have been developed, including the format used in the popular Asus Eee PC, which are also widely incompatible

This proliferation of Mini PCI-E SSDs has caused confusion and frustration among users. Many have purchased Mini PCI-E to SATA adapters designed for the Eee PC, for example, hoping to connect SATA devices to their machines. Finding compatible devices is a trial-and-error process and is played out repeatedly in forums dedicated to machines like the JooJoo and Mac Mini.

Stephen’s Stance

Apple chose another format entirely for the new MacBook Air, designing their own connector rather than selecting a standard like PCI Express or even mSATA. The variety of incompatible Mini PCI-E cards is frustrating to enthusiasts like myself, and I had hoped that the entry of a major company like Apple would bring sanity to this disorder.

As with full-sized PCI Express SSDs, Mini PCI-E SSDs hold great promise for performance, reliability, and power consumption and their compact size is perfect for a variety of applications. I look forward to examining and reviewing these devices in the future!

© sfoskett for Stephen Foskett, Pack Rat, 2010. | Unconventional SSDs: PCI Express Mini Card (Mini PCI-E)
This post was categorized as Apple, Computer History, Personal, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.