It looks like Apple will indeed re-brand Intel Light Peak as "Thunderbolt" and combine it with Mini DisplayPort!

Intel has been incredibly tight-lipped about Light Peak. Although I’ve been hounding my contacts inside the company for months, no one has spilled the beans about anything. All I know about Light Peak I learned on the Internet, as they say. Now comes another bombshell: Apple will introduce Light Peak-equipped MacBook Pros tomorrow (February 24) with “Thunderbolt”, a high-speed I/O port!

One could easily guess that Apple would rename Light Peak for its own use. It did the same with its previous high-speed I/O port, IEEE 1394, known among Apple users as FireWire. A trademarked name allows Apple to control compatibility to some extent, requiring users of the name to submit to Apple’s guidelines and perhaps pay a fee for its use. And Light Peak seems an especially poor name amid rumors that it will not use an optical connection after all!

The news comes from the German-language site, FSCKLog, and includes photos of the 13″ MacBook Pro spec sheet, the Thunderbolt logo, and even the ports on the side of the machine!

If we take this as fact (and the logo it pretty convincing) here’s what we know about Apple’s implementation of Light Peak:

1. Apple will call it “Thunderbolt” and refers to it as a “High-speed I/O” port
2. Apple will integrate Thunderbolt with the Mini DisplayPort connector (rather than a USB 3.0 port as I had guessed)
   1. The German spec sheet says “Thunderbolt-Anschluss unterstuetzt High-Speed E/A und Mini DisplayPort Geraete”, which means “Thunderbolt port supports High-Speed I/O devices and Mini DisplayPort”.
   2. The side photo shows an ordinary-looking Mini DisplayPort connector with a little thunderbolt icon next to it. Under magnification, I don’t see any sign of image manipulation in this photo.
3. Since it’s integrated with the Mini DisplayPort connector, it’s likely Apple will use the monitor as a high-speed I/O hub or breakout box
   1. Expect to see a new line of Cinema Displays with Thunderbolt-powered ports embedded in them.
   2. I bet companies like Belkin will quickly come out with Thunderbolt breakout boxes.
4. The MacBook Pro will still include FireWire 800 and (2x) USB 2.0 I/O ports, in addition to a MagSafe power port, Gigabit Ethernet port, and SDXC card slot.
5. There is no mention of USB 3.0, though I strongly suspect it will be included in the Thunderbolt spec.

Interested in learning more? You might want to check out some of my other articles about Light Peak, or my Light Peak performance comparison infographic!

Stephen’s Stance

I was skeptical that Apple would introduce Light Peak this month, though confident it would come this year. But this evidence is very convincing, if not wholly satisfying. I’ve been holding off on upgrading my three-year-old Santa Rosa MacBook Pro until Apple released some kind of serious I/O: A few USB 2.0 and FireWire ports just doesn’t cut it for my use. I do hope these new Sandy Bridge MacBook Pros meet my needs, though, because I’m itching for an upgrade!

But the specific inclusion and mention of USB 2.0 and FireWire 800 ports gives me pause. Why put USB 2.0 on board instead of USB 3.0? Why bundle “Thunderbolt” with the Mini DisplayPort connector rather than a USB 3.0 port? Why is there no mention of what Thunderbolt is useful for? I’m concerned that Thunderbolt might not be fully baked, and might not deliver the “high-speed I/O” I wanted. Early adopters could be stuck with limited compatibility and connectivity, and there is no telling if my “breakout box” concept will come to fruition. Heck, Thunderbolt could carry just video and audio for all we know!

Image credit: FSCKLog

© sfoskett for Stephen Foskett, Pack Rat, 2011. | Will Apple Call Light Peak “Thunderbolt”?
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.

Desktop NICs Won’t Work

Most of the network interfaces you will find at retail simply won’t work. Realtek is the dominant provider of Gigabit Ethernet controllers for motherboards and add-in cards right now, but none of their chips are natively supported by VMware ESX. The number-two slot seems to be Marvell’s Yukon chips, with Intel’s desktop controllers close behind. None of these will work, either.

Although it is possible to get a non-supported NIC to work in VMware ESX, it’s not a good idea. First, ESX won’t install unless it finds a supported NIC in the box. Then there’s quite a bit of fiddling to get the driver up and running. And you’re left with a potentially-weird configuration that might not support advanced features. It’s a much-better idea to locate and purchase a supported NIC.

