Up the SAN-Box – storage area networks take off – Statistical Data Included
fibre channel and new directions in network storage
1998 is the year network storage turned sexy.
Now that’s meant as a PC concept, not as a concept that’s PC. Prior to 1998, talking about storage held all the fascination of a speech at a plumbers’ convention. While the pipe-and-solder people might discuss copper versus plastic or 1.5-gallon versus 3-gallon toilets, on the computer side, it was copper versus fiber or 100Mbps versus 1Gbps. But the concept’s the same, and until recently, the unvarnished truth about network storage has hardly been the stuff of gripping drama.
So what happened in 1998? The storage area network (SAN) concept took off.
That year, more vendors at the enterprise level announced SAN storage solutions than ever before. SAN has become all the rage. Today, almost every major player in the enterprise storage arena has announced or delivered SAN product.
Never mind that others pioneered these tools years before (just ask data-protection vendor Vinca, which launched the whole SAN movement), today it’s EMC, Network Appliance, and Comdisco. In fact, SANs provide a superior level of storage access for networks. Compared with host-attached devices, SAN-attached storage systems are faster, more reliable, and easier to manage. So is an optical-drive-equipped SAN in your future?
A DATE WITH DENSITY
To understand why SAN’s promise of high-density storage may or may not be in your future, we’ll need to take a quick look at how SANs work, where storage bottlenecks are, and what solutions there are to fix these bottlenecks.
Compare, for instance, traditional host-attached storage with network-attached. Both provide convenient access with a simple setup. If there’s a fault with the server, the storage becomes inaccessible. There’s also significant overhead when using the standard Ethernet protocol. Whether it’s 10BaseT, 100BaseT, or gigabit Ethernet, there’s always the Ethernet overhead to contend with–which averages a minimum of 10 percent and could rise much higher depending on the types of devices connected and the amount of traffic each generates.
So how do we make a storage area network out of a variety of storage devices? Each device connects directly to the network cable, as would occur with the familiar network-attached storage (NAS). However, this network cable is accessed only by a set of network servers (these can be database, application, or Web servers). The loss of one or more servers does not prevent the remaining servers from accessing the SAN-attached drives. In fact, the Fibre Channel loop itself can be broken and access can continue utilizing the self-healing capabilities of the protocol.
In a SAN, the administrator can add or remove storage devices as required without downing a single network server. This provides a robust and reliable storage service. It also means that the storage service can more closely match actual need, rather than building in a lot of extra capacity to meet expected future requirements. With traditional servers, adding drives is so complex it is best to do it only when necessary. So over-capacity is a method for compensating for this complexity.
The SAN architecture does not demand a specific protocol. In fact, in the past, SANs were constructed using SCSI and could still be today using its multiple initiator capabilities. However, it is universally understood that SANs as we have come to define them use the Fibre Channel protocol and wiring.
One way to think of Fibre Channel is as a super-SCSI, built upon all the amassed development understanding gained over the last 15 years in storage since SCSI first appeared. Fibre Channel has more throughput than UltraSCSI (100MB/sec versus 80MB/sec) and supports more devices. While significant discussions persist over the more universal presence of SCSI devices versus the less-common Fibre Channel, in one key respect, they are completely different–distance. The Fibre Channel specification supports connections over single-mode fiber of up to 10KM, while SCSI remains limited to distances closer to 10 feet.
Like SCSI, however, Fibre Channel needs devices constructed with this type of connection in mind. Currently, a variety of RAID and tape subsystems are offered with built-in Fibre Channel support.
Fibre Channel also features low-overhead compared with a traditional Ethernet connection. For that reason, a Fibre Channel SAN is superior to NAS for accessing large amounts of data. So why isn’t Fibre Channel the preferred solution? As a destination, 120mm optical drives on a SAN have various limitations from capacity to speed to limitations inherent in the Fibre Channel SAN itself.
THE DENSITY BIT
One challenge to incorporating DVD/ CD-R/RW/RAM media on a SAN, such as for backup and archiving purposes, is the relatively small density of the individual discs involved. From DLT to AIT, tape delivers four to five times the capacity of a single DVD-RAM disc, with 20GB a common capacity for high-end tape devices. How can writable DVD’s 2.6GB to 4.7GB–let alone CD-R’s 650MB–hold up?
We could see increased capacity for the discs eventually, but it won’t happen anytime soon. Current laser technology stands at 780nm for CD recording and 650nm for DVD. To improve density, we need to improve lasers to the 440-450nm range (the so-called “blue” laser). What does this mean to density? The blue laser recording would increase capacity by four times–making a single DVD-RAM disc support 37.6GB double-sided. That would make it a definite contender against tender tape.
If we want to go beyond that, though, we run into a few problems. Even if lasers improve, disc capacity will still hit the wall when we reach the limitations of the plastic substrate used in the discs. At about 300nm and lower, the plastic itself is no longer transparent–requiring some new substrate technology to replace plastic–a significant barrier to the extensive (and expensive) installed base for manufacturing media.
