Slashdot Mirror
Cisco Pricing Undercut By $100M In Big Cal State University Network Project
alphadogg writes "The $100 million price differential between the Alcatel-Lucent and Cisco proposals to refresh California State University's 23-campus network revealed earlier this week was based on an identical number of switches and routers in various configurations. CSU allowed Network World to review spreadsheets calculating the eight-year total cost of ownership of each of the five bidders for the project. 'Everybody had to comply with this spreadsheet,' said CSU's director of cyberinfrastructure. 'Alcatel-Lucent won the project with a bid of $22 million. Cisco was the high bidder with a cost just under $123 million. Not only was Cisco's bid more than five-and-a-half times that of Alcatel-Lucent's, it was three times that of the next highest bidder: HP, at $41 million.'"
People who don't get competitive quotes but always buy Cisco because that's what they know.
I have always felt that Cisco had the same sort of following as Novell. Senior IT people certified up the wazoo yet unable to explain to me why Cisco was so much better. The bits that leak out of big data people like Facebook and Google seem pretty lacking in the big names. I don't see gear from HP, IBM, Dell, Cisco, etc. What I do see is white boxish or custom gear that they seem perfectly happy with.
Just a guess but my bet is that much of the business that big old companies like Cisco come from single skill IT people combined with kick ass sales people. Salespeople who sell to upper management not to the non Cisco IT people who might fact check.
So good job to the people who didn't blow an extra $100 Million.
I bet Belkin could do the job for a cool 1 million. Of course it wouldn't work, but look at the savings.
On big projects like these, the contractors try to find out how much the customer has budgeted or is willing to pay. Like, by inviting customers to strip joints, etc.
Once they have an idea of how much money the customer has, they adjust their bid to fit it.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
I figured the difference was in Monster cables.
Tic-Tac-Toe, Global Thermonuclear War, and relationships all have the same winning move.
The idea that American schools don't have enough money is absurd. America spends more per capita on its schools than any other nation in the world.
Now, that all that money is not correctly distributed among schools is clear too. And far more important than that, all the money in the world doesn't matter if mommy and daddy don't encourage and take part in junior's education. Which in makes marginal investments in failing schools pointless, because it's the entire environment of the district that's failing the students, not just the school.
...unlike those other companies, Cisco's products are carefully and lovingly fabricated in..... China?
Oops..
"The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
I wanna see the final cost after the project is done and everything is working.
$22M sounds low for a project this of this size, so I wonder if Lucent is planning to make up the difference with consulting fees.
Or maybe I'm just jaded from paying Cisco prices for so long... and also from seeing low-ball bids costing a lot more in the end.
Cisco is like Oracle. They don't need to discount their prices.
Eh? Oracle discounts, heavily. You only pay "list" to Oracle if you're a small, unimportant customer. The big fish get up to something like 99% off when Oracleewants to lock out a competitor.
but holy shit! how do they stay in business?
With luxurious profit margins. As the saying goes "A fool and his money... that's who you want to focus on."
When our name is on the back of your car, we're behind you all the way!
by winning more than a fourth of the contracts that ALU wins.
What was the Huawei bid?
Only a fool would do an important networking job without monster cables. Do you know how much packet fidelity you can lose if stray RF gets into your fiber?
I don't know if it is a bug or an intended feature but it happens from time to time.
There's a lot of PHBs who find Cisco to be "reassuringly expensive".
A bit like Oracle, et al.
No sig today...
Sure, but doesn't most of it go on luxuries and fancy facilities to attract students rather than actual teaching?
Sports facilities better then most professional teams, plasma TVs in every dorm room, etc.
No sig today...
I'm not sure that was a real quote from Cisco. It looks to me like they simply didn't want the business. In such cases, business file what are called f-you quotes, which are outrageously priced to take into account that the bidder may not currently have the capability to fill the contract, or that it would be defocusing. Priced high enough, they could sub-contract to HP, for example, and still make a lot of money.
That said, I went to our local office the other day and poked my head into the networking closet. I see the same cheap crummy wifi routers I put there before our little company got bought. Right next to them is a Cisco router worth maybe $10-20K. It's worth more than all the computers and related hardware in the office combined.
Celebrate failure, and then learn from it - Nolan Bushnell
It's worth more than all the computers and related hardware in the office combined.