Here’s what not to buy

Many inexpensive Ethernet cards and motherboards have a chip with the Realtek “digital crab” logo. None of these will work for VMware ESX.	The big “psychedelic M” identifies a Marvell controller. Skip these, too.

Selecting a Functional Home/Lab NIC

My “home/lab” network card criteria are simple:

1. They are specifically listed on the VMware ESX HCL for version 4.1 with no hacks or trickery involved
2. They cost less than $100 US
3. You can easily purchase them at retail from major online vendors (Newegg and Amazon)
4. They use PCI or PCI Express bus and have 1 or more RJ45 Gigabit Ethernet ports

It’s really amazing how few cards meet these criteria: There are really just a few cards to consider in this range.

Here’s what to buy

PCI Adapters
	The Intel Pro/1000 MT server adapter should work, and the dual-port is cheaper on Amazon (Newegg)
PCI Express (PCIe) adapters
The Intel Pro/1000 CT desktop adapter is a cheap and functional PCIe NIC (Newegg)	The Intel Pro/1000 PT server adapter is a little more expensive but potentially better-supported (Newegg)	The HP NC112T also appears to be well-supported and affordable

Although some have reported success with the very-cheap Intel Pro/1000 GT desktop adapter, I can’t recommend it. I’ve heard many negative reviews of folks trying and failing to get this adapter to work in the latest versions of VMware ESX. I think it’s worth the money to step up to the CT or PT instead!

Note also that I have not personally tried the specific adapters listed and linked here. I intend to purchase one or more over the next few months and will update this post when I do, but I welcome feedback on your experiences with them!

Stephen’s Stance

VMware ESX seems especially picky about network adapters, and the fact that it will not install without a supported NIC onboard is a real stumbling block for users. I definitely recommend picking up a well-supported NIC like the Intel Pro/1000 MT (PCI) or CT/PT (PCIe) or the HP NC112T.

My home/lab machine has two PCIe slots and two PCI slots. I had intended to use a PCI NIC, but will probably buy a Pro/1000 PT card instead. It’s affordable and called out specifically as supported in the VMware ESX HCL. Sounds good to me!

If you have a suggestion for a NIC that fits the criteria above, please do let me know. I’d love to have more choices in this list!

© sfoskett for Stephen Foskett, Pack Rat, 2011. | The Best Network Card For VMware ESX Home Lab Machines
This post was categorized as Everything, 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.

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 lure of Fibre Channel over Token Ring lies in what it tells us about Ethernet and ourselves

The buzz about Fibre Channel over Token Ring has built rapidly over the last week. Industry experts like Greg Ferro, Denton Gentry, and Joe Onisick have weighed in (as have I), and the Packet Pushers Podcast featured the news in show 12, “Get on the Ring!” Some have called out FCoTR as a foolish hoax, but the FCoTR phenomenon is not foolish. Indeed, FCoTR gives everyone in the industry the chance to reevaluate the current state of the art and has exposed real weaknesses in the Ethernet-centric future of the data center.

The Best Tech Rarely Wins

It has become something of a maxim in technology circles that “the best technology rarely wins.” While many point to the victory of VHS over Betamax, techies often look to the success of CISC over RISC, Windows over UNIX or OS/2, and PC over Macintosh. In every case, it was the plentiful availability of cheap “good enough” products that trumped any apparent technical superiority. And in every case research and development led the purported “inferior” technology to eventually surpass the capabilities of the favorite.

Electric cars, like this 1915 Detroit Electric, were popular for their cleanliness, simplicity, and quiet yet gasoline won out

The market failure of Token Ring was an ideal case study for what might be called the Betamax lesson. The token-passing scheme served to enhance both reliability and predictability. Unlike Ethernet, which focused on fault tolerance with “conversational” CSMA/CD, Token Ring took an ordered “Roberts Rules” approach with each station waiting for permission before transmitting. This mechanism could also allow a station to reserve bandwidth for a critical application. Token Ring networks were inherently faster than Ethernet as well, at 4 Mb/s, 16 Mb/s, and eventually 100 Mb/s. But the expense and complexity of cabling caused it to lose favor. This was especially true once 100BASE-TX Ethernet became common: It was fast enough for most LANs and wide support and availability made it incredibly cheap. Gigabit Token Ring was standardized in 2001 but never implemented.