Another drawback to decreasing the wavelength, and thereby increasing density, comes from heat. As lasers go from red to blue (650nm down to 440nm), the temperature of the laser almost doubles–from 30 degrees Celsius to 50 to 60 degrees. Granted, the increased temperature is low compared with Intel’s egg-cooking Pentiums, but it isn’t the interior temperature that’s the problem. Trouble is, the hotter temperature means a shorter life span–40 percent shorter. Today’s recorders average 5,000 to 5,500 hours between drive failures. A blue-laser-equipped recorder might only last 3,000 to 3,500 hours.
For a standalone recorder, this may be acceptable. It is a deal-killer for a network.
BACKUPS WITHOUT HICCUPS
Another drawback to enterprise backup and archiving using contemporary optical technology is speed. Seagate, for example, boasts that its Backup Exec for NT is capable of handling up to 28MB per minute. This respectable rate is quickly diluted with today’s hard disks–as it would take one full hour to back up a small 1.5GB hard drive–and that at its maximum speed. On a network with 100 to 500GB of storage, a few servers would take a battery of such recorders to keep backup time to a minimum.
Granted, a SAN-attached optical backup or archive system doesn’t fall prey to the shrinking backup window. Today’s network no longer has a “down” time. In the 1980s and early 90s, when 5:00 hit, most workers went home. By midnight, the network was quiet–perfect for the administrator to run backup. In 1999, that’s no longer the case, as the Internet now makes the corporate data-information, base, sales information and electronic commerce open for business 24-hours a day. So when can a lengthy backup session occur?
With a SAN, backup can occur at any time, since it adds no burden to the network. But the latency problem remains for optical versus tape backup. By the time a large array of DVD- or CD-R drives completes their backup cycle, the data is dated much older than the contents of a tape backup. How much more will it cost to reconstruct the lost data in the event of a system failure using a 120mm system versus a tape backup program?
BUILDING THE NETWORK SUPPORT NETWORK
If we look beyond the drives themselves, we also need to take note of the tower and jukebox support for SANs. As of post-AIIM April 1999, there is only one optical drive/subsystem/jukebox (from DISC) ready to run on a SAN using Fibre Channel. All the other vendors promise support as soon as market demand requires it. All acknowledge the potential benefit to using SANs for RAID and other magnetic storage. The question remains, though, of where shared CD-ROM and DVD-ROM as well as recordable optical fit in the SAN equation.
Currently, only a few vendors offer backup and archiving software supporting CD-R/CD-RW for PC systems. These include Seagate’s Backup Exec and NTI’s Backup NOW! Seagate only began offering a CD-R/RW-compatible Backup Exec version roughly 18 months ago, and Backup Now!, from longtime premastering tool provider NTI, debuted in late 1998. So the data is still a little too sketchy to make a clear assessment here.
SAN technology faces hurdles of its own for broad acceptance in the network marketplace. Current versions of Microsoft Windows NT, Novell NetWare, and the various UNIX flavors don’t natively support SAN connections like they do SCSI. So a proprietary service such as Mercury Computer Systems’ SANergy or Transoft’s SAN Manager is needed to enable SAN connections. Both support the native NT file system with extensions. Regrettably, neither supports NetWare (which is enjoying a strong resurgence in the market with the release of NetWare 5.0.) And, while popular, the NT Server has yet to amass a broad fan base among MIS professionals. As one reviewer has put it, “supporting SANs [with Windows NT] is a cruel practical joke.”
Novell itself offers support for Fibre Channel in its own offering: the High Availability Server for NetWare (HASN). However, HASN does not support a Fibre Channel SAN, only individual Fibre Channel devices. This is because HASN is currently based on NetWare 4.11 technology and not the latest NetWare 5.0. It therefore needs to mount each of those drives in the server memory. This means that the latency between the loss of one server and the time another can take over its storage devices (i.e., mount them) can be lengthy. (It varies depending on the storage capacity of the drives.) So while there’s protection from losing complete contact with the devices, there’s no continuous availability as with a SAN.
SERVING THE FUTURE
A question about the future viability of SAN technology with Fibre Channel remains. Intel is developing its own NGIO (Next Generation I/O) while a consortium of Compaq, Hewlett-Packard, and IBM are working on a competing protocol called Future I/O. Both of these potentially offer a major advantage over Fibre Channel since they feature backward compatibility with older I/O technologies. This makes a compelling argument to MIS with its immense investment in existing storage.
Finally, SAN-based technology faces a significant hurdle in expanding beyond the enterprise, large-scale site. For the small- to medium-sized network, the budget is a primary concern for network purchases. In the last six months to a year, the fine return-on-investment (ROI) equation of SAN technology has been tarnished by the plummeting cost of conventional hardware. Why invest $50,000, $150,000, or more in a fully configured SAN system if a $15,000 or $20,000 investment will do?