Debatable. It cost more than all the computers and related hardware in the office combined for sure.
I guess today is a passable day to die.
Wondering, was the offer directly from Cisco? Did the person who designed/selected the gear know what they were doing? ... e.g., we just had a project in which a company campus with something like 20 Gigabit switches (24/48 ports, access layer) and a core with 10G ability to feed to those as well as cover the DC with redundant 1G ports ... going with the usual suspect (6500) as core switch with line cards to supply up to 16 10G ports and 96 1G copper ports would have been more than twice the price than the alternative we chose, Nexus 5548 w/ two 2248 FEX chassis. ... (total of 44 SFP+, 42 SFP, with original Cisco SFPs that would add up to around 50k€ - would have been a third of the whole project budget. Using OEM/compatible modules was around 5k€). Assuming a large quantity of fiber ports in such a project, the optics alone may quickly add up to the factor mentioned above ...
Just by selecting the wrong gear, prices between different Cisco gear can already differ by a factor of 2-3
Also, instead of using overpriced (to say the least) Cisco SFP/SFP+ modules would have run the total bill up even more
You mean features like storing data and dishing it back out; or nonsense features like CPLM5 certification?
Plus you are comparing corporate Oracle to Corporate SQL. For most people all Free and Open Source would be just peachy. Most people including facebook. I rarely see the really big big big sites doing anything with any of the Oracley Microsofty IBMy stuff. They usually take Open Source and then roll their own. Sort of shows that the route to success starts with open source and ends with modified open source.
It's very possible. If you read the RFPs for some government things, you'll find things that almost no vendor can possibly adhere to. If you are a top tier vendor like Cisco, you likely CAN meet the requirements, but not cheaply. So instead of trying to compete on price, you compete on being able to fulfill all of the requirements in the RFP. You take the gamble that the people analyzing the proposals will nix the cheaper ones as non-compliant, and you are the only bidder left. Or, that the agency will cancel the RFP and rewrite a new one with different or clarified requirements. Then everyone rebids with full knowledge of each others' pricing, and hopes for the best.
Mommy and daddy don't have a clear understanding about money or economics any more than congresscritters or teachers do. If they did, they'd teach their children that formal and informal educations combined are the only way they'll really have a successful future beyond a "pawn's mate".
There are multi-level failures within the American education system. It starts with parents ultimately not understanding how to really help their children succeed and it is further exacerbated by poor management of money from local agencies all the way up through the federal Department of Education. There are also societal values at the lower- and middle-classes which can cause education which is received to be ineffective; take for example the middle-class priority of being good at a sport and getting scouted for a team in high school over academic and economic educations, or in poorer neighborhoods the priority on being tough or able to survive on the streets instead of learning something in school which might elevate them out of that cesspool.
If your family can't teach you, your teachers can't teach you, and the gov't can't teach you...then how do you learn? Vicious cycle, if you ask me.
What else can happen when an unstoppable force collides with an immovable object?
America spends more per capita on its schools than any other nation in the world.
Actually, we rank fourth
And on a percentage of GDP basis>/a> The US ranks 37th, tied with Estonia.
No, no, you're not thinking; you're just being logical. --Niels Bohr
They're doing this because Mark Benioff is Larry Ellison's ex-protoge, and their relationship took a turn for the worse and now they're in a gigantic dick measuring contest. Goes something along the lines of this-
Benioff (Presents Steve Jobs Style) "We are the original cloud, there's 'no software' and it's all a 'social enterprise"
Larry (Presents Balmer Style) "We do cloud too, and we're Oracle, so fuck you".
What a bunch of wankers (worked on SFDC and Oracle)
That's a fact. It moves data in a highly standardized way. Sure, there are some proprietary Cisco protocols here and there, but for the most part, it's all the same everywhere. Whatever Cisco does, anyone can do.
People somehow believe there's magic moving data over wires. There just isn't. And there's nothing special about Cisco's. Now, compare Cisco to Microsoft. Now *There* is some vendor lock-in. One thing depends on another thing and another thing and another. To move off of Microsoft is mind-numbingly ridiculous to imagine. But Cisco? Nah. You can replace this and that here and there and you'll be just fine. Sure, you might have to migrate away from the use of anything Cisco proprietary here and there, but for the most part? You can take your time and move bits and pieces here and there.