The Elephant in the Room

Ethernet has developed rapidly since it became the de facto data center standard. Gigabit Ethernet ports and switches are common today, and engineering has made it fairly reliable and interoperable in practice. The spread of IP networks also helped Ethernet: Many applications rely on TCP for reliable communication, masking collisions and data link errors even as CSMA/CA and improved DSPs reduced their frequency. The industry is currently shifting again, with 10 GbE becoming more common and 40 Gb and 100 Gb Ethernet standardized and rolling out in high-end switches. Convergence is the word of the day in data center circles, and Ethernet is the heir apparent to rule the converged world. To handle new traffic types (Fibre Channel and RDMA, for example), Ethernet is being extended with prioritization and reservation of bandwidth, advanced congestion management to avoid packet loss, and a mechanism to specify what capabilities are present.

From Star Trek III:

James T. Kirk: Scotty, as good as your word

Montgomery Scott: Aye, sir. The more they overthink the plumbing, the easier it is to stop up the drain. Here, Doctor, souvenirs from one surgeon to another. I took them out of her main transwarp computer drive.

Although interest (and much enthusiasm) in these Data center bridging extensions is widespread, an undercurrent of trepidation is present as well. Storagers worry whether Ethernet is really ready to handle their precious cargo, and networkers are concerned over the proprietary and complex nature of these add-ons. Independent voices also fear the gusto with which vendors are endorsing them, with Fibre Channel over Ethernet (FCoE) becoming a particular object of skepticism. Many see these moves as a land grab rather than a use case-driven expansion.

Stephen’s Stance

We may laugh at the idea of Fibre Channel over Token Ring (FCoTR) in today’s data center, but the concept also exposes real fears about a future data center dominated by converged I/O reliant on Ethernet. The putative features of FCoTR are exactly the weaknesses traditionally seen in Ethernet: Packet loss, congestion, and mediocre hardware make it unsuitable for sensitive payloads like storage. Would a resurgence of Token Ring, engineered with these in mind, really be so bad?

Although data center extensions and “big iron” equipment promise to eliminate these weaknesses, many remain concerned about an Ethernet-dominated data center. At what cost will these enhancements be delivered? Do we really need them? There is also a real technical concern that Ethernet might not withstand another round of bracing and might instead fall over on its crutches. At the very least, data center-class Ethernet is late and overweight. It is wise always to carefully consider which step to take next when so much is on the line.

© sfoskett for Stephen Foskett, Pack Rat, 2010. | The FCoTR Phenomenon Exposes the Weaknesses in Ethernet
This post was categorized as Computer History, Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Networking Basics

Let’s get the first two components of the network name out of the way:

The first part is the signaling rate in megabits per second. In layman’s terms, this is the speed of the network at hand. You are likely to come across one of the following:

• 10 megabits
• 100 megabits – Fast Ethernet
• 1000 megabits – Gigabit Ethernet, GbE, or 1000BASE-X
• 10,000 megabits – 10 Gigabit Ethernet, 10GbE, or 10GBASE-X
• 40,000 megabits – 40 Gigabit Ethernet, 40GbE, or 40GBASE-X
• 100,000 megabits – 100 Gigabit Ethernet, 100GbE, or 100GBASE-X

It may strike you as odd that the next part is always the word, “BASE”. But there is a reason for this, too. BASE refers “baseband”, meaning that this is an unfiltered line not requiring a digital modulation scheme. Back in the day, there was a 10PASS-TS version of Ethernet that used a signaling scheme similar to a modem, but baseband is dominant today.

So 100BASE refers to a Fast Ethernet connection that uses the unfiltered cable for transmission.

BASE-What?

The third part of an Ethernet network type refers to the cabling used to carry the signals. The earliest forms of Ethernet used coaxial cable, but thin twisted-pair cabling became popular in the mid-1990s. Faster versions of Ethernet also often use fiber optics rather than electrical signals.

There are a bewildering assortment of physical interconnects for Ethernet. But the naming system isn’t as complex as it might appear:

• The first letter tells us which kind of wire we are talking about:
  • “T” means twisted-pair cable (e.g. the common Cat5 in use today)
  • “K” means a copper backplane
  • “C” means balanced copper cable
  • “F” means optical cable
  • “B” uses two wavelengths over a single optical cable
  • “S” means short-range multi-mode optical cable (less than 100 m)
  • “L” means long-range single- or multi-mode optical cable (100 m to 10 km)
  • “E” means extended-range optical cable (10 km to 40 km)
  • “Z” means long-range single-mode cable at a higher wavelength
• Next is the coding scheme for data on the wire
  • “X” means 4B/5B block coding for Fast Ethernet or 8B/10B block coding for Gigabit Ethernet
  • “R” means 64B/66B block coding
• Finally, we have a number representing the number of parallel “lanes” for data
  • “1″ would mean serial (non-parallel) but is omitted instead
  • “4″ or “10″ are available for copper wire
  • Just about any other number could be used for optical lanes or wavelengths

