The new data center network is all about efficient scalability, as can be seen in all the recent press about flat networks from the likes of Cisco* and Juniper*. But efficiency must include key attributes such as latency, bandwidth utilization and cost. Latency has become a key metric in data center networks. This is true not only for financial services and clustering applications, but also for applications such as Internet Portal Data Centers (IPDCs), where a simple client web page access may spawn multiple east-west workflows that must complete quickly to maintain a good user experience. Flat networks are really multi-tier chip networks connected in a fat-tree topology. This can cause network hotspots that can be reduced by bandwidth overprovisioning at the cost of extra network equipment and cabling. Switches that provide efficient bandwidth utilization can minimize the size and cost of the network.

 

Intel now has a family of 10GbE/40GbE switch silicon that has been optimized for the data center. Its L3/L4 forwarding latency is about a factor of 2 lower than comparable products, and its high quality built-in PHY technology eliminates the need for external PHY chips in top-of-rack switch server connections, further reducing latency, board area and power. In addition, Intel® Ethernet switches provide very efficient load distribution and flow control mechanisms, optimizing bandwidth utilization and further reducing network cost. Finally, the new Intel® FM6000 switch family offers FlexPipe™ technology that allows features similar to a programmable ASIC while maintaining full processing bandwidth of 1Bpps under all conditions. Now network administrators don’t need to rip out and replace switching gear when standards such as VEPA, VN-Tag, TRILL or SPB change; they can simply download a new microcode image to the FM6000, which will support all of these as well as emerging standards.

 

Check out the complete line of Intel® Ethernet Switches.

dougb

What is Ethernet?

Posted by dougb Nov 18, 2011

     “What is Ethernet?” is a question that is asked almost six and half million times a year according to Google.  It’s great question, and here’s our effort at explaining things.


     First of all, I’m not going to go into the deep technical stuff.  This is a primer, something to get you familiar with Ethernet.


     Ethernet is a way to let computers talk to each other.  It is the central nervous system of the Internet, and it comes in many forms.  But in its most basic form Ethernet is nothing more than a highway, connecting the points that people want to visit.  Just like the highway, not a lot of people look at the road, focusing only on the destination.


     Here at the Wired blog, all we do is the highway.  We help people build the on-ramps and the off-ramps.  We make the parts of the road - both well-travelled and private roads that are hardly ever seen.  Ethernet uses packets (consider them to be like cars) which travel the Ethernet highway.

     

     And, like a real road, sometimes there are traffic jams, construction that forces detours, and traffic lights.   Commute times vary per day and per traffic, and rarely the cars don’t make it.  The cars/packets are various sizes and many formats.


     Without the highway, commerce, visiting, and day-to-day long distance activities become hard if not impossible.  Same with Ethernet.  Before Ethernet there was “sneakernet”; you had to walk your data around to those people interested in it.

 

     I recently heard that 25 million pictures per day are uploaded to one of the more famous social networking sites.  Imagine if you had to walk all those pictures to everyone who wanted to see them.


     Today Ethernet is just about everywhere.  It’s not just your PC and/or laptop.  Digital video records, Blu-Ray* players, televisions, games systems, and even your household appliances are all going to  want access to the highway. Without Ethernet our world is a very different place, one less rich and less interconnected.  So this is the simplified view of Ethernet,  one we’ll expand on in the months ahead.


     You got here via Ethernet; you’ll head to your next website via Ethernet.  So it has been since the Internet started.  And, given Ethernet’s 30 years of history, whatever comes next, after the Internet, it will still be over Ethernet.

 

Thanks for reading the Intel Wired blog, and safe driving on your Ethernet Highway...

Intel® Ethernet will be at the SC11 show in Seattle next week with some special guests, including NASA*.  Watch our twitter feed @IntelEthernet for reminders and details of the show.  We will be at booth 2121, on the fourth floor, near the 4B area.  If you stop by the booth, tell them Doug sent you!

dougb

Feature comparison per O/S

Posted by dougb Nov 3, 2011

A typical question that I get goes something like this:  “I love feature X but was wondering if was available on (insert O/S name here).”   So here is a cheat sheet on the big guns of both features and O/S.   If it is supported in only one version, it will be in the cell.  It’s going to be an HTML table, so consider yourself format warned.


 

Windows*

Linux*

FreeBSD*

VMware*

Fiber Channel over Ethernet

Yes

Yes

No

Yes

Teaming

Yes via Intel ANS

Yes via Channel Bonding

Yes,   via lagg

Yes

iSCSI

Yes

Yes

Yes

Yes

VMDq

Hyper-V

Yes

No

Yes

SR-IOV

No

KVM

Driver willing, but the kernel is missing

No

Intel® PROSet

Yes

No

No

No

RSS

Yes

Yes

Yes

Yes

IOAT

Yes

Yes

No

No

Multiple Queue Support

Yes

Yes

Yes

Yes

 

Some of the features (like iSCSI) are built into the O/S.  Some (like PROSet) are built into the Intel software.  Some features are only available on some products (like FCoE).


Thanks for using Intel® Ethernet.

 

(updated: tried to fix the cell wall colors, hopefully it worked.)

Filter Blog

By author: By date:
By tag: