Slashdot Mirror

> ... although .. after we've all finally moved onto IPv6 networking, and
> all our home systems (not just well-run geek systems but also all Joe Public's
> PCs running Windows 17) are sitting on publically routable real addresses and
> *not* behind NATs, the situation won't be as comfortable any more.

That effing stupid setup is the brainchild of some braindead internet hippies...

1) If your ISP goes down for maintenace or a "backhoe incident", two machines at home won't be able to communicate.

2) I may have a fast router at home, and 2 PC's, all with gigabit ethernet. But if it goes over my 7 mbit down / 1 mbit up ADSL connection, copying files over will take forever.

3) Copying over a few hundred gigabytes of data from my old PC to a new replacement PC would destroy my monthly bandwidth quota.

4) I do *NOT* want my ISP to know what data I have on my PCs.

The way to go is to use link-local IPV6 addresses for all machines as per http://www.brocade.com/content... e.g. and I quote

> To override a link-local address that is automatically computed for an
> interface with a manually configured address, enter commands such as the following.
>
> device(config)#interface ethernet 3/1
> device(config-if-e1000-3/1)#ipv6 address
> FE80::240:D0FF:FE48:4672 link-local
>
> These commands explicitly configure the link-local address FE80::240:D0FF:FE48:4672 for Ethernet interface 3/1.

And then use a hosts file to give simple aliases like "mom", "dad", "billy", or "sue" to each machine. Bonus points for a DD/WRT variant, or ip6tables ruleset on a Raspberry Pi that consolidates all the internal link-local addresses into one external IPv6 address as far as the outside world is concerned. Repeat after me... IPv6 NAT.

End result is, his connection is being tampered with, and he is not recieving the service he paid for.

True that, and I fully agree. BUT: the article suggests something far more evil than the evidence provided suggests, and that's what annoys me.

And like I said, transparent proxying has been done for a long time, and is actually undergoing a phase of renewed youth thanks to CDN/TIC solutions like PeerApp and this Brocade/Bluecoat solution.

A 24-port 10/100 with 2 port 10Gb will be a killer product when it emerges, is standardised, and cheap enough. Hell, I could use it NOW.

The future is here! 10GBASE-T was standardized over 5 years ago, and fiber variants before that. Every major manufacturer's midrange fixed-config edge switch lineup has a 24/48 port 10/100/1000 switch with dual 10Gb uplinks.

Just a few examples:

http://www.cisco.com/en/US/products/ps6406/index.html
http://www.extremenetworks.com/products/summit-x350.aspx
http://www.brocade.com/products/all/switches/product-details/fastiron-gs-series/index.page
http://h30094.www3.hp.com/product.asp?sku=3981100&mfg_part=J9146A&pagemode=ca

A hundred such disk drfives will set you back $13,000.

You calculated a lot of costs there, but you forgot the most important one:

The cost of your business imploding when a critical app falls over because your one hundred, 7200 RPM SATA spindles will only deliver, on a good day with the wind in their sails, 6,000 IOPS at less than 15 ms per op. Or a bad day when you experience multiple drive failures and your shoes are all squishy due to the piss that ran down your leg because you just noticed that your IOPS just dropped to 1,000/sec, your latency just went up to 100 ms and it's going to take over twenty-four hours to complete just one rebuild.

In the data center a few floors below my desk, we have a dozen database servers that average 6,000 IOPS all day long with peak times requiring 10,000 IOPS for several hours every day. Per server.

That's why it's not a hundred drives. It's a couple thousand. And they're not 1TB+ 7200 RPM SATA. They're 146 GB/300GB/400GB 15,000 RPM FC and SAS. And there's over a hundred GB of NVRAM. And everything is connected via multiple paths through two or more separate fabrics (i.e. at least two of these. Each of which will run you over $100k just to get on the floor. Plus annual support and maintenance costs.)

Regardless, it's not just moving bytes into and out of a hard drive. It's being able to replace failed components... as in every component, things like controllers, front end controllers & ports, back end controllers & ports, batteries, drive back planes, and hard drives. Being able to add physical capacity, new drives, ports, controllers, RAM, etc. It's having off site replication. Having defined RPO & RTO. Off host backups. All without service interruption or degradation.

