Learn what Russ Fellows Doesn’t Know

So how’s this for a condescending tweet?

It’s from Russ Fellows, author of the infamous FCoE “study” (which has been widely debunked for its many hilarious errors):

Interesting article (check it out). But the sad/amusing irony is that he’s wrong. How is he wrong? Here’s what Russ Fellows doesn’t know about storage:

1, 2, 4, and 8 Gbit Fibre Channel (as he points out) uses 8/10 bit encoding. That means about a 20% of the bandwidth available was lost due to encoding overhead (as Russ pointed out). That’s why 8 Gbit Fibre Channel only provides 800 MB/s of connectivity, even though 8,000 Megabits per second equates to 1,000 Megabytes per second (8000 Megabits / (8 bits per byte) = 1,000 Megabytes).

With this overhead in mind, Fibre Channel was designed to give 100 MB/s for every Gigabit of speed. It never increased the baud rate to make up for the overhead.

Ethernet, on the other hand, did increase the baud rate to make up for the overhead. Gigabit Ethernet uses the same 8/10 bit encoding, but they kicked the baud rate up to 1.25 gigabaud to make up the differences. As such, Gigabit Ethernet provides true 1 gigabit of throughput, or 125 Megabytes per second.

10 Gigabit Ethernet moved to 64/66 encoding, and kept to the approach of not letting the overhead impact throughput. 10 Gigabit Ethernet then provides 1250 Megabytes per second of throughput. The baud rate is 10.3125, giving true 10 Gigabit per second of data.

When Fibre Channel moved to the more efficient 64/66 bit encoding, rather than change the 100 MB/s per gigabit to 125 MB/s (which you get with all Ethernet speeds), they left the ratio (1 Gigabit to 100 MB/s) the same. Thus, every Gigabit = 100 MB/s, just like in previous speeds (1/2/4/8 FC). So while 16 Gbit Fibre Channel provides 1600 MB/s of throughput, the baud rate is actually only 14 gigabaud, and not true 16 Gbit. And don’t take my word for it, check out page 7 of Scott Shimomura‘s (of Brocade) presentation at the SPDE conference.

  • 1 Gbit Fibre Channel = 100 MB/s
  • 1 Gbit Ethernet = 125 MB/s
  • 2 Gbit Fibre Channel = 200 MB/s
  • 4 Gbit Fibre Channel = 400 MB/s
  • 8 Gbit Fibre Channel = 800 MB/s
  • 10 Gbit Ethernet/FCoE = 1250 MB/s
  • 16 Gbit Fibre Channel = 1600 MB/s

10 Gigabit Ethernet provides 1250 MB/s, providing true 10 Gigabit Ethernet, and not putting the slight overhead into the equation. So while 10 Gigabit Ethernet is true 10 Gigabit, 16 Gigabit Fibre Channel is actually 14 Gigabit Fibre Channel (14.025, to be exact).

And that’s what Russ Fellows doesn’t know. His entire article is based on a false premise: Thinking that the move to 64/66 makes 16 Gbit pass more than twice as much traffic as 8 Gbit. But it’s not. He says that with 8 Gbit FC, 1+1 = 1.6 (when compared to 16 Gbit FC), which is factually incorrect for the reasons I’ve just explained. Yes, 64/66 bit encoding is more efficient. But they dropped the baud rate, negating the efficiency gains

8 Gigabit Fibre Channel provides 800 Megabytes per second of data transfer. 16 Gigabit Fibre Channel (really 14 Gigabit Fibre Channel) provides 1600 Megabytes per second of data transfer. 800 + 800 = 1600.

Sorry Russ, 1+1 really does equal 2. Even in Fibre Channel.


OTV AEDs Are Like Highlanders

While prepping for CCIE Data Center and playing around with a lab environment, I ran into a problem I’d like to share.

I was setting up a basic OTV setup with three VDCs running OTV, connecting to a core VDC running the multicast core (which is a lot easier than it sounds). I’m running it in a lab environment we have at Firefly, but I’m not going by our normal lab guide, instead making it up as I go along in order to save some time, and make sure I can stand up OTV without a lab guide.

Each VDC will set up an adjacency with the other two, with the core VDC providing unicast and multicast connectivity.  That part was pretty easy to setup (even the multicast part, which had previously freaked me the shit out). Each VDC would be its own site, so no redundant AEDs.

