Re:In My Opinion, Cisco Should Be Worried
on
Google Router Rumors
·
· Score: 4, Interesting
Isn't Juniper's business plan to install FreeBSD on cheap embedded hardware and pretend that it's special-secret-proprietary-magic? I wouldn't be surprised if Google could undercut them, for in-house use at the very least.
This is not really true. On the higher end Juniper boxes, while the control plane is running FreeBSD, the real work is done on the forwarding plane which is comprised of custom ASICs. You can't route at an enterprise or carrier level using commodity hardware.
If Google is building an in-house router, it's down to the hardware design level. Either they're developing their own ASICs (plausible) or they're using merchant silicon (even more plausible) and rolling their own OS and chassis.
Your post, it makes no sense. Have you ever used a GPS receiver? Mine almost *never* loses signal, and in the mountains you get a great signal.
I can't comment on the mountains, be he/she is right about cities with tall buildings. Driving in New York City with a GPS can be interesting if you're relying on it. Sometimes it can't even make heads or tails of what direction you're moving.
The tricky part is that vendors who sell the untaxed stuff retail are not allowed to have pumps that are capable of dispensing directly into a vehicle (primarily accomplished by making the hose too short to reach).
Call me crazy, but what prevents me from filling a tank and then using that to fill my car/truck?
The 2960 may be a "gig switch" but it is not a gig switch. Get a 3750. Better yet, get an HP 3500yl and be done with it. Unless you have real pull for pricing with Cisco (some of us do), HP makes way more sense with the 3500yl and the 8200zl than Cisco with 3750 and 6500.
Uhm. It's common courtesy to expand an acronym the first time you use it. While I would normally assume that NOC means "Network Operations Center," it makes no sense in this context. And as for "COGS," I'm not sure where to begin.
The only shops that actually look at cost/GB as a measuring stick are small shops, or shops with very specific needs.
Large corporations, government and high tech companies are usually more concerned with management costs, retention, migration and so forth.
This is simply not true. There are plenty of commodity storage requirements that do not require Fibre Channel or even NetApp level NAS. On the other end of the spectrum, cost/GB might not be a huge factor, but the cost of getting necessary IOPS is certainly a factor.
I work on Wall St. and we have multiple PB of storage. We have tons of EMC. However, things like the Sun X4500 and similar products from HP are changing the game. Couple that with being able to do 48 ports of line-rate 10GigE in a 1 RMU stackable, per priority pause coming into use, and Data Center Ethernet down the road and you have many reasons to seriously reconsider the scope of your fibre channel deployment.
Gig-E speed is about 30MB/s in the real world. This is with a crossover cable, machine to machine. I've tested and verified this over a number of platforms, including expensive server systems.
Either your "real world" resides in a parallel universe where things aren't what they seem, or your test was flawed.
I recently tested my MacBook Pro to my iMac over GigE through a switch (using iperf). With jumbo frames enabled, I got 986 Mbps. With jumbo frames disabled, I got about 870 Mbps. Not sure where your "real world" 30MB/s comes from.
You can use the Solaris 10 Operating System at home or at work -- without paying a license fee. Download the Solaris 10 OS today - FREE -- when you register your system (as part of the download or by going through the registration portion of the download) you will receive an Entitlement Document, which grants you unlimited rights to use Solaris registered machines. Sun offers this no-cost licensing program for all use - even commercial use, even on multi-processor machines.
... to say that software RAID is almost invariably a poor solution. It is woefully slow compared to even a slow hardware RAID implementation.
Spend a few bucks and get the right hardware. It is not expensive these days.
This may have been true years ago, but it's not anymore. Modern CPUs can handle parity computations without a problem. As long as your controllers can support the throughput needed, there is no need for hardware RAID. After all, we have ZFS.
Storage is undergoing a massive paradigm shift and folks like EMC are being caught with their pants down. Their spindle cost and price per GB is just too high.
Encryption only works because brute forcing the scheme on current hardware is ridiculously time consuming. Encrypting with today's standards does not protect against future advances in computing power.
Do you know much about encryption? Actually, no, you've already answered that. Today's encryption standards sufficiently protect against any conceivable increase in computing power in your and many generations of your offspring's lifetimes.
