Accelerate Your Mac! - 1998 Review of dual SCSI Controller Raid Tests

Lab Tests: Dual PCI Ultra-Wide SCSI Controller Raid
By Mike Breeden
Published: February 1998

Part 2 of a continuing series on SCSI controllers, disk drivers and Raid software.

Intro | Cards Tested | ATTO Benchmark | FWB Benchtest | Time Drive | MacBench | Summary

Accelerate Labs Tests: Mac PCI Ultra-Wide SCSI Controllers
- Dual SCSI Card Raid Tests -

Introduction:
This is the second installment of my SCSI Card Comparison Report. The first installment reported on tests with a single card attached to a single 2.2GB Ultra-Wide SCSI Quantum Atlas II drive, using both Conley's SoftRaid 2 and FWB's Hard Disk ToolKit drivers. This installment compares the performance of a two drive Raid array using pairs of each SCSI card (ATTO ExpressPCI, Adaptec 2940UW, Initio Miles and HammerStorage JackHammer SE) using all the popular Mac Raid Software (ATTO Raid, Anubis Raid, Conley SoftRaid 2.0, FWB Raid ToolKit 2.06 and Adaptec Remus Lite v1.4).
A dual card setup (each driving one or more drives, in this case one each) usually provides higher performance than either a dual-channel card (2 SCSI channels on one card) or a single SCSI card. The final installment in this series shows the results of the same two drive Raid array with only a single SCSI card for comparison.
Since most of my writing takes place between midnight and 6am, please bring any typos you find to my attention.
Test System:
The base system used for test was a refurbished PowerTower Pro 180, purchased for $1380. It was chosen for its expansion potential and low cost, since most of the base components would not be used in the final system (CPU card, ram and video card). The hard disk(s) used in the test were a pair of Quantum Atlas II 2.2GB, 7200rpm, 512k cache drives partitioned as a single 4GB Raid volume using the various software packages mention previously.
Due to issues between the current Adaptec 2940UW and the fastest CPU card I've tested, the PowerForce 250 (running reliably at 292mhz with a 1:1 cache), the stock 180mhz 604E processor was used for this comparison. [Update: Powerlogix's 1999 Powerforce control software should resolve this issue by disabling speculative processing.] Keep in mind that performance would increase somewhat with a faster CPU.
The operating system used was OS 7.6.1 with the standard array of Quicktime 2.5 and Quickdraw 3D 1.5.3, PC exchange, Connectix's Speed Doubler 2 and Motorola's LibMoto extensions active. Installed Ram was 384MB, consisting of four 64MB and four 32MB Dimms on loan from MacGurus. A disk cache setting of 512K was used for all Raid tests, rather than the 4MB cache used for the single card/single drive tests in Part 1. The video card used during this test was an ATI Nexus GA.
This report shows the results of pairs of each of the four card models with ATTO Raid 1.1, CharisMac Anubis Raid v1.36, Conley SoftRaid 2.0, FWB Raid ToolKit 2.06 and Adaptec's Remus Lite 1.4. For comparisons of scores with the same software using only a single SCSI card, see the Single SCSI card Raid Test report. Stay tuned to my main Accelerate Your Mac! site for update notices as well as other performance, compatibility and hardware news and sale notices updated several times each day.
Future Installments:
The next installment will show single card/dual drive array results using all the cards and software listed here. After reading the dual card/dual drive tests in this report, you may want to review the single card/single drive test report if you haven't already.
Viewed as a whole, this series will clearly show the best card and driver software in almost every possible scenario. Keep in mind these tests were run on a 180mhz 604E CPU based machine and that scores would improve with a faster CPU card. Since the Adaptec 2940UW has problems working properly with the PowerForce G3 CPU cards, I was forced to use the much slower stock PowerTower Pro 180 CPU card for all the tests.

Feedback

I welcome comments about this review and/or the products involved.

I'd like to thank the following companies for providing products for test, saving me precious funds - Adaptec, CharisMac, Conley Corporation, Initio, and a special thanks to MacGurus for providing pairs of many of the cards and Raid ToolKit.
If you find this review and site helpful, please let these companies know their efforts are appreciated.

You can follow my preferred path through the report by continuing to the next page, or use the links below to jump to a specific page.

Index of Dual Card/Dual Drive SCSI Raid Comparison
Intro | Cards Tested | ATTO Benchmark | FWB Benchtest | Time Drive | MacBench
= Results Summary = | Go to Single SCSI Card Raid Tests | Go to Single Card/Drive Tests
- or -
Back to WWW.XLR8YOURMAC.COM

Copyright © 1998.
Thanks to my better half for tolerating all the time I spend on this site.