On each OTV VDC, I setup the following as per my pre-OTV checklist:

  • Bi-directional IPv4 unicast connectivity to each join interface (I used a single OSPF area)
  • MTU of 9216 end-to-end (easy since OTV requires M line cards, and it’s just an MTU command on the interface)
  • An OTV site VLAN which requires:
    • That the VLAN is configured on the VDC
    • That the VLAN is active on a physical port that is up
  • Multicast configuration
    • IP pim sparse-mode configuration on every interface, end-to-end
    • IP igmp version 3 on every interface end-to-end
    • Rendezvous point (RP) configured on the loopback address of the core VDC (I used the bidir tag)

So I got all that configured and then configured the OTV setup. Very basic:

feature otv

otv site-vlan 10

interface Overlay1
  otv join-interface Ethernet1/2
  otv control-group
  otv data-group
  otv extend-vlan 100
  no shutdown
otv site-identifier 0000.0000.0002

ip pim rp-address group-list
ip pim ssm range

The only difference between the three OTV VDC configurations was the site-identifier and the join interface. Everything else was identical, pretty easy configuration. But… it didn’t work. Shit. Time for some show commands:

N7K-11-vdc-2# show otv adjacency
Overlay Adjacency database
Overlay-Interface Overlay1 :
Hostname System-ID Dest Addr Up Time State
VDC-3 18ef.63e9.5d43 01:36:52 UP
vdc-4 18ef.63e9.5d44 01:41:57 UP

OK, so the adjacencies are built. I’ve at least got IP4 unicast and multicast going on. How about “show otv”?

N7K-11-vdc-2# show otv

OTV Overlay Information
Site Identifier 0000.0000.0002

Overlay interface Overlay1

 VPN name : Overlay1
 VPN state : UP
 Extended vlans : 100 (Total:1)
 Control group :
 Data group range(s) :
 Join interface(s) : Eth1/2 (
 Site vlan : 11 (up)
 AED-Capable : No (Site-ID mismatch)
 Capability : Multicast-Reachable

Site-ID mismatch? What the shit? They’re supposed to mismatch. I try another command:

N7K-11-vdc-2# show otv site

Dual Adjacency State Description
 Full - Both site and overlay adjacency up
 Partial - Either site/overlay adjacency down
 Down - Both adjacencies are down (Neighbor is down/unreachable)
 (!) - Site-ID mismatch detected

Local Edge Device Information:
 Hostname vdc-2
 System-ID 18ef.63e9.5d42
 Site-Identifier 0000.0000.0002
 Site-VLAN 11 State is Up

Site Information for Overlay1:

Local device is not AED-Capable (Site-ID mismatch)
Neighbor Edge Devices in Site: 1

Hostname System-ID Adjacency- Adjacency- AED-

 State Uptime Capable

VDC-3 18ef.63e9.5d43 Partial (!) 00:17:39 Yes

Now this show command confused me for a while. I was trying to figure out the Site-ID mismatch. I was also wondering why I could see VDC-3 but couldn’t see VDC-4. Then it dawned on me (after am embarrassing amount of time) I’m not supposed to. I’m not supposed to see VDC-3, either. The “show site” command is only looking at the local area. For my configuration, I shouldn’t see any other VDCs with “show otv site”.

This means that there’s some type of Layer 2 connectivity between the different sites. VDC-3 and VDC-4 both somehow see each other as Layer 2 adjacent. That shouldn’t happen if they’re supposedly on remote sites. This is a lab environment, so there’s some sort of Layer 2 connectivity for the Site-VLAN that I need to kill.

OTV edge devices are like highlanders, if there’s Layer 2 adjacency, they sense each other.


“I could sense you by your VLAN”

It probably happened on the interface that I assigned the site-VLAN to as an access port. A VLAN will not show “active” unless you have an active physical link (interface VLANs don’t count).

So I went through and re-configured the site VLAN. Instead of VLAN 10 (which was probably active on the other ends of those interfaces somehow) I created new VLANs, and used a unique VLAN for each VDC. The site-VLANs do not need to be identical between sites. I put the VLAN on a physical link that was up, and voila.

In the real world, you probably won’t run into this. However, it’s possible if there are other Layer 2 interconnects going on in your data center (perhaps dark fiber) or you’re transitioning from one DCI to another, you may hit this.

Top 5 Reasons The Evaluator Group Screwed Up

It’s been a while since the trainwreck of a “study” commissioned by Brocade and performed by The Evaluator Group,  but it’s still being discussed in various storage circles (and that’s not good news for Brocade). Some pretty much parroted the results, seemingly without reading the actual test. Then got all pissy when confronted about it.  I did a piece on my interpretations of the results, as did Dave Alexander of WWT and J Metz of Cisco. Our mutual conclusion can be best summed up with a single animated GIF.



But since a bit of time has passed, I’ve had time to absorb Dave and J’s opinions, as well as others, I’ve come up with a list of the Top 5 Reasons by The Evaluator Group Screwed Up. This isn’t the complete list, of course, but some of the more glaring problems. Let’s start with #1:

Reason #1: I Have No Idea What I’m Doing

Their hilariously bad conclusion to the higher variance in response times and higher CPU usage was that it was the cause of the software initiators. Except, they didn’t use software initiators. The had actually configured hardware initiators, and didn’t know it. Let that sink in: They’re charged with performing an evaluation, without knowing what they’re doing.

