Are we hearing the DDS swan song? Maybe, but only the OEMs will call the tune.
The DDS market is, admittedly, a substantial piece of pie. Gartner/Dataquest reports total factory revenues for 2001 was just over $650 million. Drive manufacture is currently split between four companies (See Figure 1), with Hewlett-Packard and Seagate sharing the North American market. The original technology developers of DDS, Sony, and Hewlett-Packard, have both formally announced that DDS-4 is the last generation of DDS. There will be no DDS-5. This classes DDS as an end-of-life product.
Alive and Kicking
For a technology whose days are said to be numbered, DDS is as robust as any tape technology in its class. At the 2002 PlanetStorage conference, veteran tape analyst Fara Yale said in a formal presentation that "although multiple competing technologies are targeting this space, none have made any major impact on DDS drive sales." DDS-1 saw its last shipment in 2000, completing a 12-year lifetime. Capacities on DDS-2 through DDS-4 range from 4-20GB with transfer rates running from 366KB/sec to 3MB/sec DCLZ data compression is applicable to all formats. The tape media is a highcoercivity metal particulate formulation.
What is the secret of DDS' continuing appeal? Bob Hawkins, Seagate's product line marketing manager for DDS products points to three elements. "First, there is a huge installed base. Next, interchange: There are multiple vendors for media and drives...there are no proprietary issues. Finally, DDS has a low media cost." He also observes that Seagate's DDS business is "strong and consistent."
Hawkins also notes that DDS is still viewed as a market leader and that the business is likely to extend well into the decade and beyond. Analysts confirm this point of view, forecasting only gradual declines in unit shipments into the middle of the decade (see Figure 2).
If the DDS market is as strong as the players indicate, why are so many different technologies trying to stake a claim in this space now? Analyst Robert Abraham at Freeman Reports suggests that the issue is time to market. Said Abraham: "The process has started; there is business to get right now. The people who get that business will be the incumbents." And that explains the urgency that some vendors feel; it is difficult to take market share from an incumbent. Not impossible, but difficult.
If in fact there is sound business reasoning to lay claim to the title of DDS successor, which technologies represent likely candidates? Several technologies, announced and otherwise, have thrown their hats into the ring.
One of the contenders is the VXA architecture, owned by Exabyte Corporation through its acquisition of Ecrix. VXA is the first tape technology to read and write data in packets. Packetizing opens up new modes of operation that lead to improved reliability and performance.
Additionally, VXA can operate at variable speed. In effect, this eliminates the need for backhitching. No backhitching means higher effective throughput and less wear and tear on drives and media. VXA also offers overscan, in which the heads can read data from any physical location on the tape, without having to follow tracks from beginning to end. This capability gives a boost to reliability, data retrieval, and interchange.
Exabyte's Kelly Beavers believes that within three to four years, the tape world will be split between LTO and VXA products. Said Beavers: "We tell OEMs to replace DDS-4 with VXA-2 drives. This gives them four times the size and twice the speed."
Beavers reviewed some of the DDS understudies. "Travan won't work. The choice wasn't made, and users need a bigger, faster drive." He also noted: "DDS came up where other drivers were bigger, faster, and more costly. This rules out Sony ATT. ATT-3 is too expensive; no one will pay a premium for bigger and faster products."
He also looked at issues of media. "You need affordable media, and a choice of sizes. Short tapes are for incremental backups, and long tapes are for archival. This is an option that DLT doesn't offer." To illustrate the point, Beavers points to the price decline in DDS media: "It's a fire sale."
AIT, Sony's 8mm Contender
VXA is only one of the 8mm technologies seeking to succeed DDS. The other is the AIT format from Sony (the inventors of DDS). All' is characterized by innovations like Memory-In-Cassette (MIC), larger capacities, high data transfer rates, and high reliability.
The AIT format was a new mid-range tape technology addressing new positions in capacity and performance, as well as access to data. The AIT recording technology today offers uncompressed capacities including 25GB, 35GB, 50GB, and 100GB, all using the compact 8mm cartridge form-factor. The higher compression specifications for capacity and performance for the AlT family are achieved through the incorporation of the adaptive lossless data compression (ALDC) technology, which delivers an average 2.6 to 1 compression ratio.
The MIC innovation deserves special attention. Higher data-search rates are achieved through the assistance of MIC technology, which provides greater intelligence to a tape drive and media combination through the use of a flash-memory chip incorporated into the data cassette. The MIC chip contains key parameters such as a tape log, search map, and application information that allow flexible management of the media and its contents.
The AIT format is based on helical-scan recording using advanced metal-evaporated (AME) media. With its pure metal magnetization layer and protective "diamond-like" carbon coating (DLC), a much higher recording density can be implemented. Additionally, the low tape speed and low tape tension mean a gentle handling of the media over the recording heads, promoting both higher mechanism and media life.