For example, suppose we want to have a high-availability system so if one server goes down, we still have access to our storage. We could do this easily through a SAN, but we could also invest $5000 or $7500 in a server-class machine with ATAPI drives, then load Vinca’s Standby Server software on it to mirror our primary server. For $15,000 or less, we have full data protection using familiar, commodity hardware and software.
THE FRUITS OF FIBRE
Okay, we concede that optical backup systems won’t make SAN an instant hit with users. What will? There are two possible scenarios–one as a destination for data, the other as the origin.
Deep archiving, matched with the unprecedented distance supported by Fibre Channel, can be run off-site for data vaulting purposes on a SAN. Being offsite, it provides an additional protection against localized disasters (such as a fire, explosion, or flood). Unlike wide-area network (WAN) links, the FC link is instantaneous and full gigabit speed. A system such as the Plasmon M500 jukebox or the NSM Galaxy DVD-RAM jukebox could serve in this capacity.
As an origin, a DVD-equipped SAN could be ideal for a special effects house, media development center, or television station where real-time video editing and production occurs and where the cost of the connection is more than matched by the reduced production time.
As an origin, an optical subsystem on a SAN would provide ISPs with an ideal shared resource over the Web which would be impregnable to hackers. The native SAN allows more than one Web server to access the data stored on its devices. It also provides for continuous network growth often seen with ISPs without downing those servers.
And not even the lack of direct device support for Fibre Channel is as great a hurdle as it might seem. There are several vendors, including Hewlett-Packard, ATTO, and MicroNet, which offer bridges from SCSI to Fibre Channel. ATTO’s FibreBridge, for example, supports any existing SCSI device up to and including UltraSCSI. MicroNet’s FibreFlex, which combines FC and SCSI technology, offers users the ability to connect devices over longer distances than SCSI itself while retaining the advantages of the SCSI destination device.
Theoretically then, one could attach any SCSI optical jukebox to an FC-to-SCSI bridge and provide the network with DVD or CD service. Whether this extra step can remain transparent to the user (for example, will backup software work well with these bridges?) remains to be seen, but the promised support in Transoft’s SAN Manager and Mercury’s SANergy could make this happen.
SAN IS NOT NAS BACKWARDS
SAN technology is a remarkably useful addition to enterprise RAID and tape backup. Its universal server access, speed, and abundant hardware support make it a recommended solution for those tasks. For offsite data vaulting, one or more optical jukeboxes on a SAN provide a superior solution to traditional tape. For optical storage, however, an NAS approach provides equally strong jukebox and tower CD and DVD-ROM access with less attendant cost and easier management.
companies mentioned in this article
ATTO Technology, Inc.
40 Hazelwood Drive, Building 106, Amherst, NY
14228; 716/691-1999; Fax 716/691-9353;
http://www.attotech.com; INFOLINK #400
6111 North River Road, Rosemont, IL
60018; 800/321-1111, 847/698-3000;
http://www.comdisco.com; INFOLINK #401
372 Turquoise Street, Milpitas, CA 95035;
406/934-7000; Fax 408/934-7007;
http://www.discjuke.com; INFOLINK #404
35 Parkwood Drive, Hopkinton, MA 01746;
Information Storage Americas, 800 South
Taft Avenue, Loveland, CO 80537;
800/826-4111, 970/679-5227; Fax
Mercury Computer Systems, Inc.
1999 Riverneck Road, Chelmsford, MA
01824-2820; 978/256-1300; Fax
MicroNet Technology, Inc.
80 Technology, Irvine, CA 92618; 714/453-6000;
http://www.micronet.com; INFOLINK #413
Network Appliance, Inc.
2770 San Tomas Expressway, Santa Clara,
CA 95051; 888/463-8277, 408/367-3000;
Fax 408/367-3151; http://www.
networkappliance.com; INFOLINK #414
NewTech Infosystems, Inc. (NTI)
2081 Business Center Drive, Suite 260,
Irvine, CA 92612; 949/622-6970; Fax
Seagate Software, Inc.
400 International Parkway, Heathrow, FL
32746; 800/327-2232, 407/531-7500;
Fax 407/531-7670; http://www.seagatesoftware.com;
Transoft Networks, Inc.
425 East Cota Street, Santa Barbara, CA
http://vwvw.transoft.net; INFOLINK #420
1201 North 800 East, 0rem, UT 84097;
801/223-3100; Fax 801/223-3107;
http://www.vinca.com; INFOLINK #421
NETWORK OBSERVER columnist David Doering
(email@example.com), an EMedia Professional
contributing editor, is also senior analyst with
TechVoice Inc., an Orem, Utah-based consultancy.
Comments? Email us at firstname.lastname@example.org, or
check the masthead for other ways to contact us.
COPYRIGHT 1999 Online, Inc.
COPYRIGHT 2000 Gale Group