The Cisco UCS VIC 1240 hardware CNA’s were utilized.  Referring to them as software initiators caused some confusion. The Cisco VIC is a hardware initiator and we configured them with virtual HBAs. Evaluator Group has no knowledge of the internal architecture of the VIC or its driver.  Our commentary of the possible cause for higher CPU utilization is our opinion and further analysis would be required to pinpoint the specific root cause.

Of course, it wasn’t the software initiator. They didn’t use a software initiator, but they were so clueless, they didn’t know they’d actually used a hardware initiator. Without knowing how they performed their tests (since they didn’t publish their methodology) it’s purely speculation, but it looks like the problem was caused by congestion (from them architecting the UCS solution incorrectly).

Reason #2: They’re Hilariously Bad At Math.

They claimed FCoE required 50% more cables, based on the fact that there were 50% more cables in the FCoE solution than the FC solution. Which makes sense… except that the FC system had zero Ethernet.

That’s right, in the HP/Fibre Channel solution, each blade had absolutely zero Ethernet connectivity. In the Cisco UCS solution, every blade had full Ethernet and Fibre Channel connectivity.  None. Zilch. Why did they do that? Probably because had they included any network connectivity to the HP system, the cable count would have shifted to FCoE’s favor.  Let me state this again, because it’s astonishingly stupid: They claimed FCoE (which included Ethernet and FC connectivity) required more cables without including any network connectivity for the HP/FC system. 


Also, they made some power/cooling claims, despite the fact that the UCS solution didn’t require a separate FC switch (it’s capable of being a full-fledged Fibre Channel switch by itself), though the HP solution would have required a separate pair of Ethernet switches (which wasn’t included). So yeah, their math is a bit off. Had they done things, you know, correctly, the power, cooling, and cable count would have flipped in favor of FCoE.

Reason #3: UCS is Hard, You Guys!

They whinged about UCS being more difficult to setup. Anytime you’re dealing with unfamiliar technology, it’s natural that it’s going to be more difficult. However, they claimed that they had zero experience with HP as well (seriously, who at Brocade hired these guys?) How easy is UCS? Here is a video done from Amsterdam where a couple of Cisco techs added a new chassis and blade and had it booted up and running ESXi in less than 30 minutes from in the box to booted. Cisco UCS is different than other blade systems, but it’s also very easy (and very quick) to stand up. And keep in mind, the video I linked was done in Amsterdam, so they were probably baked   

Reason #4: It Contradicts Everyone Else’s Results (Especially those that know what they’re doing)

For the past couple of years, VMware and NetApp have been doing performance tests on various storage protocols. Here’s one from a few years ago, which includes (native) 4 and 8 Gbit Fibre Channel, 10 Gbit FCoE, 10 Gbit iSCSI, and 10 Gbit NFS. The conclusion? The protocol doesn’t much matter. They all came out about the same when normalized for bandwidth. The big difference is in the storage backend. At least they published their methodology (I’m looking at you, Evaluator Group). Here’s one from Demartek that shows a mixture of storage protocols saturating 10 Gbit Ethernet. Again, the limitation is only the link speed itself, not the protocol. And again, again, Demartek published their methodology.

Reason #5: How Did They Set Everything Up? Magic!

Most of the time with these commissioned reports, the details of how it’s configured are given so that the results can be reproduced and audited. How did the Evaluator Group set up their environment?


As far as I can tell, magic. There’s several things they could have easily gotten wrong with the UCS setup, and given their mistake about software/hardware initiators, quite likely. They didn’t even mention which storage vendor they used.

So there you have it. A bit of a re-hash, but hey, it was a dumb report. The upside though is that it did provide me with some entertainment.

Fibre Channel: The Heart of New SDN Solutions

From Juniper to Cisco to VMware, companies are spouting up new SDN solutions. Juniper’s Contrail, Cisco’s ACI, VMware’s NSX, and more are all vying to be the next generation of data center networking. What is surprising, however, is what’s at the heart of these new technologies.

Is it VXLAN, NVGRE, Openflow? Nope. It’s Fibre Channel.


If you think about it, it makes sense. Fibre Channel has been doing fabrics since before we ever called Ethernet fabrics, well, fabrics. And this isn’t the first time that Fibre Channel has shown up in unusual places. There’s a version of Fibre Channel that runs inside certain airplanes, including jet fighters like the F-22.


Keep the skies safe from FCoE (sponsored by the Evaluator Group)

New generation of switches have been capable of Data Center Bridging (DCB), which enables Fibre Channel over Ethernet. These chips are also capable of doing native Fibre Channel So rather than build complicated VPLS fabrics or routed networks, various data center switching companies are leveraging the inherent Fibre Channel capabilities of the merchant silicon and building Fibre Channel-based underlay networks to support an IP-based overlay.

