|
||
|
|
|
              |
|
 |
 |
||||||||||||||||
|
 |
|
||||||||||||||
By William Van Winkle |
||||||||||||||||
DRIVES II: SPEED AND SAFETY Besides externals, there’s hot news afoot in the world of internal hard drives, and for once it’s not just about the interface, rotation speed, and/or capacity. When was the last time you saw internal hard drives take any sort of leap in functionality? Well, get ready, because now we have two leaps, and they’re both going to be important in the coming months. Now that we’ve done our bashing on flash memory aiding PCs, let’s backtrack. You’ve no doubt heard by now about hybrid hard drives, meaning conventional hard drives with 256MB or so of flash memory added on them as a sort of secondary cache. Windows Vista utilizes this extra memory under the name DriveBoost. News on DriveBoost has been about as whisper quiet as the drives it works with, and this may be because Vista just isn’t living up to expectations on performance increases with the extra flash resources. Some people anticipate that DriveBoost’s acceleration situation will improve with the first Vista Service Pack release.
But don’t write off hybrid drives just yet. Even if the speed during regular system operation turns out to be a wash, the caching function of the flash memory will work to reduce the number of times the system needs to read from the drive platters. This means less spin time, longer drive life, and—-most importantly for mobile users—-lower power draw. In theory, Vista should also be able to store certain boot files to the flash memory in order to help improve system boot times. Be aware, though, that it takes many reboots in order for this function to have any effect. Samsung and Seagate have been the two key vendors for hybrid hard drives to date. On the security side, we turn again to Seagate but add Hitachi to the mix. Seagate has been trumpeting its Full Disc Encryption (FDE) technology, more recently branded under the name DriveTrust, for what seems a very long time. The idea behind it seemed brilliant. Essentially, FDE entails mounting an AES encryption chip into a hard drive’s electronics, enabling a client-side application to manage the encryption, and making sure there’s a motherboard BIOS in the middle that understands how to interact with both of those elements. The advantage of a discrete chip is that there’s no additional software cost and no need for the CPU to expend cycles on the encrypt/decrypt load. What you should end up with, when enabled, is a system in which everything written to the hard drive from the first OS installation files out to the smallest everyday Word docs is locked behind bulletproof AES encryption. If someone steals the file, it looks like nothing but gibberish, and the only way to unlock the drive is with a hashed, user-selected password. With FDE, notebook theft no longer means a compromise of sensitive information.
That was the plan. But there were hiccups in the software. Nobody had BIOS support. Etcetera. Finally, though, Seagate delivered the 2.5” Momentus 5400 FDE.2, the world’s first FDE drive available in mass quantities. It also supported key integration with a Trusted Platform Module-—a significant point when you start talking about vPro systems. The Momentus 5400 FDE.2 now ships in capacities up to 160GB. As companies become ever more enmeshed in compliance regulations, products such as the FDE.2 will grow in importance. Their potential to reduce liability alone will more than justify their use. More recently, Hitachi Global Storage followed suit with its Bulk Data Encryption technology, which is more or less identical in concept to Seagate’s FDE. The technology is applied as an option to the 2.5” Travelstar 7K200, a 7,200 RPM, 200GB jewel that claims spec sheet superiority over its Seagate rival. For an excellent look at what to expect in Hitachi’s implementation, Google the phrase: Hitachi How To Guide Bulk Data Encryption Technology. You’ll find a three-page PDF to get you on your way. Full disk encryption is just starting to make its way from the mobile space into desktops. Seagate’s DB35 drive, optimized for use in digital video recorders, was the first 3.5” model to get the feature, which only makes sense when you consider content providers’ worries about protecting user recordings. For the rest of the world, Seagate announced in September its new Barracuda FDE series, sporting the same DriveTrust technology as the 2.5” drives, in capacities up to 1TB.
We should point out that we have not reviewed any of these encryption-enabled drives here at RAM. The question is not whether the product works—-we’re sure it does—-but rather how easy it is for a system builder to deploy. If compatibility is limited to only a handful of notebooks or desktop motherboards, or if installation is too cumbersome to be practical across large numbers of systems, then we still don’t have a viable channel solution. Further examination will tell for sure. For now, just be aware that these solutions are out and may be exactly what your customers need to meet their growing security needs. And if the drives aren’t exactly what you need just yet, odds are that they will be soon. CASES: THINK BIG...AND SMALL Once upon a time, desktops and mini tower cases were the norm. ATX form factor, five bays, and anything but beige-—have a nice day. But usage models have changed, even if the basic home/SOHO/small biz format of choice hasn’t yet. You might have caught a clue on this point last year with the demise of BTX, which was slated to supplant ATX in the industry. The argument was that NetBurst-based processors (Pentium D and prior models) were running too hot for ATX designs. The airflow patterns in ATX systems were all wrong. BTX would fix that, enabling bigger systems to run cooler while also facilitating smaller systems than ATX could accommodate.
Of course, BTX turned out to be a bust, both because manufacturers failed to see a significant enough advantage over ATX and because the lower thermals of the Core microarchitecture wiped away (for a while) the paranoia about systems getting too hot. The interesting thing is that, while BTX was still on the rise, most enthusiasm and adoption in the channel focused not on the mainstream desktop and mini tower systems but in tower and small form factor designs. The change in form factor was gravitating to the areas in which change and value-add was most prevalent. This is why the picoBTX design was the last to evaporate. It was the niche in which change was most prevalent. Is it any wonder that Shuttle has been able to stay successful in small form factor systems by not following the ATX standard? In a market where change is rapid, you can’t always succeed by adhering to the norm. Part of the reason there’s so much change at the top and bottom system sizes is because buyers are reconfiguring their components and peripherals. A power user no longer has one meaty GPU; he has two-—maybe even four if you count NVIDIA’s GeForce 7950 GX2 in an SLI configuration. And don’t forget that we’re expecting to see three graphics cards in enthusiast machines before long. Then there are CPUs to consider. AMD devotees may well opt for the company’s Quad FX platform, which plants two high-end CPUs on a motherboard, just like a 2P server. With the rise of 64-bit software and operating systems, we’re seeing more memory modules going into systems. All of this means more heat and a need for more space. Enthusiasts love to put bling in their boxes-—everything from cold cathode tubes to specially etched panel windows to bay-mounted rheostat controls—-and that means a need for big cases. Not least of all, as people store more and more personal media, particularly of the HD variety, the need for bays in which to store an ever-increasing number of hard drives becomes paramount. Old-school desktops could accommodate two hard drives, maybe three. But this is the age in which RAID 5 capability comes included in chipsets for free, and data redundancy is king. For users who want to protect their files, three drives is now the starting point, and you, the reseller, need to help provide them with plenty of room to grow. How many drive bays does that mean? Who knows? How many can you get? All of this presumes that this case is a “power box” being used either for extremely demanding tasks (such as gaming or workstation-class encoding) or as a central storage repository. Either way, people who buy these large cases are not averse to thinking in value-add terms. They know that their applications require a higher caliber of component. Cooler Master and Thermaltake are two of the best vendors in this field with strong channel presences. Lian Li also makes exceptional towers, but the company has yet to make much of an effort toward system builders.
