Showing posts with label IPv6. Show all posts
Showing posts with label IPv6. Show all posts

Thursday, October 21, 2010

One Giant Step for Interop, Two Small Steps for IPv6 in the Data Center

By David Gross

Interop came out with a press release at its own show yesterday saying that it is giving back virtually all of the 45.x.x.x block of IPv4 addresses that have been allocated to it since 1995.   Yes, this one trade event has set 16 million IP addresses free, or nearly one half of one percent of all IPv4 addresses available for public use.   I hope Hurricane Electric adjusts the IPv4 depletion clock on their website.

Nonetheless, the news on IPv6 keeps rolling, with NetLogic (NETL) announcing a new TCAM chip, or as their marketing department prefers to call it, a "knowledge-based processor".  Underneath the glossy product category name, this new memory chip, the NL10k, supports up to 1.6 billion decisions per second for IPv6 routing, which is an important development for core routers, but not for the $1,000 per port Top-of-Rack switches we've been hearing about over the last week.  Part of the reason those products are so cheap is that they don't give you too many options over where to send data when the cables attached them only go for 10 meters.   So they don't need fancy memory chips or sophisticated operating systems.

The NetLogic chip can store up to 256,000 IPv6 addresses, and 1 million IPv4 addresses.   TCAMs, or Ternary Content Address Memories, work by adding a third state of "x" to the normal 0 and 1 binary combination.   This "x" is needed to block out the host address in the routing lookup, which greatly reduces the time needed to forward the packet compared to a binary CAM, and dramatically more so than if the addresses were stored in conventional RAM.

The use of TCAMs grew dramatically in the middle of last decade with increasing shipments of 10 gigabit ports on data networking devices, which needed the chips to maintain performance at high line rates while searching through routing tables.   NetLogic was the leading supplier in this segment, and Wall Street fell in love with its stock in 2005.   IPv6 presents another networking development that could accelerate NetLogic's growth, although I still think it will progress far more slowly than people panicking about IPv4 address depletion expect.

Wednesday, October 13, 2010

IPv6 - For Real This Time?

By David Gross

For the last 10 years, we've heard that the Internet will run out of IPv4 addresses in 18 months.   All sorts of panicky statements have come out of the IETF and other engineering groups, but the Internet hasn't fallen apart, and everyone has just kept going with an old-fashioned "dotted quad".

Any business that's ordered any kind of bandwidth service has seen how precious IPv4 addresses are.   Getting your numbers from your service provider is like being handed some carefully guarded secret.   But as scarce as IPv4 addresses have become, the threat of running out of addresses has done little to provoke a shift to the long ago standardized, but never really implemented IPv6, which with 2 to the 128th addresses, has sufficient space to tag just about every physical (or virtual object) on the planet with an IPv6 identifier.

There are a number of reasons behind IPv4's persistence, but one of the most important has been the use of MAC addressing, which adds another 16 bits to a device's address, or 65,536x the number of addressable devices compared to IPv4.    Additionally, the Ethernet switches which pass traffic based on MAC address have been shipping in greater numbers relative to routers.   And they're generally cheaper.   Moreover, XenServer and VMWare let you tag each virtual machine with its own MAC ID, which is one factor that has prevented virtualized servers from blowing through IPv4's 32-bit address space.

But after years of hype and literally going nowhere in the market, IPv6 is seeing new life.   A major source of this resurgence is the Federal Government.   Now before anyone claims the government can make anything happen, I have to point out some of the odd things they've done that private industry never adopted, which in turn forced them to retreat and follow the private sector's lead.   One was GOSIP, a TCP/IP competitor that gave us the seven layer OSI model we now reference all the time, but very few actual implementations.   The government planned to make GOSIP compliance a requirement in the early 90s.   But after all kinds of wasted dollars on expensive routers and software, and the growth of the Internet, it got just a little bit overwhelmed by the growing use of TCP/IP. 

The Federal Goverment also was one of the few places in the mid-to-late 90s where you could see ATM-to-the-Desktop NICs actually installed on people's computers.   Most corporations, and certainly their shareholders, preferred their IT managers to use 100 Meg Fast Ethernet adapers, which offered 4x the bandwidth for $300 less than the $300 price of ATM-to-the-Desktop NICs. 

So I get just a little bit skeptical when the government starts issuing networking mandates, which the U.S. CIO has done recently, by mandating that public-facing Federal websites deploy IPv6 by September 30, 2012, and that Federal agencies upgrade their internal networks and servers to IPv6 by September 30, 2014.   Reminds me a lot of the August 1990 mandate for GOSIP compliance.

One of the big differences between IPv6 and GOSIP is that there are actual profit-making companies getting behind IPv6.  Hurricane Electric, a.k.a HE.net, is bringing IPv6-based IP transit into Telx, Equinix, and co-lo centers across the country, and just expanded its own co-lo facility in the Bay Area.   Verisign declared this week that it plans to keep its lead in IPv6, with feature parity between IPv4 DNS services and IPv6 DNS services, according to this article in Network World.   But in the same story, the company admits IPv4 DNS queries are still outpacing IPv6 DNS queries by a ratio of approximately 1100 to 1.   I wouldn't be surprised if the 1 originated from a cubicle in a government building.