Buffer-to-buffer (B2B) credit system and losslessness of Fibre Channel, plus the new 32/128 Gigabit interfaces with the newest Fibre Channel standard are all being leveraged for these underlays. I find it surprising that so many companies are adopting this, you’d think it’d be just Brocade. But Cisco, Arista (who notoriously shunned FCoE) and Juniper are all on board with new or announced SDN offerings that are based mostly or in part on Fibre Channel.

However, most of the switches from various vendors are primarily Ethernet today, so the 10/40 Gigabit interfaces can run FCoE until more switches are available with native FC interfaces. Of course, these switches will still be required to have a number of native Ethernet ports in order to connect to border networks that aren’t part of the overlay network, so there will be still a need for Ethernet. But it seems the market has spoken, and they want Fibre Channel.


Hey, Remember vTax?

Hey, remember vTax/vRAM? It’s dead and gone, but with 6 Terabyte of RAM servers now available, imagine what could have been (your insanely high licensing costs).

Set the wayback machine to 2011, when VMware introduced vSphere version 5. It had some really great enhancements over version 4, but no one was talking about the new features. Instead, they talked about the new licensing scheme and how much it sucked.


While some defended VMware’s position, most were critical, and my own opinion… let’s just say I’ve likely ensured I’ll never be employed by VMware. Fortunately, VMware came to their senses and realized what a bone-headed, dumbass move that vRAM/vTax was, and repealed the vRAM licensing one year later in 2012. So while I don’t want to beat a dead horse (which, seriously, disturbing idiom), I do think it’s worth looking back for just a moment to see how monumentally stupid that licensing scheme was for customers, and serve as a lesson in the economies of scaling for the x86 platform, and as a reminder about the ramifications of CapEx versus OpEx-oriented licensing.

Why am I thinking about this almost 2 years after they got rid of vRAM/vTax? I’ve been reading up on the newly released Intel’s E7 v2 processors, and among the updates to Intel’s high-end server chip is the ability to have 24 DIMMs per socket (the previous limit was 12) and the support of 64 GB DIMMs. This means that a 4-way motherboard (which you can order now from Cisco, HP, and others) can support up to 6 TB of RAM, using 96 DIMM slots and 64 GB DIMMs. And you’d get up to 60 cores/120 threads with that much RAM, too.

And I remembered one (of many) aspects about vRAM that I found horrible, which was just how quickly costs could spiral out of control, because server vendors (which weren’t happy about vRAM either) are cramming more and more RAM into these servers.

The original vRAM licensing with vSphere 5 was that for every socket you paid for, you were entitled to/limited to 48 GB of vRAM with Enterprise Plus. To be fair the licensing scheme didn’t care how much physical RAM (pRAM) you had, only how much of the RAM was consumed by spun-up VMs (vRAM). With vSphere 4 (and the current vSphere licensing, thankfully), RAM had been essentially free: you only paid per socket. You could use as much RAM as you could cram into a server.  But with the vRAM licensing, if you had a dual-socket motherboard with 256 GB of RAM you would have to buy 6 licenses instead of 2. At the time, 256 GB servers weren’t super common, but you could order them from the various server vendors (IBM, Cisco, HP, etc.). So with vSphere 4, you would have paid about $7,000 to license that system. With vSphere 5, assuming you used all the RAM, you’d pay about $21,000 to license the system, a 300% increase in licensing costs. And that was day 1.

Now lets see how much it would cost to license a system with 6 TB of RAM. If you use the original vRAM allotment amounts from 2011, each socket granted you 48 GB of vRAM with Enterprise Plus (they did up the allotments after all of the backlash, but that ammended vRAM licensing model was so convoluted you literally needed an application to tell you how much you owed). That means to use all 6 TB (and after all, why would you buy that much RAM and not use it), you would need 128 socket licences, which would have cost $448,000 in licensing. A cluster of 4 vSphere hosts would cost just shy of $2 million to license. With current, non-insane licensing, the same 4-way 6 TB server costs a whopping $14,000. That’s a 32,000% price differential. 

Again, this is all old news. VMware got rid of the awful licensing, so it’s a non-issue now. But still important to remember what almost happened, and how insane licensing costs could have been just a few years later.


My graph from 2011 was pretty accurate.

Rumor has it VMware is having trouble getting customers to go for OpEx-oriented licensing for NSX. While VMware hasn’t publicly discussed licensing, it’s a poorly kept secret that VMware is looking to charge for NSX on a per VM, per month basis. The number I’d been hearing is $10 per month ($120 per year), per VM. I’ve also heard as high as $40, and as low as $5. But whatever the numbers are, VMware is gunning for OpEx-oriented licensing, and no one seems to be biting. And it’s not the technology, everyone agrees that it’s pretty nifty, but the licensing terms are a concern. NSX is viewed as network infrastructure, and in that world we’re used to CapEx-oriented licensing. Some of VMware’s products are OpEx-oriented, but their attempt to switch vSphere over to OpEx was disastrous. And it seems to be the same for NSX.

