Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
You will have to be careful with the Intel gear. It works ok for most deployments but with my setup I had problems. I deployed first with 520SRs (on two Dell R710) and found that they enter a mode where they latch/lock up. The would respond to pings but not send any data. This caused all sorts of issues. I had to use Broadcom instead, this was a real pain since we bought 4 Intel cards, now having to buy 4 Broadcom ones. Most of my clients use Broadcom and were working fine at 10G for the past 4 years. DRBD for my setup supports 2 volumes that can be read/written to at a rate of 640Mbytes/sec each via network bonding. That is with roundtrip commit (protocol C). James -----Original Message----- From: drbd-user-bounces at lists.linbit.com [mailto:drbd-user-bounces at lists.linbit.com] On Behalf Of Shaun Thomas Sent: Friday, January 11, 2013 9:53 AM To: Andy Dills Cc: drbd-user at lists.linbit.com Subject: Re: [DRBD-user] Poor fsync performance On 01/11/2013 08:10 AM, Andy Dills wrote: > Any positive recommendations on 10Gb NICs? Intel is something of a gold standard in that area. We use them exclusively. Stay far, far away from anything with a Broadcom chip. A major benefit is that the 10Gb cards usually have so much bandwidth, you can drop all of your other 1Gb interfaces and reduce cabling slightly. We *were* using dual bonded onboard 1Gb for regular traffic, and 10Gb for DRBD only over a crossover, but found we lost nothing by just doing everything over the 10Gb. Good luck! -- Shaun Thomas OptionsHouse | 141 W. Jackson Blvd. | Suite 500 | Chicago IL, 60604 312-676-8870 sthomas at optionshouse.com ______________________________________________ See http://www.peak6.com/email_disclaimer/ for terms and conditions related to this email _______________________________________________ drbd-user mailing list drbd-user at lists.linbit.com http://lists.linbit.com/mailman/listinfo/drbd-user