In addition, since helical-scan recording does not require fast tape speeds, the re-positioning time of the drive is much faster than linear serpentine technologies, providing better performance over a wide range of host data rates. Helical-scan also eliminates the need of linear technologies to "serpentine" the tape (moving back and forth hundreds of times) in order to fill the cartridge with data, but rather uses only one pass from end-to-end in order to read or write the entire cassette
John Woelburn at Sony identifies AIT as a natural fit, calling out AIT-3 as the logical successor. AIT-3 is the third-generation AIT technology implementation, doubling the capacity and performance of the prior generation. AIT-3 is backward read and write compatible with all AIT-2 and AIT-1 media, and offers a native capacity of 100GB and a native data transfer rate of 12MB/sec in a 3.5-inch form-factor. The increased capacity of AIT-3 was achieved by doubling the track density (expressed as recording tracks per inch) compared to the AIT-1 and AIT-2, while the doubling of the data transfer rate was achieved by simultaneously utilizing two recording heads on-tape.
Said Woelburn: "Sony will continue to promote AIT as the DDS upgrade. They share a common form factor and a common recording style. It's certainly not as big as the four-inch cartridges." Also noted is a clear upgrade path with no compatibility issues.
When asked about Sony's current commitment to DDS, Woelburn noted, "DDS enjoys a big installed base, and Sony is still producing the drives. But we'll not be producing a next generation; we think that AIT fits the bill with less risk. And for customers looking at automation, AIT makes sense."
Is 8mm the Way?
Eight millimeter is not the only option as regards DDS replacement. Benchmark offers a DLT drive family that has gained both market share and recognition.
Last June, Benchmark introduced its second-generation half-high form factor DLT tape drive and its accompanying media, extending the ValuSmart tape family. ValuSmart now features Benchmark's half-height design that incorporates important core technology components, offering multiple operating points for greater application opportunity. The ValuSmart family integrates leading technologies including magneto-resistive (MR) heads for greater density and durability, integrated circuitry for compact design and reduced power consumption, and simplified market tested mechanics for form factor and reliability.
Steve Berens at Benchmark acknowledges the staying power of DDS. "DDS has been very effective by lowering its price point. They've held off what could have been a catastrophic decline."
Benchmark does not identify its product line as replacements for current DDS drives, but for the next generation. Said Berens: "We are replacing what would have been DDS-5." This strategy is a recognition of DDS-4's comparatively recent ramp to volume. Berens related that when the disk drive industry saw its decline, it slowed the introduction of DDS-4. Only recently has DDS-4 outshipped DDS-3.
"Basically," noted Berens, "people don't want another form factor." But to Berens, this resistance reaches to 8mm products. "Relatively speaking, DDS is replacing 8mm. It's hard for 8mm to be the understudy for DDS when you're declining faster than the technology that you're trying to replace." Berens refers to. analyst figures that identify 8mm taking a much more severe decline than the roughly 14% slip that DDS has experienced.
But any decline in the 8mm space may not be permanent by any means. Freeman Reports Compact Tape Outlook 2002 says: "Growth is expected to resume [in the 8mm space] immediately, however, due to the availability of new high-end models and well-defined migration plans for the market participants."
Waiting in the Wings
There are others who are looking for a place in the DDS replacement sunlight. IBM will be expecting a shot; they have OEMed an external VXA drive from Exabyte, and specifically position the drive for DDS replacement.
At press time, Quantum Corporation, leader in DLT tape technology, was contemplating a new low-end strategy. A Quantum spokes-person confirmed that the company will compete in the low-end space, but was not prepared to share that strategy.
An especially interesting potential is the unconfirmed rumor that Seagate is preparing a hybrid DDS drive...sort of a DDS-4-plus that would offer increased capacity and transfer rate. When queried on this point, Seagate's Hawkins said that the company was still investigating opportunities and was not ready to announce a product or a technology. Sony's Woelburn said that another DDS might have some appeal, but is putting off the inevitable.
It's the OEMs
The need for DDS replacement is likely as disk capacities and speeds become greater; eventually the need to get more on one tape cartridge will tax DDS beyond the breaking point. Then a decision will have to be made. As Gartner's Yale observes: "This technology will live on in sizable number for several more years until major OEMs offer and market re-placement products."
There is little doubt that the OEMs bear the burden of decision, but is that decision far away? Sony's Woelbum noted that "Some OEMs are facing the decision this year." Additionally, he noted that one or two OEMs are looking at a market requirement for a DDS-4+. This would be the hybrid that Seagate is. rumored to be developing.
All of the DDS understudies waiting in the wings have one thing in common. Job one for all the candidates is, as Exabyte's Beavers said, "to sell the OEMs." Until this is done, contention for the DDS replacement spotlight will persist.
Editor's Note: Please consult the other articles in this special tape section to check technology status on Travan, SLR, and ADR technologies. Additionally, consult the sidebars on media and automation for a rounded view of technology status and direction.
[FIGURE 2 OMITTED]
Figure 1 MARKET SHARES--WORLDWIDE REVENUE DAT DRIVES Hewlett-Packard 46% Seagate 31% Sony 21% MKE 2% Note: Table made from pie chart
|Printer friendly Cite/link Email Feedback|
|Publication:||Computer Technology Review|
|Date:||Aug 1, 2002|
|Previous Article:||Fabric virtualization: the roadmap: Visualize, virtualize, automate. (Business of Technology).|
|Next Article:||Intel and Microsoft absquatulate! Is all hope lost for InfiniBand?|