Changing Data Center Workloads

Networking-wise, I’ve spent my career in the data center. I’m pursuing the CCIE Data Center. I study virtualization, storage, and DC networking. Right now, the landscape in the network is constantly changing, as it has been for the past 15 years. However, with SDN, merchant silicon, overlay networks, and more, the rate of change in a data center network seems to be accelerating.


Things are changing fast in data center networking. You get the picture

Whenever you have a high rate of change, you’ll end up with a lot of questions such as:

  • Where does this leave the current equipment I’ve got now?
  • Would SDN solve any of the issues I’m having?
  • What the hell is SDN, anyway?
  • I’m buying vendor X, should I look into vendor Y?
  • What features should I be looking for in a data center networking device?

I’m not actually going to answer any of these questions in this article. I am, however, going to profile some of the common workloads that you find in data centers currently. Your data center may have one, a few, or all of these workloads. It may not have any of them. Your data center may have one of the workloads listed, but my description and/or requirements is way off. All certainly possible. These are generalizations, and with all generalizations your mileage may vary. With that disclaimer out of the way, strap in. Let’s go for a ride.

Traditional Virtualization

It’s interesting to say that something which only exploded into the data center in a big way in about 2008 as now being “traditional”, rather than “new-fangled”. But that’s the situation we have here. Traditional virtualization workload  is centered primarily around VMware vSphere. There are other traditional virtualization products of course, such as Red Hat’s RHEV, Xen, and Microsoft Hyper-V, but VMware has the largest market share for this by far.

  • Latency is not a huge concern (30 usecs not a big deal)
  • Layer 2 adjacencies are mandatory (required for vMotion)
  • Large Layer 2 domains (thousands of hosts layer 2 adjacent)
  • Converged infrastructures (storage and data running on the same wires, FCoE, iSCSI, NFS, SMB3, etc.)
  • Buffer requirements aren’t typically super high. Bursting isn’t much of an issue for most workloads of this type.
  • Fibre Channel is often the storage protocol of choice, along with NFS and some iSCSI as well

Cisco has been especially successful in this realm with the Nexus line because of vPC, FabricPath, OTV and (to a much lesser extent) LISP, as they address some of the challenges with workload mobility (though not all of them, such as the speed of light). Arista, Juniper, and many others also compete in this particular realm, but Cisco is the market leader.

With the multi-pathing Layer 2 technologies such as SPB, TRILL, Cisco FabricPath, and Brocade VCS (the latter two are based on TRILL), you can build multi-spine leaf/spine networks/CLOS networks that you can’t with spanning-tree based networks, even with MLAG.

This type of network is what I typically see in data centers today. However, there is a shift towards Layer 3 networks and cloud workloads over traditional virtualization, so it will be interested to see how long traditional virtualization lasts.


VDI (Virtual Desktops) are a workload with the exact same requirements as traditional virtualization, with one main difference: The storage requirements are much, much higher.

  • Latency is not as important (most DC-grade switches would qualify), especially since latency is measured in milliseconds for remote desktop users
  • Layer 2 adjacencies are mandatory (required for vMotion)
  • Large Layer 2 domains
  • Converged infrastructures
  • Buffer requirements aren’t typically very high
  • High-end storage backends. All about the IOPs, y’all

For storage here, IOPs are the biggest concern. VDI eats IOPs like candy.

Legacy Workloads

This is the old, old school. And by old school, I mean late 90s, early 2000s. Before virtualization changed the landscape. There’s still quite a few crusty old servers, with uptimes measured in years, running long-abandoned applications. The problem is, these types of applications are usually running something mission critical and/or significant revenue generating. Organizations just haven’t found a way out of it yet. And hey, they’re working right now. Often running on proprietary Unix systems, they couldn’t or wouldn’t be migrated to a virtualized environment (where it would be much easier to deal with).

The hardware still works, so why change something that works? Because it would be tough to find more. It’s also probably out of vendor-supported service.

  • Latency? Who cares. Is it less than 1 second? Good enough.
  • Layer 2 adjacencies, if even required, are typically very small, typically just needed for the local clustering application (which is usually just stink-out-loud awful)
  • 100 megabit and gigabit Ethernet typically. 10 Gigabit? That’s science-fiction talk!
  • Buffers? You mean like, what shines the floor?


My own personal opinion is that this is the only place where Cisco Catalyst switches belong in a data center, and even then only because they’re already there. If you’re going with Cisco, I think everything else (and everything new) in the DC should be Nexus.

Cloud Workloads (Private Cloud)

If you look at a cloud workload, it looks very similar to the previous traditional virtualization workload. They both use VMs sitting on top of hypervisors. They both have underlying infrastructure of compute, network, and storage to support these VMs. The difference is primarily is in the operational model.