Looks fun but I am still waiting for 3ware Solaris drivers.
3ware is redundant on Solaris. There's no reason to be doing hardware RAID if you can do ZFS. Take all your drives on 3ware and put them on commodity controllers.
General purpose hardware today is fast enough that dedicated RAID controllers are getting nearly obsolete.
I'm not most people. I'm just me. Like at least some of the "developers" mentioned in the title of this article I develop software that has to run on case-sensitive filesystems.
Can you explain why your software has to run on case-sensitive filesystems? Do you mean that your software must use a case sensitive file system to work or simply that your software can run on a case sensitive file system if need be?
We can argue about what might have happened forever. The crucial detail is that she was served coffee hot enough to put her in the hospital with serious burns. Not scalds, burns.
I really find this laughable. The coffee couldn't have been any hotter than about 100 degrees celsius. Guess what, when I get coffee or tea, I expect it to be that temperature. Only a retard wouldn't.
I missed the part where she went and poured the coffee on her legs, so I'm going to chalk that up to desperate rationalization on your part.
Spilling it on her legs and deliberately pouring it on her legs are effectively one in the same. Both are her fault, not McDonald's fault, and not anyone else's fault. Her fault.
Except there is, because McDonald's intentionally kept their coffee hotter than needed because they thought customers would be driving long distances and not drinking the coffee until much later.
What about the person who went to McDonald's because their coffee was hotter? If she didn't want burns, she shouldn't have poured coffee on her lap. That's it. This is not a 2 year old touching a hot stove. This is an adult handling a beverage that, by definition, is made at or near boiling temperatures.
Instead of this generating a proper feedback loop of "don't pour hot stuff on lap" to the general population, it makes people think "I can be a retard and get paid for it!" The saving grace is that if male to do this around reproductive age, he might never get a chance to breed.
That woman who sued over hot coffee was not simply whining about scalding her hands. She went to the hospital with 3rd degree burns. Probably the coffee had been reheated in a microwave. One hazard of heating liquids this way is that you can make them superhot [wikipedia.org] without causing them to boil.
Boohoo. I doubt the coffee was reheated in a microwave. Even if it was, the superheating effect generally doesn't happen with styrofoam or paper cups because the insides are pretty rough. Even if it does happen, as soon as you touch the container, the liquid flash boils. I've done it (in ceramic as it's a smooth surface). So even if was superheated, it would have flash boiled by the time she got the cup, hence reducing the temperature to a maximum of 100 degrees celsius. I don't know about you, but I pretty much assume coffee and tea is going to be 100 degrees celsius. That's why I don't pour it over my legs. If I were to, I would suffer the consequences, maybe including not having my stupid genes copied to another human.
There have been frivolous lawsuits, definitely true. The scalding coffee was not. Other coffee vendors around the city were, at the highest temperature, 20 degrees Fahrenheit cooler than McDonald's coffee.
It absolutely was frivolous. Here's a tip, a "life lesson" if you will: If you buy something that is, by definition, hot, don't pour it all over your legs. I have no sympathy for people who do stupid things and then blame other people. There is no argument here. Period.
that's not a solution when you have applications processing information, and you switch over while your in the middle of processing requests. In my situation there isn't a single second the system isn't fielding 100's of requests. basicly it involves a hand shake where the client makes a requests and expects an answer, if you switch over the new system won't know the client is expecting an answer so you'd have to re engineer a black box system to do it somehow.
Ever heard of connection draining? You build systems with the expectation that they will fail. Any component at any given point in time should be expected to be broken, because it will be at some point. If your system can't handle bringing down a server for maintenance, then you have far bigger problems than picking a good OS. Good luck to you.
Slashdot Mirror
Isn't Juniper's business plan to install FreeBSD on cheap embedded hardware and pretend that it's special-secret-proprietary-magic? I wouldn't be surprised if Google could undercut them, for in-house use at the very least.
This is not really true. On the higher end Juniper boxes, while the control plane is running FreeBSD, the real work is done on the forwarding plane which is comprised of custom ASICs. You can't route at an enterprise or carrier level using commodity hardware.