The good news is is that you might spend $1,000,000 on keeping all that going, but you won't be paying out tens of thousands of dollars a minute in salary and hundreds of thousands of dollars in lost business when the home brew solution explodes. And if, heaven forbid, the worst happens and your data center happens to burn to the ground, your DR site will be able to come online in ten minutes and every committed transaction will be sitting there on the drives ready for your apps, customers and employees.

Their Fibre Channel switches SUCK. I had the most downtime I've ever had in my career recently due to their sucky designs. They massively oversubscribe things and they have way too few hardware B2B credits. Their big switches are an absolute joke. Look at Brocade's DCX vs a 9513, the Brocade can do full 8Gb full duplex bandwidth between every port on a 48 port blade and has 256Gbps of backplane which doesn't even get touched if your storage and hosts are on the same blade because the blades do local switching. By comparison the brand new gen 2 9513 fabric has 96Gbps and no local switching between ASIC's! In the real world even the old 48000 series is better with "only" 64Gbps of backbone due to the local switching. After having been bitten by Cisco it will be a loooong time before I even reconsider them.

p.s.
For a good comparison see this page, yeah I know it's from Brocade but it's all truthful information.

There's a newer power calculator at

http://brocade.com/power

The DCX is under half a watt per Gbit.

The 9513 not 528 x 4Gb ports -- take another look at that web page. There's no local switching on Cisco FC linecards, so with 48Gb per slot, it's really just 528 x 1Gb ports. A port can do 4Gb, but it's stealing bandwidth from neighboring ports because of that 48Gb limit.

Brocade can do more bandwidth per slot -- 64Gb in the 48000 (16 x 4Gb ports per slot), and you can get 4Gb on up to 48 ports per slot if you locally switch the other 32 ports.

The DCX has 256Gb per slot and can do 4Gb on every port without needing to locally switch. It can do 256 ports of 8Gb, any to any, and 384 if you use local switching.

The DCX also has a 1/2 terabit interconnect so you can attach two chassis and get 768 ports talking at a better subscription ratio that the 528 ports in the 9513.

The 9132's run SAN-OS and offer all of the features one would expec, so I think they actuall did it themselves this time.

I dont know what the old McData's were like in terms of power. Brocade was proud of their 1W / Gb power envelope (4w per port). The new 9132's look like they are close to that. For what is worth, brocade has a power calculator for the 9513 vs thier 4800 at http://www.brocade.com/products/directors/power_draw_density_calculator.jsp (the numbers look close to real world, so it is a good start). The new Brocade DCX pulls about the same power. More marketing dirt can be found here: http://www.brocade.com/products/competitive/directors.jsp and http://www.brocade.com/products/competitive/references.jsp (I'm sure cisco has a similar page for brocade equipment)

Of course, only cisco offers 528 4G ports on a single chassis (as long as you cable stuff to avoid oversubscribing the backplane. Again, in theory this can be done but in practice one just needs to be a bit more careful).

And lord help me if I ever manage an environemnt that requires that many san ports. I have enough to do now as it is. :-)

The 9132's run SAN-OS and offer all of the features one would expec, so I think they actuall did it themselves this time.

I dont know what the old McData's were like in terms of power. Brocade was proud of their 1W / Gb power envelope (4w per port). The new 9132's look like they are close to that. For what is worth, brocade has a power calculator for the 9513 vs thier 4800 at http://www.brocade.com/products/directors/power_draw_density_calculator.jsp (the numbers look close to real world, so it is a good start). The new Brocade DCX pulls about the same power. More marketing dirt can be found here: http://www.brocade.com/products/competitive/directors.jsp and http://www.brocade.com/products/competitive/references.jsp (I'm sure cisco has a similar page for brocade equipment)

Of course, only cisco offers 528 4G ports on a single chassis (as long as you cable stuff to avoid oversubscribing the backplane. Again, in theory this can be done but in practice one just needs to be a bit more careful).

And lord help me if I ever manage an environemnt that requires that many san ports. I have enough to do now as it is. :-)

The 9132's run SAN-OS and offer all of the features one would expec, so I think they actuall did it themselves this time.