That doesn't quite count if you're talking about Cisco phones... that's kind of an all or nothing scenario there... within limits. One thing is certain though -- Cisco needs to be humbled.
No. That's not true. It takes brain power. There's a wealth of information out there because pretty much, it has all been done before in one form or another.
It seems cheaper in the short run to buy something off the shelf and put it up. But when you keep paying for it over and over and over again, you might begin to realize that people are cheaper in the long run.
Besides that, do you think the likes of Google STARTED out with billions of dollars? How about Facebook and the others like them? They started with some pretty smart people which turned out to be a far better investment than paying for licensed off-the-shelf stuff.
Actually, Forth kind of sucks. I prefer LISP.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
More News:
The sky is blue
I live in Seattle, you insensitive clod!
There's no place I could be, since I've found Serenity...
99%? Doubtful.. But hefty discounts indeed, and Oracle is still obscenely expensive
So many injustices..so little time..
I'm not sure that was a real quote from Cisco. It looks to me like they simply didn't want the business.
Not really, looking at the spreadsheets, it's typical pricing for Cisco. Especially once they started quoting Nexus-backed infrastructure with OTV to stretch layer-2. You'd be surprised at how many people have been biting off on massive OTV and Nexus costs with no competitive analysis. Looks like Cal State just did an objective analysis without marketing hype, and kudos to them.
You clearly don't know what "per capita" means.
I have to completely agree with this. I've been involved with several large-scale RFPs, and this is exactly how it goes. The only thing I'd add is that like clockwork, any party that doesn't win threatens to sue someone. It happens every time. They must be teaching this in business school or something. I've never seen a more childish group of people.
I'm partial to HP gear, and I always claimed that it has quite decent TCO even in very small scale deployments (we have 5k worth of gear, not 40M). People who buy Cisco must be getting a lot of free pussy or something.
Yes, in a small deployment, HP is very cost effective and works well. Even Netgear works well if your budget is $5K.
But in my experience, Cisco does win in manageability and scalability of larger networks. And if things stop working, you can get a knowledgable engineer on the line quickly. We're a pretty small customer (~200 switches), and when we ran into a weird problem with some switches in our environment caused by an IOS bug, we got a custom patch for the problem, which Cisco rolled into the next IOS release.
But you have to understand that there is a quality difference, and a quite substantial one at that. Cisco's gear, while far from perfect, is massively better than any Alcatell-Lucent stuff. Alcatell-Lucent makes one-off routers. They make it, they ship it, they never look at it again. So if you find your bgp crashes every 2 days ... that just becomes a fact of life (I actually replaced bgp peering on an alcatell with a firewall rule redirecting any bgp sessions to a quagga daemon running on linux, which then ibgp peered with the alcatell. This was actually more robust than their own implementation).
I've seen plenty of linux setups doing the same (network/firewall) job as cisco gear, hell I've even seen a freebsd setup that used pf as a replacement for missing routing features in the freebsd kernel with millions of rules total (which worked), but all those setups did have something in common : horribly hard to figure out what they were doing, and the vast majority of them were configured by admins without a clue of networking. This meant obvious holes, and even when it was done competently, made the setup as a whole trivial to ddos. (my remote-bgp solution for the alcatell monsters was one such thing : attack port 179, play around a bit with packet ttl's and the whole network would die. Nobody ever figured that out though).
Cisco's hardware routers, properly configured, are unddossable. The same cannot be said for their firewalls though.
What I'm saying, as a network admin : please, please, please don't install cheap "one-off" equipment as a supposed replacement for quality hardware. And if you do, expect to be quoted at least $100/day more than everybody else for consultancy to fix the mess, and expect it to take a month longer than normal (that still may make cheap network equipment worth it, though do remember than having your network offline every 30 minutes for a month is bound to make customers extremely unhappy, and more consultants won't help). Not because it's not cisco (juniper is fine, though even more expensive).
Just look at West Virginia's last big IT project and you'll see how Cisco stays in business.
A similar situation also happens (although I have no idea if it happened in this case).
The "preferred" vendor "assists" in writing the RFP. They get it written so they can fulfill it economically while competitors can't. Sometimes this is done by specing features missing from competitor's product or insuring competitors have to spec higher end products to pick up one "required" feature that the "assisting" vendor just happens to include in a lower priced model.
