Reviews Leaderboard Database Reference Search StorageReview Discussion Reliability Survey Search About StorageReview.com Contents

Seagate Cheetah 36XL ST336705LW
  March 13, 2001 Author: Eugene Ra  

Seagate Cheetah 36XL ST336705LW Available Capacities *
Model Number
Capacity
ST336705LW
36.7 GB
ST318405LW
18.4 GB
ST39205LW
9.2 GB
* The benchmark scores presented in this review represent expected performance across the entire line.
Estimated Flagship Price: $599 (60 GB)
Evaluation unit provided by Seagate Technologies.


Introduction

Seagate's venerable Cheetah series now features five-generations of tried-and-true 10,000 RPM spindle speed technology. The Cheetah 4LP broke new ground as the first drive to feature 10k RPM speeds as well as a sub 8 ms seek time. Its successor, the Cheetah 9LP, elevated things to a new level by shaving seek time down to just 5.2 milliseconds. Since then, however, Seagate's 10k improvements have been evolutionary at best. Though the 18LP and the 36LP took advantage of ever-increasing areal densities to deliver higher capacities and transfer rates, access times remained constant. Seagate's next-generation unit, however, finally features a reduced seek time. Do we speak of the 36XL?

Well, no, not quite . With its "XL" designation, the Cheetah 36XL is more of a "half-generation" update. The 36LP, after all, already delivered a Seagate low-profile 36 gig 10k unit. So what does the 4.5th generation 36XL offer? Improved areal densities and a resulting platter count decrease, of course. The 36XL requires just 4 platters do deliver its 36 gigs of capacity. Buffer size, at 4 megs, remains unchanged. Spindle speed, of course, is 10k.

Seek time diverges a bit from predecessors... upwards, amazingly. While the Cheetahs 9LP, 18LP, and 36LP all featured a 5.2 millisecond rating, the 36XL comes in a hair above, specified at 5.4 ms. Why? Perhaps the increased areal density... the 36XL packs 50% more data per platter than the 36LP. We'd figure that decreased platter count, however, would offset the extra precision that higher densities require. As we'll demonstrate, all this hair-splitting is unnecessary- specified seek times aren't always reliable.

Seagate positions the Cheetah 36XL squarely at the mainstream SCSI workstation and server markets. Unlike the higher-end 73LP, the 36XL is available only with the Ultra160 SCSI interface... there aren't any FC-AL versions. An industry-standard 5-year warranty backs these enterprise-class drives.

Let's see how the 36XL's low-level measurements stack up.


WB99/Win2k Low-Level Measurements

[an error occurred while processing the directive]

Click here to examine the STR graph for this drive

WinBench 99 measures the Cheetah 36XL's access time at 8.5 milliseconds. Accounting for 3 milliseconds of rotational latency yields a measured seek time of 5.5 milliseconds... right on the specs, something a bit unusual for Seagate these days. This means that despite its slightly higher specified seek time, the Cheetah 36XL measures in a full half millisecond lower than the 36LP.

Interestingly, though the Cheetah 36XL's higher areal density allows it to improve on the 36LP with an outer-zone transfer rate of 40.1 MB/sec, it doesn't quite match up to the highest rates offered by the Fujitsu MAJ or Quantum Atlas 10k II. Transfer rates "decay" very nicely, though, with nearly 24 gigs of the drive offering rates in excess of 36 megs/sec.


WB99/Win2k WinMarks

[an error occurred while processing the directive]
The Cheetah 36XL significantly improves on the 36LP's WinMark scores, coming within spitting distance of the record-setting Quantum Atlas 10k II. The 36XL's scores in both the Business and High-End WinMarks lag those of the Atlas by a mere 3%. Its Business Disk WinMark score of 8.6 MB/sec represents a healthy 20+% improvement over the 36LP.


IOMeter Performance

[an error occurred while processing the directive]
Note: The IOMeter indices are a normalized average of Light, Medium, and Heavy loads.

Quantum's Atlas 10k II rules the roost when it comes to the IOMeter File Server Index. Its score of 242 bests all drives save only the Cheetah X15. The Cheetah 36XL's score of 240, however, places it less than 1% behind the Atlas.

A Workstation Index comparison favors the Cheetah 36XL. Its score of 253 leads the Quantum by a significant 12%. Database Index scores are very close... the Cheetah leads by a mere 1%.


Conclusion

The 36XL's reduced platter count allows it to run warm but not hot in our rather small testbed case without using active cooling. Integration into larger cases shouldn't be a problem. Though seeks are fairly quiet, this review sample featured an idle noise harmonic that vibrated the entire case, creating a noticeable though not overly distracting hum. A previous sample that we've used didn't exhibit this harmonic- we'd hazard that the noise was peculiar only to our review samples and isn't representative of the vast majority of 36XLs.

The StorageReview.com Safe Buy Award



Simply put, this designation means we'd purchase this product without regret. Sure, there may be a slightly better, slightly faster, and/or slightly less-expensive model from a competitor, but you can't go wrong with this particular unit. This award is applicable, of course, to all units at the top of their class, but also applies to units that, though not quite best-of-class, provide a strong showing nonetheless. Overall, despite its "evolutionary" nature, the Cheetah 36XL does an admirable job in improving the performance (and, as one hopes from reduced part counts, reliability) of Seagate's celebrated line. A quick glance at StorageReview.com sponsor Hyper Microsystems reveals that while the 36XL's only slight more expensive than the Cheetah 36LP, an Atlas 10k II may be had for significantly less. This isn't surprising- after all, the 36XL is a brand-spanking-new unit while the 10k II has been around for a while. Seagate's pricing may be come more competitive after a few months. When it comes to absolute performance, however, the Cheetah 36XL establishes itself as the 10k RPM drive to beat.

[an error occurred while processing the directive]

HOME | ARTICLES | LEADERBOARD | PERFORMANCE DATABASE | REFERENCE GUIDE
COMMUNITY | RELIABILITY SURVEY | SUPPORT SR! | ABOUT SR |

Copyright © 1998-2005 StorageReview.com, Inc. All rights reserved.
Write: Webmaster