It’s often described as the difference between pets and cattle. With traditional virtualization, you have pets. You care what happens to these VMs. They have HA and DRS and other technologies to care for them. They’re given clever names, like Bart and Lisa, or Happy and Sleepy. With cloud VMs, they’re not given fun names. We don’t do vMotion/Live Migration with them. When we need them, they’re spun up. When they’re not, they’re destroyed. We don’t back them up, we don’t care if the host they reside on dies so long as there are other hosts carrying the workload. The workload is automatically sharded across the available hosts using logic in the application. Instead of backups, templates are used to create new VMs when the workload increases. And when the workload decreases, some of the VMs get destroyed. State is not kept on any single VM, instead the state of the application (and underlying database) is sharded to the available systems.

This is very different than traditional virtualization. Because the workload distribution is handled with the application, we don’t need to do vMotion and thus have Layer 2 adjacencies. This makes it much more flexible for the network architects to put together network to support this type of workload. Storage with this type of workload also tends to be IP-based (NFS, iSCSI) rather than FC-based (native Fibre Channel or FCoE).

With cloud-based workloads, there’s also a huge self-service component. VMs are spun-up and managed by developers or end-users, rather than the IT staff. There’s typically some type of portal that end-users can use to spin up/down resources. Chargebacks are also a component, so that even in a private cloud setting, there’s a resource cost associated and can be tracked.

OpenStack is a popular choice for these cloud workloads, as is Amazon and Windows Azure. The former is a private cloud, with the later two being public cloud.

  • Latency requirements are mostly the same as traditional virtualization
  • Because vMotion isn’t required, it’s all Layer 3, all the time
  • Storage is mostly IP-based, running on the same network infrastructure (not as much Fibre Channel)
  • Buffer requirements are typically the same as traditional virtualization
  • VXLAN/NVGRE burned into the chips for SDN/Overlays

You can use much cheaper switches for this type of network, since the advanced Layer 2 features (OTV, FabricPath, SPB/TRILL, VCS) aren’t needed. You can build a very simple Layer 3 mesh using inexpensive and lower power 10/40/100 Gbit ports.

However, features such as VXLAN/NVGRE encap/decap is increasingly important. The new Trident2 chips from Broadcom support this now, and several vendors, including Cisco, Juniper, and Arista all have switches based on this new SoC (switch-on-chip) from Broadcom.

High Frequency Trading

This is a very specialized market, and one that has very specialized requirements.

  • Latency is of the utmost concern. To the point of making sure ports are on the same ASIC. Latency is measured in nano-seconds, microseconds are an eternity
  • 10 Gbit at the very least
  • Money is typically not a concern
  • Over-subscription is non-existent (again, money no concern)
  • Buffers are a trade off, they can increase latency but also prevent packet loss

This is a very niche market, one that Arista dominates. Cisco and a few other vendors have small inroads here, using the same merchant silicon that Arista uses, however Arista has had huge experience in this market. Every tick of the clock can mean hundreds of thousands of dollars in a single trade, so companies have no problem throwing huge amounts of money at this issue to shave every last nanosecond off of latency.

Hadoop/Big Data

  • Latency is of high concern
  • Large buffers are critical
  • Over-subscription is low
  • Layer 2 adjacency is neither required nor desired
  • Layer 3 Leaf/spine networks
  • Storage is distributed, sharded over IP

Arista has also extremely successful in this market. They glue PC RAM onto their switch boards to provide huge buffers (around 760 MB) to each port, so it can absorb quite a bit of bursty traffic, which occurs a lot in these types of setups. That’s about .6 seconds of buffering a 10 Gbit link. Huge buffers will not prevent congestion, but they do help absorb situations where you might be overwhelmed for a short period of time.

Since nodes don’t need to be Layer 2 adjacent, simple Layer 3 ECMP networks can be created using inexpensive and basic switches. You don’t need features like FabricPath, TRILL, SPB, OTV. Just fast, inexpensive, low power ports. 10 Gigabit is the bare minimum for these networks, with 40 and 100 Gbit used for connectivity to the spines. Arista (especially with their 7500E platform) does very well in this area. Cisco is moving into this area with the Nexus 9000 line, which was announced late last year.



Understanding the requirements for the various workloads may help you determine the right switches for you. It’s interesting to see how quickly the market is changing. Perhaps 2 years ago, the large-Layer 2 networks seemed like the immediate future. Then all of a sudden Layer 3 mesh networks became popular again. Then you’ve got SDN like VMware’s NSX and Cisco’s ACI on top of that. Interesting times, man. Interesting times.

FCoE versus FC Farce (I’m Tellin’ All Y’All It’s Sabotage!)

Updates 2/6/2014:

  • @JohnKohler noticed that the UCS Manager screenshot used (see below) is from a UCS Emulator, not any system they used for testing.
  • Evaluator Group promises answers to questions that both I and Dave Alexander (@ucs_dave) have brought up.