Why is there an "insightful" mod and why isn't it "-1"? If I wanted insight, I wouldn't be reading
You either pay for the brainpower upfront with pre-configured hardware/software, or make up the difference with smart (but expensive) people. The trick is which lasts longer, or is a better long-term investment.
I could employ a crappy IT admin who only knows windows, and pay him $40k and the enterprise Win suite and license seats for $20k total; or a decent linux admin at $60k and virtually unlimited server and seats running CentOS and Ubuntu for $0.
This is of course a very simplified way of seeing things. There are many more variables to consider, such as how competent is the end-user with computers?
When we had an RFP for new network equipment Cisco's proposal was twice the price of the next bid. In the end Juniper won the contract. Of all the proposals only the party offering Avaya/Nortel made a judgement error and ignored a few requirements which they could have fulfilled with a bit more expensive kit and still come out as the winner. The requirements were pretty high but not impossible. We had proposals with kit from Cisco, Avaya/Nortel, H3C, Juniper and Alcatel.
If the price for the Cisco proposal was not a real price one wonders why they submit a proposal at all. Writing the proposal costs money and if they submit a f-you quote they know the will get nothing in return.
Because of idiots with a massive budget to spend or get cut. The contract was with Verizon, providing Cisco hardware. Someone wanted top-of-the-line, whether it was appropriate or not.
It all depends on who you are. At my last company, the local Cisco account team wanted us as a reference account. Our discount from Cisco (before distribution and the partner took their cut) was 59%. Our final discount was 55% or 57% (depending on whether we went through disti or direct from Cisco).
At my current company, our discount from Cisco (again before disti and partner) is 55%. Our price after partner and disti is 47% (yes, our partner is taking a ridiculous cut).
The problem is that most customers don't have any idea what list price is, and so they have no clue whether they are even getting a reasonable price. Sadly, Cisco, along with their partners, do their best to hide this information from their customers.
huh! That's even less than your source says we spend on the military, which initially surprised me. But, my first question was: Did they account for spending by the states in their estimate? Further, in trying to sort out the source for the data, the site seemed to repeatedly cite itself in an orobouros-like fashion. Do you have another source?
Is data from The United Nations acceptable?
No, no, you're not thinking; you're just being logical. --Niels Bohr
CIO's and other high level executives didn't get to where they are by taking big risks. They are generally very conservative and play it by the book. When the time comes to make a big decision like infrastructure or database or enterprise software they are not going to make that decision unilaterally, even if they are perfectly capable of making it themselves. They are going to bring in "experts", the ones you refer to as high paid consultants, and have them conduct studies and analysis and come up with a recommendation. The CIO will then almost certainly follow that recommendation. Why? Because if something goes wrong he hasn't put his neck out on the line. He has taken the prudent step of bringing in experts and is following their opinions.
Consultants are not necessarily any smarter than anyone else but they often have the advantage of specializing in a given field. When you hire them they can focus on a single task and not get drawn into office politics or have to split their time amongst several projects at once. Sure, some of them are lousy...just like there are lousy doctors and lousy carpenters and lousy hairdressers. But in business appearing to do the right thing is often as important as doing the right thing.
It's the same reason that many companies go the "packaged software" route, rather than develop it in house. If something goes wrong you pick up the phone and report the problem and the vendor is on the hook to fix it. If you build it in house and your lead programmer quits the next day then what do you do?
But times are changing. For many of the Web 2.0 companies (Facebook, LinkedIn, etc.) buying packaged software like Oracle is prohibitively expensive when you are starting out. So they use MySQL or Postgre SQL and they are very, very good products. In the right hands they can be just as good as anything from Oracle or Microsoft. It allows them to focus capital on other things and bring ideas to market without incurring extra debt. It's a smart move and it has been proven to be as scalable as anything else out there.
They wanted a router which could handle both a T1 (so a slow legacy software router with obsolete interfaces) and a fiber (most likely 1Gbps ethernet). The 3945 is not a particularly stupid choice if those are the requirements and you like Cisco.