If Google is building an in-house router, it's down to the hardware design level. Either they're developing their own ASICs (plausible) or they're using merchant silicon (even more plausible) and rolling their own OS and chassis.
Your post, it makes no sense. Have you ever used a GPS receiver? Mine almost *never* loses signal, and in the mountains you get a great signal.
I can't comment on the mountains, be he/she is right about cities with tall buildings. Driving in New York City with a GPS can be interesting if you're relying on it. Sometimes it can't even make heads or tails of what direction you're moving.
In this case, it's not that they want more money; they want to keep getting the same amount as always. Slight difference.
Talk about naïve. The government always wants more money. It's why they are evil. Especially since I get way less back for the money I pay in taxes.
The tricky part is that vendors who sell the untaxed stuff retail are not allowed to have pumps that are capable of dispensing directly into a vehicle (primarily accomplished by making the hose too short to reach).
Call me crazy, but what prevents me from filling a tank and then using that to fill my car/truck?
The 2960 may be a "gig switch" but it is not a gig switch. Get a 3750. Better yet, get an HP 3500yl and be done with it. Unless you have real pull for pricing with Cisco (some of us do), HP makes way more sense with the 3500yl and the 8200zl than Cisco with 3750 and 6500.
Uhm. It's common courtesy to expand an acronym the first time you use it. While I would normally assume that NOC means "Network Operations Center," it makes no sense in this context. And as for "COGS," I'm not sure where to begin.
Routing TCP/IP Volume I (ISBN-13: 978-1587052026) also falls into the "classics" category of Cisco Press.
Arista Networks. Formally known as Arastra.
Self-correction: That should be formerly.
Who's doing 48 ports of 10GbE? That's pretty damn impressive.
Arista Networks. Formally known as Arastra.
The only shops that actually look at cost/GB as a measuring stick are small shops, or shops with very specific needs.
Large corporations, government and high tech companies are usually more concerned with management costs, retention, migration and so forth.
This is simply not true. There are plenty of commodity storage requirements that do not require Fibre Channel or even NetApp level NAS. On the other end of the spectrum, cost/GB might not be a huge factor, but the cost of getting necessary IOPS is certainly a factor.
I work on Wall St. and we have multiple PB of storage. We have tons of EMC. However, things like the Sun X4500 and similar products from HP are changing the game. Couple that with being able to do 48 ports of line-rate 10GigE in a 1 RMU stackable, per priority pause coming into use, and Data Center Ethernet down the road and you have many reasons to seriously reconsider the scope of your fibre channel deployment.
Gig-E speed is about 30MB/s in the real world. This is with a crossover cable, machine to machine. I've tested and verified this over a number of platforms, including expensive server systems.
Either your "real world" resides in a parallel universe where things aren't what they seem, or your test was flawed.
I recently tested my MacBook Pro to my iMac over GigE through a switch (using iperf). With jumbo frames enabled, I got 986 Mbps. With jumbo frames disabled, I got about 870 Mbps. Not sure where your "real world" 30MB/s comes from.
Solaris isn't free, it is a 90 day trial. I went to setup a file server using Solaris and quit after reading the fine print.
Uh. Complete and utter FUD. From the horse's mouth:
You can use the Solaris 10 Operating System at home or at work -- without paying a license fee. Download the Solaris 10 OS today - FREE -- when you register your system (as part of the download or by going through the registration portion of the download) you will receive an Entitlement Document, which grants you unlimited rights to use Solaris registered machines. Sun offers this no-cost licensing program for all use - even commercial use, even on multi-processor machines.
... to say that software RAID is almost invariably a poor solution. It is woefully slow compared to even a slow hardware RAID implementation.
Spend a few bucks and get the right hardware. It is not expensive these days.
This may have been true years ago, but it's not anymore. Modern CPUs can handle parity computations without a problem. As long as your controllers can support the throughput needed, there is no need for hardware RAID. After all, we have ZFS.
Storage is undergoing a massive paradigm shift and folks like EMC are being caught with their pants down. Their spindle cost and price per GB is just too high.
yeah, go get a 2u server with a 3ware raid card - $2000 and it can run at disk speeds.
Or just use commodity SATA and ZFS. Much better than 3ware.