On my way back from South America/Antarctica, I was pointed to a bake-off/performance test commissioned by Brocade and performed by a company called Evaluator Group. It compared the performance of edge FCoE (non-multi-hop FCoE) to native 16 Gbit FC. The FCoE test was done on a Cisco UCS blade system connecting to a Brocade switch, and the FC was done on an HP C7000 chassis system connecting to the same switch. At first glance, it would seem to show that FC is superior to FCoE for a number of reasons.

I’m not a Cisco fanboy, but I am a Cisco UCS fanboy, so I took great interest in the report. (I also work for a Cisco Learning Partner as an instructor and courseware developer.) But I also like Brocade, and have a huge amount of respect for many of Brocade employees that I have met over the years. These are great and smart people, and they serve their customers well. 

First, a little bit about these types of reports. They’re pretty standard in the industry, and they’re commissioned by one company to showcase superiority of a product or solution against one or more of their competitors. They can produce some interesting information, but most of the time it’s a case of: “Here’s our product in a best-case scenario versus other products in a mediocre-to-worst case scenario.” No company would release a test showing other products superior to theirs of course, so they’re only released when a particular company comes out on top, or (most likely) the parameters are changed until they do. As such, they’re typically taken with a grain of salt. In certain markets, such as the load balancer market, vendors will make it rain with these reports on a regular basis. 

But for this particular report, I found several substantial issues with it which I’d like to share. It’s kind of a trainwreck. Let’s start with the biggest issue, one that is rather embarrassing.

What The Frak?

On page 17 check out the Evaluator Group comments:

“…This indicates the primary factor for higher CPU utilization within the FCoE test was due to using a software initiator, rather than a dedicated HBA. In general, software initiators require more server CPU cycles than do hardware initiators, often negating any cost advantages.”

For one, no shit. Hardware initiators will perform better than software initiators. However, the Cisco VIC 1240 card (which according to page 21 was included in the UCS blades) is a hardware initiator card. Being a CNA (converged network adaptor) the OS would see a native FC interface. With ESXi you don’t even need to install extra drivers, the FC interfaces just show up. Setting up a software FCoE initiator would actually be quite a bit more difficult to get going, which might account for why it took so long to configure UCS. Configuring a hardware vHBA in UCS is quite easy (it can be done in literally less than a minute).

Using software initiators against hardware FC interfaces is beyond a nit-pick in a performance test. It would be downright sabotage.


I asked the @Evalutor_Group account if they really did software initiators:

And they responded in the affirmative.

Wow. First of all, software FCoE initiators is absolutely not standard configuration for UCS. In the three years I’ve been configuring and teaching UCS, I’ve never seen or even heard of FCoE software initiators being used, either in production or in a testing environment. The only reason you *might* want to do FCoE software initiators is when you’ve got the Intel mezzanine card (which is not a CNA, just an Ethernet card), and want to test FCoE. However, on page 21 it shows the UCS blades has having the VIC 1240 cards, not the Intel card.

So where were they getting that it was standard configuration?

They countered with a reference to a UCS document regarding the VIC:

and then here:

Which is this document here.

The part they referenced is as follows:

The adapter offerings are:

■ Cisco Virtual Interface Cards (VICs)

Cisco developed Virtual Interface Cards (VICs) to provide flexibility to create multiple NIC and HBA devices. The VICs also support adapter Fabric Extender and Virtual Machine Fabric Extender technologies.

■ Converged Network Adapters (CNAs)

Emulex and QLogic Converged Network Adapters (CNAs) consolidate Ethernet and Storage (FC) traffic on the Unified Fabric by supporting FCoE.

■ Cisco UCS Storage Accelerator Adapters

Cisco UCS Storage Accelerator adapters are designed specifically for the Cisco UCS B-series M3 blade servers and integrate seamlessly to allow improvement in performance and relief of I/O bottlenecks.

Wait… I think they think that the VIC card is a software-only FCoE card. It appears they came to that conclusion because the VIC doesn’t specifically mention it’s a CNA in this particular document (other UCS documents clearly and correctly indicate that the VIC card is a CNA). Because it mentions the VICs separately from the traditional CNAs from Emulex and Qlogic, it seems they believe it not to be a CNA, and thus a software card.

So it may be they did use hardware initiators, and mistakenly called them software initiators. Or they actually did configure software initiators, and did a very unfair test.

No matter how you slice it, it’s troubling. On one hand, if they did configure software initiators, they either ignorantly or willfully sabotaged the FCoE results. If they just didn’t understand the basic VIC concept, it means they setup a test without understand the most basic aspects of the Cisco UCS system. We’re talking 101 level stuff, too.  I suspect it’s the later, but since the only configuration of any of the devices they shared was a worthless screenshot of UCS manager, I can’t be sure.

This lack of understanding could have a significant impact on the results. For instance, on page 14 the response time starts to get worse for FCoE at about the 1200 MB/s mark. That’s roughly the max for a single 10 Gbit Ethernet FCoE link (1250 MB/s). While not definitive, it could mean that the traffic was going over only one of the links from the chassis to the Fabric Interconnect, or the traffic distribution was way off. My guess is they didn’t check the link utilization, or even know how, or how to fix it if it were off.