Examples

Now let’s look at some examples:

• Back in the day, 10BASE-T became more common than coaxial 10BASE2. It was a simple 10 megabit baseband signal over common twisted-pair.
• When Fast Ethernet first rolled out, there was some concern that traditional (usually Cat3) cabling couldn’t handle 100 megabits. Some early implementations used four copper pairs (100BASE-T4) or fiber optics (100BASE-FX), but nearly every 100 megabit Ethernet connection today is 100BASE-TX, using plain two pairs on plain Cat5 cable.
• Gigabit Ethernet had a similar history. Many were concerned that two pairs on unshielded Cat5 wiring could not handle 1000 megabits per second, so optical (1000BASE-SX) and balanced shielded wiring (1000BASE-CX) were specified. Although an unshielded 2-pair standard was developed (1000BASE-TX), it never really caught on. Therefore, today’s predominant gigabit LAN connection, 1000BASE-T, uses all four pairs of unshielded twister-pair wiring on a Cat5 cable (see note 1).
• The 10 Gigabit Ethernet world has mostly shifted to the block coding scheme from Fibre Channel, 64B/66B, which is denoted by the letter “R”. This gives us a family of fiber optic cables (10GBASE-SR, LR, ER, etc), and a copper backplane interconnect (10GBASE-KR). The earlier copper wiring standard (10GBASE-CX4) used InfiniBand-like 4-lane cables and 8B/10B signaling, as did 10GBASE-KX4 on the backplane. A backwards-compatible twisted-pair 10GBASE-T has also been developed, but work continues to make it power-efficient enough to be practical (see note 2).
• Looking ahead, we see Higher-Speed Ethernet emerging: 40GBASE-KR4 for backplane use, multi-mode optical 40GBASE-SR4 and 100GBASE-SR10, and long-range single-mode optical 40GBASE-LR4 and 100GBASE-LR10.

As you can see, all this alphabet soup does have some consistency. Common unshielded twisted pair wiring is all “BASE-T”, optics are denoted according to their range (“S”, “L”, “E”), and backplanes use “K” copper. Clear as mud?

Note 1: Lots of people (and even equipment makers) incorrectly refer to common Gigabit Ethernet as “1000BASE-TX”, but this really should be called “1000BASE-T”.

Note 2: We will probably never see a 10GBASE-TX, which would use just 2 pairs of unshielded twisted pair copper wiring.

© sfoskett for Stephen Foskett, Pack Rat, 2010. | 1000Base-What?
This post was categorized as Computer History, Enterprise storage, Gestalt IT, Personal. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Update: Check out the latest multi-vendor iSCSI post!

I usually don’t write about other peoples’ articles on this blog, preferring to stick to my own independent work. But this time I’m making an exception.

If you use or are interested in VMware ESX 3.x and iSCSI, you simply must go read Chad Sakac’s post on the topic. Co-written with just about everyone in the industry (including EMC, VMware, NetApp, Dell/EqualLogic, and HP/Lefthand), Chad has put together a “cheat sheet” on the ins and outs of iSCSI connectivity and performance in ESX 3.x.

Top takeaways (and I’ve been preaching about these for a while myself, too!)

• Ethernet link aggregation doesn’t buy you anything in iSCSI environments
• iSCSI HBA’s don’t buy you much other than boot-from-SAN in ESX, either
• The most common configuration (ESX software iSCSI) is limited to about 160 MB/s per iSCSI target over one-gigabit Ethernet, but that’s probably fine for most applications
• Adding multiple iSCSI targets adds performance across the board, but configurations vary by array
• Maximum per-target performance comes from guest-side software iSCSI, which can make use of multiple Ethernet links to push each array as fast as it can go

Thanks for putting together such a great article, guys!

This post can also be found on Gestalt IT: Essential Reading for VMware ESX iSCSI Users!

© sfoskett for Stephen Foskett, Pack Rat, 2009. | Essential Reading for VMware ESX iSCSI Users!
This post was categorized as Enterprise storage, Gestalt IT, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.