It's not exactly mooching, I pay them the $180 per year it costs for my line. For the main line my dad is getting more pool minutes for significantly less than he used to pay. He gets like 6,000 minutes for less than he used to pay for 2,000 minutes and now he has no overage charges which would often double his bill. My wife uses a couple hundred minutes a month of which probably 90% are mobile to mobile which doesn't come out of the pool minutes. Oh yeah and the ? is definitely warranted, the reason my employer picks up my service is that I am oncall 1 week a month, but I probably average 4 calls per year so it works out well I think =)
10GbE is reasonable today. Quadrics has a 96 port switch for only $300 per port and adapters are only $1K (eg NC510C from HP or the PXLA8591CX4 from Intel). Sure you can get 2Gb FC for around this same price point but 4Gb is significantly more. Brocade wants $500 per port with no SFP's and only 32 of the 64 ports enabled for the Silkwork 4900, fully configured you're at greater than $1,500 per port. Qlogic is similar for the SANbox 9000.
or spend $60/mo for the same level of service from Speakeasy without a landline
You keep using that word, I do not think it means what you think it means. Now if you mean the same advertised bandwidth then it's possible, but they aren't remotely the same service either from a TOS perspective or from a customer service perspective.
I spend $15/month ($9.99+fees and taxes) for my VoIP landline with unlimited in state calling, $15/month for my wifes phone on my folks family share plan and $0 on my cell (provided by work). That adds up to a grand total of $360 for total communications budget. My company also picks up my internet. I guess the telcos must dislike people like me, which is great because the feeling is mutual =)
Nah I see this as a lower cost way to distribute from the FC switch back to the storage users eg the servers. Most storage is also presented by some kind of storage array, very very little is JBOD presented directly by a switch. This is mostly due to the lack of management of JBOD as well as the fact that the performance improvement of placing a bunch of intelligent cache in front of the storage pool is huge.
Latency and bandwidth are comparable for copper and fiber ethernet solutions today, the drawback to copper is you need to be within 15m of the switch. This isn't so bad in a small datacenter but in a larger facility you would either need switches all over the place (preferably in 2's for redundant path) or you'd need to go fiber which eliminates a good percentage of the cost savings. FiberChannel used to have copper as a low cost option but it's not there in the 4Gb world and even in the 2Gb space it was so exotic that there was almost no cost savings due to lack of economies of scale.
Huh? Our piddly 150 spindle SAN could keep a 10Gb link saturated no problem if we had a workload that needed it. In fact that's less than 7MB/s per drive, about one tenth of what these drives are capable of for bulk reads or about one fifth for bulk writes. Even for totally random I/O 150 spindles could probably keep that sized pipe filled.
According to this netapp paper even NFS over 10GbE is lower latency than 4Gb FC. I imagine if the processing overhead isn't too high or offload cards become available then this would be significantly faster than 4Gb FC. As far as bandwidth 10>4 even with the overhead of ethernet framing, especially if you can stand the latency of packing two or more FC frames into an ethernet jumbo frames.
As long as a server is within the distance limit of copper, 10GE is about 3-4x cheaper then even 2Gb FC. We've also had a heck of a lot more stability out of our 6500 series switches then we have out of our 9140's and the 9500's are extremely expensive if you have a need for under 3 cards worth of ports.
Any DC running ESX 3.5 on certified hardware can shutdown and powerup host servers on demand, it's a new experimental feature. Depending on your profile you can shutdown during business hours to just essential servers and add capacity for nightly batches or have servers online during the day to handle user demand and scale back for nighttime lulls.
The problem of internet optimization stopped being a research subject years ago and is more of a business problem. Peering relationships and lowest cost routing mean that traffic will often travel a suboptimal route from a networking perspective because it is the best route from a providers financial perspective.
Everything recent from Linksys is IPv6 capable, not sure about other vendors. Also Cisco gear old enough to not support IPv6 is so old you can't even buy support for it anymore, businesses should not be running on that kind of gear anyway.
Well, "recent desktop" in the Microsoft world means Windows 2000+. Standards compliant IPv6 wasn't in the standard Linux kernel until 2.6 so you'd need something of the Fedora Core 2 vintage or newer, that's still 3+ years old. Now the problem is more at the app level since many app vendors never bother to build their app against the IPv6 libs, let alone test IPv6 functionality. Of course if the fed had made it a functional requirement rather than a capability requirement you can bet that all the large software packages would be IPv6 certified by now =)
There are plenty of digital radio protocols that use a very well defined slice of bandwidth, without any more bleed over than traditional AM or FM radio analog broadcasts
Bad example since as a percentage of channel width FM has about the widest guard area of any communications method. Irrespective of that the 2.4Ghz spectrum is ISM so as long as they meet FCC regs for total power and don't bleed too much into non-ISM spectrum they can be as rude as they want to be.
They just need to reallocate some blocks, MIT has a Class A, 4 Class B's and a host of Class C's. That's enough to get most countries online. HP has TWO class A's thanks to the consumption of Compaq/DEC, ham's have a class A as does Xerox and Halliburton. Combined that makes for 100+ million additional IP's to become available if a couple large organizations simply re-ip. Now I know a large scale re-ip can be painful, but they have years to do it if they start now.