Antec has done the best job to date of combining high-quality manufacturing, smart feature adoption, and solid channel partnership in its tower chassis. For enthusiasts, Antec’s Nine Hundred is one of the slickest cases we’ve seen yet. Replete with nine drive bays, two front-loading drive cages, four LED case fans (and an option for two more), and clear side paneling, the Nine Hundred is tops for demanding power users. For those less concerned with eye candy, we like Antec’s Performance One P190 + 1200 tower, which blends attributes such as RoHS compliance and noise-dampening features with nine drive bays and pull-out drive cages. You’ll also find perks such as special cable management features, a 1200W dual power supply system (a 650W PSU for the motherboard and cards with a 550W PSU for drives and peripherals), and even an internal gooseneck LED lamp for when you’re working in those dark chassis corners. Now, assuming you put all of those drives into a beefy tower case, what happens to the requirements placed on other network systems? We’re down to discussing something like a server/client model. The PCs in the den and kitchen don’t need much storage capacity if most of their data is being stored on a central repository elsewhere. Obviously, this applies in the SOHO and small business world, too. Many of these clients only need a mainstream CPU sitting on a motherboard with integrated graphics and a network controller in the shadow of a slim optical drive. As long as your customer doesn’t need a lot of peripherals soaking up cubic liters, then you can entice him with the benefits of a tiny footprint, more eco-friendly power consumption, chic style, and nearly silent operation. Taking an extreme example, check out the X series of barebones systems from Shuttle. These amazing little machines measure only 8.27” x 11.76” x 2.1” and are based on Intel’s Centrino mobile computing platform. Essentially, they’re PCs the size of a hardback novel running notebook innards. The X200, for example, runs on Intel’s 945GM/ICH7M chipset, complete with GMA 950 integrated graphics, and supports mobile CPUs up to the Core 2 Duo as well as dual-channel DDR2-677 SO-DIMMs. With an external power brick, the PC is essentially silent, plus it includes 802.11b/g wireless connectivity. The barebones includes a slimline combo optical drive and can fit a full 3.5” SATA drive, giving it desktop-class drive capacity. There are options for a TV tuner, media center remote, and even a fitted carrying case. For those who like the messages surrounding small form factor but want more conventional performance and expansion capabilities, we’re firm believers in the “cube” form factor of traditional Shuttle XPCs, now in the form of models such as the SP35P2 Pro. This unit is based on Intel’s P35 chipset, which makes it compatible with any processor from the Core 2 line, including the upcoming 45nm parts. The system measures only 12.8” x 8.3” x 8.7” and is almost too dense in features to detail here. Ports include everything from optical and coax SPDIF to Gigabit Ethernet to a pair of eSATA connections tied back to Intel’s powerful ICH9R southbridge. A three-part heatsink apparatus Shuttle calls OASIS covers the chipset and MOSFET modules to help reduce noise even below the normally whisper-quiet levels enabled by Shuttle’s 400W, 80 PLUS-certified power supply. Through ingenious use of space and cable management, the P2 chassis easily houses up to three hard drives and can still accommodate one x16 PCIe graphics card and one PCI adapter. We have yet to find a more value-add solution for downsizing a full-featured desktop, and that’s even before figuring in the overclocking options or integrated biometric (fingerprint) security system. “Our systems integrators love the SP35P2 because it addresses so many common SMB needs,” says Sheila Dy, business development director at Shuttle Computer. “Customers want longer product life spans and something that’s going to be a real productivity workhorse without looking and sounding like one. I mean things like less fan noise, less heat, less energy consumption. You want powerful expansion capabilities without a lot of dongles and breakout boxes and ugly ports. And when you create important data to keep on a PC, you need to lock it down tight, especially in this age of identity theft. Shuttle has been considering all these things for years, and we’ve evolved what we feel is hands-down the smartest, best solution in the channel.” Whether you want to think big or little, the goal is not to think like everybody else. When it can make sense for your customer, push them out of that mini tower mindset. Alternatives can offer them so much more for their money.
MULTI-MADNESS You might compare monitors to cases. The norm is a humdrum 17” LCD panel, much akin to a beige mini tower. You’re never going to make much money on a 17” monitor. Try to thrive in the middle ground and you’ll get washed out. But move toward the smaller and larger extremes and you’re likely to find opportunity. According to Q2 numbers for 2007 from DisplaySearch, 17” SXGA panels had 34% of the worldwide market, but that marks an 8% quarter-over-quarter decline. Counterintuitively, 15” LCDs only own 7.6% of the market, but that’s a 43% gain from Q1. (Chalk it up to volume miracles in the emerging markets.) Meanwhile, 19” widescreens showed 19% growth, and 22” wides blew off the doors with 61% Q-over-Q numbers. Size does matter. Of course, this is America, land of the super size, where the only thing better than one big widescreen display is two big widescreen displays. And the only thing better than that is three or four. The numbers speak for themselves. People want more pixels in their field of view at once. Gamers want a more immersive experience. Financial professionals need to track ever more data in real-time. And ordinary office workers, content producers, and even students can do more work in less time if they don’t have to constantly toggle between windows for the dozen or so apps we all keep open these days. For years, studies have shown a productivity increase of over 10% when moving from a single- to dual-monitor desktop. So why do so few users adopt a multi-monitor setup? You can pick up a 20” LCD now for just over $200! It might not be the best monitor on the market, but still—you get the point. We take it as a given that multi-monitor setups are the way of the future. Honestly, we can’t understand why more people don’t adopt dual-head output. Most of us at RAM use three or four displays on our desks, and all of us still pine for more pixels from time to time. You should be urging multi-monitor for your clients first because of its productivity benefits. Anything you can do to improve your customer’s business is ultimately going to benefit your own business. And naturally, nabbing an extra monitor sale or two on a single PC sale doesn’t hurt a bit.
