& cplSiteName &

Two Gigabits, One Vendor

Light Reading
1/5/2002
50%
50%

A new technology called 2-Gbit/s Fibre Channel emerged in 2001, promising speeds fast enough to deliver the Holy Grail of storage-area networking – real-time access to offsite data.

Vendors of SAN equipment eagerly lined up to support the new spec. It sounded promising, so Light Reading commissioned the independent lab Network Test Inc. to evaluate 2-Gbit/s fabric switches. Test equipment maker Spirent Communications contributed the testing kit and engineering support.

Coming in first: QLogic Corp.

Coming in last: QLogic Corp.

There's a Logic at work here. No fewer than five switch makers said they’d have equipment ready by year’s end. But when vendors actually had to submit equipment for testing, only one – QLogic Corp. (Nasdaq: QLGC) was willing to go through with it.

Some vendors said they just didn’t have the equipment or the people. One company – Vixel Corp. (Nasdaq: VIXL) – entered the test and then withdrew. The dominant player – Brocade Communications Systems Inc. (Nasdaq: BRCD) – simply said no.

Not a pretty picture, by any means. The most likely explanation is that – as with any new technology – there are still significant bugs to work out in these time-to-market products.

On the positive side, the fact that QLogic submitted its product meant that we were able to go ahead and perform the first ever public test of 2-Gbit/s FC technology. Not simply the first 2-Gbit/s switch test, mind you. Nor just the first 2-Gbit/s fabric test. It was first ever public 2 Gbit/s test, period.

What did we learn? The good news is that QLogic’s SANbox2 switches worked very, very well. The switches really do run at 2-Gbit/s or close to it. Better yet, the boxes posted the lowest latency numbers ever recorded in any test conducted by Network Test. About the only downside was some elevation in latency in our more advanced tests.

Table 1: Table for One

QLogic Corp. Aliso Viejo, California
949-389-6000 www.qlogic.com
Product and version tested SANbox2, version 1.1
Maximum ports per switch/per cascade 16/3,348
Switch architecture Cross-connect
Topologies supported Point-to-point, arbitrated loop, switched fabric, cascade, mesh
Traffic classes supported 2, 3
Management methods Telnet, serial, in- and out-of-band SNMP, Java-based Web application
Management software support Linux, Solaris, Windows 9x/Me/NT/2000
Redundant features Power supplies, fans
Price as tested $17,500


It’s a promising start, at least for QLogic. Now it’s up to the other vendors to show they, too, can deliver working 2-Gbit/s solutions.

Read ahead for details:

Excuses, Excuses
Fibre Channel in a Nutshell
First Things First
Size Does Matter
Delay Tactics
Switch Jitters
Failsafe
Hold That Line
All for One

(5)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
dnewman
50%
50%
dnewman,
User Rank: Light Beer
12/4/2012 | 11:02:27 PM
re: Two Gigabits, One Vendor

Two things:

1. In preparing for Light Reading/Byte & Switch 2-Gbit/s SAN test, I was struck by the lack of public data on traffic characteristics in production SAN settings. This lack of traffic analysis makes it harder to develop meaningful synthetic benchmarks.

Contrast this with IP traffic, where organizations like CAIDA (www.caida.org) and Merit (www.merit.edu) offer rich Internet traffic sources, along with rigorous analysis of the traffic.

A repository of traces would be a service to the storage industry. Any interest?

2. I've never been comfortable with the anonymous format of these message boards, and effective now I'm not going to respond to anonymous posts.

At the bottom of this message is my name, email address, and a disclosure statement. I'll be happy to respond to any query -- provided you're willing to identify yourself as well. Thanks.

David Newman
Network Test
dnewman@networktest.com

Disclosure: No position in QLGC or any other vendor invited to participate in the Light Reading/Byte & Switch SAN test.


jacobi
50%
50%
jacobi,
User Rank: Light Beer
12/4/2012 | 10:58:10 PM
re: Two Gigabits, One Vendor
Mr. Newman,
A very interesting benchmark. Especially, the exceptional low-latency, low-jitter merits are very impressive, and seem to test the testing equipment nicely.
Could you please elaborate about the abovementioned performance metrics, and detail in which offered load to the switch were the merits being recorded?

Shimshon Jacobi
TeraCross
jacobi@teracross.com


dnewman
50%
50%
dnewman,
User Rank: Light Beer
12/4/2012 | 10:57:12 PM
re: Two Gigabits, One Vendor
Hi Shimson (if I may),

"Especially, the exceptional low-latency, low-jitter merits are very impressive, and seem to test the testing equipment nicely."

Impressive indeed. The published resolution of the test instrument is +/- 100 nanoseconds.

We really should start including the accuracy of test instruments when we publish results. A few weeks ago this was suggested on the IETF's benchmarking mailing list, and I completely agree.