No part of this sites content or images are to be reproduced or distributed in any form without written permission. Please ask before using quotes or images.
All brand or product names mentioned here are properties of their respective companies.
Users of this site must read and are bound by the terms/conditions of use


Lab Tests: Dual PCI Ultra-Wide SCSI Controller Raid By Mike Breeden Published: February 1998
Part 2 of a continuing series on SCSI controllers, disk drivers and Raid software.
Intro \| Cards Tested \| ATTO Benchmark \| FWB Benchtest \| Time Drive \| MacBench \| Summary
Accelerate Labs Tests: Mac PCI Ultra-Wide SCSI Controllers - Dual SCSI Card Raid Tests -
Introduction: This is the second installment of my SCSI Card Comparison Report. The first installment reported on tests with a single card attached to a single 2.2GB Ultra-Wide SCSI Quantum Atlas II drive, using both Conley's SoftRaid 2 and FWB's Hard Disk ToolKit drivers. This installment compares the performance of a two drive Raid array using pairs of each SCSI card (ATTO ExpressPCI, Adaptec 2940UW, Initio Miles and HammerStorage JackHammer SE) using all the popular Mac Raid Software (ATTO Raid, Anubis Raid, Conley SoftRaid 2.0, FWB Raid ToolKit 2.06 and Adaptec Remus Lite v1.4). A dual card setup (each driving one or more drives, in this case one each) usually provides higher performance than either a dual-channel card (2 SCSI channels on one card) or a single SCSI card. The final installment in this series shows the results of the same two drive Raid array with only a single SCSI card for comparison. Since most of my writing takes place between midnight and 6am, please bring any typos you find to my attention. Test System: The base system used for test was a refurbished PowerTower Pro 180, purchased for $1380. It was chosen for its expansion potential and low cost, since most of the base components would not be used in the final system (CPU card, ram and video card). The hard disk(s) used in the test were a pair of Quantum Atlas II 2.2GB, 7200rpm, 512k cache drives partitioned as a single 4GB Raid volume using the various software packages mention previously. Due to issues between the current Adaptec 2940UW and the fastest CPU card I've tested, the PowerForce 250 (running reliably at 292mhz with a 1:1 cache), the stock 180mhz 604E processor was used for this comparison. [Update: Powerlogix's 1999 Powerforce control software should resolve this issue by disabling speculative processing.] Keep in mind that performance would increase somewhat with a faster CPU. The operating system used was OS 7.6.1 with the standard array of Quicktime 2.5 and Quickdraw 3D 1.5.3, PC exchange, Connectix's Speed Doubler 2 and Motorola's LibMoto extensions active. Installed Ram was 384MB, consisting of four 64MB and four 32MB Dimms on loan from MacGurus. A disk cache setting of 512K was used for all Raid tests, rather than the 4MB cache used for the single card/single drive tests in Part 1. The video card used during this test was an ATI Nexus GA. This report shows the results of pairs of each of the four card models with ATTO Raid 1.1, CharisMac Anubis Raid v1.36, Conley SoftRaid 2.0, FWB Raid ToolKit 2.06 and Adaptec's Remus Lite 1.4. For comparisons of scores with the same software using only a single SCSI card, see the Single SCSI card Raid Test report. Stay tuned to my main Accelerate Your Mac! site for update notices as well as other performance, compatibility and hardware news and sale notices updated several times each day. Future Installments: The next installment will show single card/dual drive array results using all the cards and software listed here. After reading the dual card/dual drive tests in this report, you may want to review the single card/single drive test report if you haven't already. Viewed as a whole, this series will clearly show the best card and driver software in almost every possible scenario. Keep in mind these tests were run on a 180mhz 604E CPU based machine and that scores would improve with a faster CPU card. Since the Adaptec 2940UW has problems working properly with the PowerForce G3 CPU cards, I was forced to use the much slower stock PowerTower Pro 180 CPU card for all the tests.
Feedback
I welcome comments about this review and/or the products involved. I'd like to thank the following companies for providing products for test, saving me precious funds - Adaptec, CharisMac, Conley Corporation, Initio, and a special thanks to MacGurus for providing pairs of many of the cards and Raid ToolKit. If you find this review and site helpful, please let these companies know their efforts are appreciated.

You can follow my preferred path through the report by continuing to the next page, or use the links below to jump to a specific page.

Index of Dual Card/Dual Drive SCSI Raid Comparison Intro \| Cards Tested \| ATTO Benchmark \| FWB Benchtest \| Time Drive \| MacBench = Results Summary = \| Go to Single SCSI Card Raid Tests \| Go to Single Card/Drive Tests - or - Back to WWW.XLR8YOURMAC.COM
Copyright © 1998. Thanks to my better half for tolerating all the time I spend on this site. No part of this sites content or images are to be reproduced or distributed in any form without written permission. Please ask before using quotes or images. All brand or product names mentioned here are properties of their respective companies. Users of this site must read and are bound by the terms/conditions of use