Encryption only works because brute forcing the scheme on current hardware is ridiculously time consuming. Encrypting with today's standards does not protect against future advances in computing power.
Do you know much about encryption? Actually, no, you've already answered that. Today's encryption standards sufficiently protect against any conceivable increase in computing power in your and many generations of your offspring's lifetimes.
Lose two hard drives, your RAID is still AOK
That's an awfully simplistic statement, and only applicable for RAID-Z2. A simple mirror or RAID-Z can (understandably) only sustain a single failure.
Hot swap?
With the right disk and controller? Absolutely. AHCI supports hot-swap with SATA drives.
Looks fun but I am still waiting for 3ware Solaris drivers.
3ware is redundant on Solaris. There's no reason to be doing hardware RAID if you can do ZFS. Take all your drives on 3ware and put them on commodity controllers.
General purpose hardware today is fast enough that dedicated RAID controllers are getting nearly obsolete.
I'm not most people. I'm just me. Like at least some of the "developers" mentioned in the title of this article I develop software that has to run on case-sensitive filesystems.
Can you explain why your software has to run on case-sensitive filesystems? Do you mean that your software must use a case sensitive file system to work or simply that your software can run on a case sensitive file system if need be?
We can argue about what might have happened forever. The crucial detail is that she was served coffee hot enough to put her in the hospital with serious burns. Not scalds, burns.
I really find this laughable. The coffee couldn't have been any hotter than about 100 degrees celsius. Guess what, when I get coffee or tea, I expect it to be that temperature. Only a retard wouldn't.
I missed the part where she went and poured the coffee on her legs, so I'm going to chalk that up to desperate rationalization on your part.
Spilling it on her legs and deliberately pouring it on her legs are effectively one in the same. Both are her fault, not McDonald's fault, and not anyone else's fault. Her fault.
Except there is, because McDonald's intentionally kept their coffee hotter than needed because they thought customers would be driving long distances and not drinking the coffee until much later.
What about the person who went to McDonald's because their coffee was hotter? If she didn't want burns, she shouldn't have poured coffee on her lap. That's it. This is not a 2 year old touching a hot stove. This is an adult handling a beverage that, by definition, is made at or near boiling temperatures.
Instead of this generating a proper feedback loop of "don't pour hot stuff on lap" to the general population, it makes people think "I can be a retard and get paid for it!" The saving grace is that if male to do this around reproductive age, he might never get a chance to breed.
That woman who sued over hot coffee was not simply whining about scalding her hands. She went to the hospital with 3rd degree burns. Probably the coffee had been reheated in a microwave. One hazard of heating liquids this way is that you can make them superhot [wikipedia.org] without causing them to boil.
Boohoo. I doubt the coffee was reheated in a microwave. Even if it was, the superheating effect generally doesn't happen with styrofoam or paper cups because the insides are pretty rough. Even if it does happen, as soon as you touch the container, the liquid flash boils. I've done it (in ceramic as it's a smooth surface). So even if was superheated, it would have flash boiled by the time she got the cup, hence reducing the temperature to a maximum of 100 degrees celsius. I don't know about you, but I pretty much assume coffee and tea is going to be 100 degrees celsius. That's why I don't pour it over my legs. If I were to, I would suffer the consequences, maybe including not having my stupid genes copied to another human.
There have been frivolous lawsuits, definitely true. The scalding coffee was not. Other coffee vendors around the city were, at the highest temperature, 20 degrees Fahrenheit cooler than McDonald's coffee.
It absolutely was frivolous. Here's a tip, a "life lesson" if you will: If you buy something that is, by definition, hot, don't pour it all over your legs. I have no sympathy for people who do stupid things and then blame other people. There is no argument here. Period.
that's not a solution when you have applications processing information, and you switch over while your in the middle of processing requests. In my situation there isn't a single second the system isn't fielding 100's of requests. basicly it involves a hand shake where the client makes a requests and expects an answer, if you switch over the new system won't know the client is expecting an answer so you'd have to re engineer a black box system to do it somehow.
Ever heard of connection draining? You build systems with the expectation that they will fail. Any component at any given point in time should be expected to be broken, because it will be at some point. If your system can't handle bringing down a server for maintenance, then you have far bigger problems than picking a good OS. Good luck to you.