"Could you please elaborate about the abovementioned performance metrics, and detail in which offered load to the switch were the merits being recorded?"

RFC 2544, section 26.2, specifies that latency (and therefore jitter, which is derived) should be measured at the throughput rate. That's standard operating procedure in all our tests, including this one.

There have been suggestions that latency should be measured at rates below the throughput level as well. I agree; there is a very interesting "hockey stick" effect with latency, especially as the offered load approaches the throughput rate. The shape of the hockey stick is different for every device, too.

Summarizing load and latency for multiple offered loads, multiple frame sizes, and multiple vendors may pose an interesting presentation challenge. Maybe Light Reading will start mailing out 3D glasses to its readers. :)

One measurement that is NOT valid is latency in the presence of frame loss. This is really a measure of a device's queue depth (buffering) rather than its basic delay characteristics.

There isn't yet one standard definition of jitter. The definition we use here is the absolute value of the difference in arrival times between any two frames. Average jitter is simply the mean of all these observations.

Regards,
David Newman
Network Test
dnewman@networktest.com

Disclosure: Neither long nor short any FC switch vendor at the time of testing.








wolves
50%
50%
wolves,
User Rank: Light Beer
12/4/2012 | 10:50:21 PM
re: Two Gigabits, One Vendor
thanks for your information.
maybe next time you can test a SAN solution(not only a switch) in emulated production enviroment.

roger
yzy@ztek.com.cn
dnewman
50%
50%
dnewman,
User Rank: Light Beer
12/4/2012 | 10:50:06 PM
re: Two Gigabits, One Vendor
Hi Roger,

"maybe next time you can test a SAN solution(not only a switch) in emulated production enviroment."

Thanks for your message. Testing end-to-end is a great idea. One of the challenges we'll face is getting the emulation right; for example, what are meaningful utilization levels, frame size distributions, and so forth.

As I've said before, the lack of freely available FC traffic analysis is a big hindrance. I'd be glad to take part in an effort to make more FC traffic info available, similar to the way that, say, CAIDA publishes Internet traffic analysis.

Regards,
David Newman
Network Test
dnewman@networktest.com

Disclosure: Neither long nor short any FC vendor at test time.
More Blogs from Tests
Content Switch Test Highlights:
* 3 market-ready switches
* Great scaleability
* Edge goes to Netscaler
Edge Router Test Highlights:
* Great VPN Delivery
* Great Scaleability
* Slight edge to Laurel
Multiservice Switch Test Highlights:
*Alcatel Stars
*Terrific ATM
*Good MPLS
*Other Vendors MIA
VPN Test Highlights:
* Max t-put: 965.8 Mbit/s
* Min latency: 19.2 ms
* Thumbs: two, up
Step 1 in Light Reading's ultimate test of optical grooming switches
* What carriers want
* What's behind the test
* Potential players
Featured Video
From The Founder
John Chambers is still as passionate about business and innovation as he ever was at Cisco, finds Steve Saunders.
Flash Poll
Upcoming Live Events
September 12, 2018, Los Angeles, CA
September 24-26, 2018, Westin Westminster, Denver
October 9, 2018, The Westin Times Square, New York
October 23, 2018, Georgia World Congress Centre, Atlanta, GA
November 6, 2018, London, United Kingdom
November 7-8, 2018, London, United Kingdom
November 8, 2018, The Montcalm by Marble Arch, London
November 15, 2018, The Westin Times Square, New York
December 4-6, 2018, Lisbon, Portugal
All Upcoming Live Events
Hot Topics
Telecom Jargonosaurus Part 1: Repeat Offenders
Iain Morris, News Editor, 7/13/2018
AT&T's Stankey Serves Up a Stinker at HBO
Iain Morris, News Editor, 7/10/2018
Broadcom Buys CA – Huh?
Mitch Wagner, Executive Editor, Light Reading, 7/11/2018
Verizon Taps Malady as Acting CTO
Dan Jones, Mobile Editor, 7/12/2018
FCC's Rosenworcel: US 'Falling Behind' on 5G
Iain Morris, News Editor, 7/13/2018
Live Digital Audio

A CSP's digital transformation involves so much more than technology. Crucial – and often most challenging – is the cultural transformation that goes along with it. As Sigma's Chief Technology Officer, Catherine Michel has extensive experience with technology as she leads the company's entire product portfolio and strategy. But she's also no stranger to merging technology and culture, having taken a company — Tribold — from inception to acquisition (by Sigma in 2013), and she continues to advise service providers on how to drive their own transformations. This impressive female leader and vocal advocate for other women in the industry will join Women in Comms for a live radio show to discuss all things digital transformation, including the cultural transformation that goes along with it.

Like Us on Facebook
Twitter Feed