The trick is to enable your systems now for an easy upgrade path to a multi-monitor setup later when the ROI light finally clicks on over buyers’ heads. Some customers, especially those saying they only want $500 cubicle PCs, are loathe to discuss multi-monitor setups because it implies adding expensive graphics cards to the system. That’s bunk. If you’re smart about how you configure your machines, you’ll pay the extra few dollars for something like Intel’s DQ35JO or ASUS’ M2AVM, based on the Q35 and AMD 690G chipsets, respectively. Both boards support dual-monitor output (one VGA, one DVI) directly from integrated graphics on the motherboard. Tack on a dual-output PCI Express card to the ASUS board and you’ve got quad-monitor support. On the Intel side, you can add a $25 ADD2 card for additional monitor support. The beauty of the DQ35JO board is that it’s also a new-generation vPro motherboard, which paves the way for a host of future service and remote management opportunities-—another hot futureproofing angle we don’t have space to explore here. The potential downside to using integrated graphics for multi-monitor support is that you’re limited to IGP (integrated graphics processor)-class performance. If all the customer needs is 2D work in Office and Web browsing, no problem. Today’s IGPs are more than up to the task. But for more substantial 3D work or in situations where more resource-intensive multitasking is common, look to enable customers with a pair of dual-output PCI Express cards on any motherboard that supports ATI’s CrossFire or NVIDIA’s SLI technologies. Hopefully, as high-end boards start to adopt three and even four slots, this will enable six- and even eight-head monitor configurations. Whether consumers will need that many displays is a different story. Many of the 3D performance and image enhancement benefits of CrossFire/SLI won’t be enabled if you’re working on more than one screen, but with CrossFire/SLI mode disabled, a pair of sub-$100 cards will still deliver excellent results on up to four screens. For larger environments where manageability and user profiles, not to mention application validation and guaranteed uptime, become more important, check out the lines of workstation cards from AMD/ATI and NVIDIA. Specifically, these are the FireMV and Quadro NVS brands, respectively. Both vendors offer PCI and PCI Express adapters that, when used in combination, can scale well beyond four displays. They tend not to use the most cutting-edge GPUs in order to be more stable and to adapt sometimes into half-height cards, some of which will offer passive cooling. If you’re aiming for a two- to four-monitor setup, a pair of consumer cards will usually cost less, but in something like a Shuttle box, you may only have one slot available for graphics, in which case a FireMV or Quadro NVS card is the smart move for maximum display options and scalability. What if your customer has a notebook and still wants a multi-monitor setup? Cloning the screen to a single LCD monitor is easy. Just about any notebook can do that. But if you want to spread the desktop across two or three monitors, now that’s something else. Check out Matrox’s TripleHead2Go, a brilliant little box that can work in conjunction with a notebook’s integrated display and dual-link output to deliver up to quad-head operation: the notebook’s screen plus three additional monitors. There are dual- and triple-head models available for both analog and digital displays. With the TripleHead unit, you can have a three-monitor desktop of up to 3840x1024 resolution, and there’s even a bezel-management setting that helps smooth the transition of scenes flowing from one screen to the next. Finally, there’s a smart way to provide the same multi-monitor future-proofing to mobile workers as there is for desktop types. Now that you’ve got the tools to enable the solution, it’s just a matter of showing customers the superior experience multi-monitor adoption can offer. We could make the same point about all of the future-proofing plays we’ve detailed above. You can talk about benefits and investment protection until you’re blue in the face. Customers need to see and touch these ideas. Show those 45nm and quad-core benchmarks in action. Play the blue laser HD content on multiple screens. Make the future-proofing value tangible, and prepare to invoice accordingly. |
||||||||||||||||
 Back to top
|
||||||||||||||||
Copyright © 2007 RAM Magazine. All rights reserved.
Do not duplicate or redistribute in any form. |
||||||||||||||||