Conclusions First?

One of the more odd aspects of this report are where you found some of the conclusions, such as this one:

“Evaluator Group believes that Fibre Channel connectivity is required in order to achieve the full benefits that solid-state storage is able to provide.”

That was page 1. The first page of the report is a little weird to make a conclusion like that. Makes you sound a little… biased. These reports usually try to be impartial (despite being commissioned by a particular vendor). This one starts right out with an opinion.

Lack of Configuration

The amount of configuration they provide for the setup very sparse. For the UCS side, all they provide is one pretty worthless screenshot. Same for the HP system. For the UCS part, it would be important to know how they configured the vHBAs, and how they configured the two 10 Gbit links from the IOM to the chassis. Where they configured for the preferred Fabric Port Channel, or static pinning? So there’s no way to duplicate this test. That’s not very transparent.

Speaking of configuration, one of the issues they had with the FCoE side was how long it took them to stand up a UCS system. On page 11, second paragraph, they mention they need the help of a VAR to get everything configured. They even made a comment on page 10:

“…this approach was less intuitive during installation than other enterprise systems Evalutor Group has tested.”


So wow, you’ve got an environment that you’re unfamiliar with (we’ll see just how unfamiliar in a minute), and it took you *gasp* longer to configure? I’m a bit of an expert on Cisco UCS. I’m kind of a big deal. I have many paper-bound UCS books and my apartment smells of rich mahogany. I teach it regularly. And I’m not nearly as familiar with the C7000 system from HP, so I’d be willing to bet that it would take me longer to stand up an HP system than it would a Cisco UCS system. Anyone want in on that action?

Older Version? Update 2-6-2014: Screenshot Is A Fake


@JohnKohler noticed that on the screenshot on page 23 that the serial number in the screenshot is “1”, which means the screenshot is not from any physical instance of a UCS Manager, but the UCS Emulator. The date (if accurate on the host machine) shows June of 2010, so it’s a very, very old screenshot (probably of 1.3 or 1.4). So we have no idea what version of UCS they used for these tests (and more importantly, how they were configured). 

Based on the screenshot on page 23, the UCS version is 2.0. How can you tell? Take a look where it says “Unconfigured Ports”. As of 2.1, Cisco changed the way ports are shown in the GUI. 2.1 and later do not have a sub-menu for unconfigured ports. Only 2.0 and prior.


In version 2.1, there’s no “Unconfigured Ports” sub-menu. 


From Page 21, you can see “Unconfigured Ports”, indicating it’s UCS 2.0 or earlier.

If the test was done in November 2013, that would put it one major revision behind, as 2.1 was released in November of 2012 (2.2 was released in Dec 2013). We don’t know where they got the equipment, but if it acquired anytime in the past year, it likely came with 2.1 already installed (but not definite). To get to 2.0, they’d have to downgrade. I’m not sure why they would have done that, and if they brought in a VAR with certified UCS people they would have likely recommended 2.1. With 2.1, they could have directly connected the storage array to the UCS fabric interconnects. UCS 2.1 can do Fibre Channel zoning, and can function as a standard Fibre Channel switch. The Brocade switch wouldn’t have been needed, and the links to the storage array would be 8 Gbit instead of 16 Gbit. 

They also don’t mention the solid state array vendor, so we don’t know if there was capability to do FCoE directly to the storage array. Though not terribly common yet, FCoE connection to a storage array is done in production environments and would have benefited the UCS configuration if it were possible (and would be the preferred way if competing with 16 Gbit FC). There would have been the ability to do an LACP port channel between the Fabric Interconnects, providing better load distribution and redundancy.

Power Mad

The claim that the UCS system uses more power is laughable, especially since they specifically mentioned this setup is not how it would be deployed in production. Nothing about this setup was production worthy, and it wasn’t supposed to be. It’s fine for this type of test, but not for production. It’s non-HA, and using only 2 blades would be a waste of a blade system (and power, for either HP or Cisco). If you were only using a handful of servers, buying pizza boxes would be far more economical. Blades from either HP or Cisco only make sense past a certain number to justify the enclosures, networking, etc. If you want a good comparison, do 16 blades, or 40 blades, or 80 blades. Also include the Ethernet network connectivity. The UCS configuration has full Ethernet connectivity, the HP configuration as shown has squat.

Even by competitive report standards, this one is utter bunk. If I was Brocade, I would pull the report. With all the technology mistakes and ill-found conclusions, it’s embarrassing for them. It’s quite easy for Cisco to rip it to shreds.

(Just so there’s no confusion, no one paid me or asked me to write this. In fact, I’m still on PTO. I wrote this because someone is wrong on the Internet…)


Get every new post delivered to your Inbox.

Join 76 other followers