Now, split those requirements in two, and you can use a) any old T1 router and b) a random 1Gbps ethernet CPE. a) costs practically nothing refurbished. If you go with Cisco and you're a government agency, you can probably get an ME3400 with advanced license for under $1000. Juniper EX2200C + license is a bit cheaper than that, OneAccess is a lot cheaper.
Extra costs for the 2 router solution: A power bar with 2 outlets and an ethernet cable...
Finally! A year of moderation! Ready for 2019?
To price itself out of the market. I recall when I worked at the RI Sec of State's office we did a major move of several units within the department to a new space in a different building. I had to go out and spec pricing for switches, routers, and security gear. For basic core networking I looked at Cisco and HP. For the features I required, namely easy management, VLAN, etc. both offered it but the Cisco gear was 3 times the price of the HP. Cisco essentially thinks that because it is the predominant vendor for networking hardware that they can charge a heavy premium. They have also bought up competitors whenever they could to limit the market.
Reading the article it's easy to see that there was a huge discrepancy in capabilities, at least to anyone familiar with the various product lines. Cisco proposed a very high end solution, for instance offering up their Nexus solution for the data centers. Alcatel-Lucent simply doesn't have anything similar, although they could build a fine data center solution with slightly less bells and whistles. HP, well, they make some great switching devices, but their L3 routing capabilities are woefully short of both Cisco and Alcatel-Lucents. In fact, that's my biggest clue something went wrong here, if an HP solution is being compared to Nexus, well, that's about as far on opposite ends of the networking spectrum as you can get.
These bids were not at all for the same thing, which tells me the university did a very bad job of writing the RFP. If you put out an RFP saying "I need a car that can take 2 people 100 miles" that spec can be met by a Lamborghini Aventador and a Nissan Versa. The reality is probably neither are appropriate for someone who wants a good value, middle of the road solution.
I have no doubt Cisco could offer up a solution with the same capabilities as Alcatel-Lucent or HP for a competitive price, and no one knows why they didn't do that here. Also, even with similar hardware capabilities speced Cisco software has a lot of features the other vendors simply do not have. Are they worth millions extra? Probably not, but they are worth some extra. If the university had competent people writing the RFP they could have pointed to features that reduce manpower needs and gotten more appropriately priced equipment.
Having written and reviewed a number of RFP's, one of our criteria was the spread on the responses. When it is this large something has likely gone wrong with the RFP process, and it needs to be rebid with better specifications. Back to my car example you can throw in things like it needs to run on regular gas (no more Aventador, or other high end cars), or that it needs to have at least 15 cubic feet of trunk space (no more Versa), and put yourself in a much more reasonable range.
Rather than picking the low bid here the university needs to take a serious look at their requirements, and put out a revised RFP.
I'll agree that support of learning by the family is significant. Not *quite* essential, given an optimal for learning school environment, but even then significant. (And no school in the country, in the world as far as I know, offers an optimal for learning environment. It requires 1-3 teachers/child, e.g. [and not even always the same 1-3 teachers.])
OTOH, my wife teaches music and art in her studio, and she reports that children whose mothers don't go to work tend to have children that are about twice as easy to teach. These parents are obviously committed to their children's learning, or they wouldn't enroll with my wife, but so are the parents that work. The probable conclusion that I draw from this is that working full-time strongly interfers with doing a good job as a parent.
Again, there was I study that I read reporting on class size vs. learning in highschool. That reported a sharp decline in learning when the class size exceeded 18. There was a smooth curve drawn, but I don't know how well it actually fit the data. Still, there was a slow decline in learning rate from class size of 15 approaching 18, 18 was the knee of the curve, and there was a fairly rapid rise beyond 18. One needs to wonder what other effects were present, e.g., what other factors do students that good to schools with a small class size have in common. Still, it's not a totally unreasonable conclusion, given the difficulty that high school teachers have in controlling their classes (any effort spent in controlling the class obviously takes away from effort that could be spent instructing), and also attending to each students unique needs can be time consuming. Large classes prohibit that.
That said, the "teach to the test" is guaranteed to produce students inferior in every way except in test taking. And it's not clear that they'll be much good even in taking different kinds of tests. It's not clearly an entirely wasted effort. Twice a year would be reasonable. Say once at the start of the school year, and once at the end. More frequently is a waste of time, resources, effort, attention, learning, money, etc.
I think we've pushed this "anyone can grow up to be president" thing too far.