I dont know what the old McData's were like in terms of power. Brocade was proud of their 1W / Gb power envelope (4w per port). The new 9132's look like they are close to that. For what is worth, brocade has a power calculator for the 9513 vs thier 4800 at http://www.brocade.com/products/directors/power_draw_density_calculator.jsp (the numbers look close to real world, so it is a good start). The new Brocade DCX pulls about the same power. More marketing dirt can be found here: http://www.brocade.com/products/competitive/directors.jsp and http://www.brocade.com/products/competitive/references.jsp (I'm sure cisco has a similar page for brocade equipment)

Of course, only cisco offers 528 4G ports on a single chassis (as long as you cable stuff to avoid oversubscribing the backplane. Again, in theory this can be done but in practice one just needs to be a bit more careful).

And lord help me if I ever manage an environemnt that requires that many san ports. I have enough to do now as it is. :-)

Speaking of WAFS, brocade had a product suite based on an architecture called FAN's (file area networks). Originally it was several cobbled together disparate bits of software and an "appliance" running windows server 2003 - though i believe the components that make up tapestry now look more like they belong together rather then the way they used to look where it was very obvious the products were all from different vendors and had different design paradigms. Take a look though, http://www.brocade.com/products/tapestry.jsp (brocade arent the only ones that do this, so look around).

And if you look here: http://en.wikipedia.org/wiki/File_Area_Network - this is the generic term for most of the technology involved, file area networks.

Assuming your running windows everywhere (which wouldn't be a leap) then its not a bad solution - the on-site box is literally a "branch office in a box" solution that incorporates wafs/distributed locked/etc and runs a version of windows server, which i believe can be a AD server as well. But the point of it all is that the remote side has no real date unto itself (Everything goes back to head office) but can manage everything at a remote site (including such things as printers) as well as being easy to replace (in fact, its supposed to be constructed in such a way that if the branch office box fails, people shouldn't notice, everything just starts going back to "head office" in a seamless way). Supposedly its operates over very small amounts of bandwidth, but i can imagine the first time someone opens a large file being a painful excersize.

Still, ive not seen the product except in demo's, but i have heard good things about it.

It's any company that needs a major amount of storage (tera/peta/exa bytes) either at their main site or even remote sister sites would utilize these products. Companies like EMC and EqualLogic make device frames that house hundreds of regular hard drives and are presented to servers as a local drive despite not being local. Because the storage is in a frame and the frames are intelligent, the frames can do data replication to DR sites without a server ever needing to know about it. I know EMC makes both fibre channel-based and IP-based hardware and, based on the headline of the linked article, EqualLogic makes IP-based (iSCSI) SAN equipment. iSCSI allows a corporation to utilize an existing ethernet/IP network without investing hundreds of thousands in a separate storage area network (i.e. fibre channel, see Brocade Comm. Systems).

SANs enable proper separation of servers and data and come with their own tools. Servers can be setup in a cluster with each node being configured to utilize storage allocated on the SAN but only an active node has the storage mounted. Upon failover the passive node can take over and users don't notice any downtime. The data is independent of the server hosting it. It can get expensive though. EMC sells its DMX Symmetrix frames for about $1.4 million (last time I saw a value) depending on capacity ($1.4M probably gets you about a petabyte but don't quote me on the price; it's been too long to remember accurately).

Brocade would be known by a fair few IT engineers that have dealt SANs and enterprise level servers.

The Firewire cards needed to build a cluster can cost as little as 10% as much as the required FiberChannel hardware

Not to mention the FiberChannel switch. The Brocade fiber switch we use to tie our three SGI Origins to our SAN's storage RAID was over CA$12K when we bought it.

What about InfiniBand, which all the major PC hardware design people seem to be involved with? This takes a "switched fabric" approach to linking function blocks together, via Switches (which is where Brocade hopes to be the next Cisco). Need more CPU power or more memory? Hot-plug a module into the Infiniband Switch. Version 1.0 of the spec. is available for download at the site, for those interested.

The successor to IDE is already on the way: Serial ATA. Reportedly, PC makers like it because the thin cables allow them to build smaller systems with better cooling. V1.0 is not going to be much faster than UltraATA/100, but they say there's room for growth there.

Plus, you can have fibre channel (not fiber) hard drives right now, from Seagate (example), IBM (example), etc., and the big storage guys are heading that way too. Fibre Channel doesn't always mean Optical - these drives use a 40-pin "copper" connection, which can be a cable or a backplane (for hot-plugging). The SCSI-3 protocol is carried over the Fibre Channel interface, meaning that with a FC driver loaded, the drives look like SCSI devices.

Anyone see a trend here? It's the end of the parallel interface in all its forms, much as USB and FireWire are replacing the humble parallel port...