Yeah and now everyone will be screwed. The local franchise agreement is the only reason I got cable internet access in a timely manner. Adelphia was in be-acquired mode and wasn't about to put any expenses on the books but our city council put pressure on them to live up to the terms of their franchise agreement and finish the rollout on time. If it weren't for the franchise agreement we would have been screwed and had no broadband for at least a year or two because TW wouldn't have done the rollout right after the acquisition.
Well the theory about limiting current works until there is something wrong with an adjacent high voltage line and current is either transfered or induced powerful enough to cause a spark. This is known to have taken down several airliners over the years. If you were to make an intrinsically safe circuit that was powered by light it would be MUCH more difficult for such a fault to occur. I'm not an electrical engineer but assuming JDS can design such a circuit then it would in fact be more safe than running a conductor from the wiring bay into the fuel tanks.
Damn, it showed up in preview, here's another attempt. They aren't exactly expensive at $3.36 ($2.69 in quantity) for a 1.5ft cable. Hell even a 50' cable is only $12 =) Of course they'd be a hell of a lot more expensive at BB or CC but that's what you get for instant gratification and expensive realestate.
The speed of SATA isn't wasted in theory, SATA allows for a tree topology up to 3 levels deep and up to 128 devices. This has never been a problem in the past because the physical specs didn't allow for SATA to be the host connection so since you were bridging to a separate physical transport you would hide the details of the drives from the host. With SATA 1.2 you now have eSATA which is usable as a host connection. I expect to see cheap storage shelves that are nothing but a SATA->eSATA hub/mux and power distribution box. This would allow you to use one HBA for a few external enclosures just like with SCSI.
Eh, I'm not getting blasted most of the time, my teens are well behind me. On the one occasion that I did get a call when I was not in a condition to be working I plainly said so and reminded my boss I was not on call that week. Doing a password reset after 1-3 beers is not something that has any potential to harm my employer and allows me to actually have a life.
For a sysadmin continuous access can actually enhance the balance. I don't have to drive in for every little problem because I have integrated lights out on all my servers so only a true hardware failure requires a trip into the office on off hours. Having a smartphone makes it even better because now instead of having to have a laptop with me or doing it from my home workstation I can now do many non-complicated tasks from anywhere (such as at the bar).
Slashdot Mirror
It's not exactly mooching, I pay them the $180 per year it costs for my line. For the main line my dad is getting more pool minutes for significantly less than he used to pay. He gets like 6,000 minutes for less than he used to pay for 2,000 minutes and now he has no overage charges which would often double his bill. My wife uses a couple hundred minutes a month of which probably 90% are mobile to mobile which doesn't come out of the pool minutes. Oh yeah and the ? is definitely warranted, the reason my employer picks up my service is that I am oncall 1 week a month, but I probably average 4 calls per year so it works out well I think =)
10GbE is reasonable today. Quadrics has a 96 port switch for only $300 per port and adapters are only $1K (eg NC510C from HP or the PXLA8591CX4 from Intel). Sure you can get 2Gb FC for around this same price point but 4Gb is significantly more. Brocade wants $500 per port with no SFP's and only 32 of the 64 ports enabled for the Silkwork 4900, fully configured you're at greater than $1,500 per port. Qlogic is similar for the SANbox 9000.
or spend $60/mo for the same level of service from Speakeasy without a landline
You keep using that word, I do not think it means what you think it means. Now if you mean the same advertised bandwidth then it's possible, but they aren't remotely the same service either from a TOS perspective or from a customer service perspective.
I spend $15/month ($9.99+fees and taxes) for my VoIP landline with unlimited in state calling, $15/month for my wifes phone on my folks family share plan and $0 on my cell (provided by work). That adds up to a grand total of $360 for total communications budget. My company also picks up my internet. I guess the telcos must dislike people like me, which is great because the feeling is mutual =)
Nah I see this as a lower cost way to distribute from the FC switch back to the storage users eg the servers. Most storage is also presented by some kind of storage array, very very little is JBOD presented directly by a switch. This is mostly due to the lack of management of JBOD as well as the fact that the performance improvement of placing a bunch of intelligent cache in front of the storage pool is huge.
Latency and bandwidth are comparable for copper and fiber ethernet solutions today, the drawback to copper is you need to be within 15m of the switch. This isn't so bad in a small datacenter but in a larger facility you would either need switches all over the place (preferably in 2's for redundant path) or you'd need to go fiber which eliminates a good percentage of the cost savings. FiberChannel used to have copper as a low cost option but it's not there in the 4Gb world and even in the 2Gb space it was so exotic that there was almost no cost savings due to lack of economies of scale.
Huh? Our piddly 150 spindle SAN could keep a 10Gb link saturated no problem if we had a workload that needed it. In fact that's less than 7MB/s per drive, about one tenth of what these drives are capable of for bulk reads or about one fifth for bulk writes. Even for totally random I/O 150 spindles could probably keep that sized pipe filled.
According to this netapp paper even NFS over 10GbE is lower latency than 4Gb FC. I imagine if the processing overhead isn't too high or offload cards become available then this would be significantly faster than 4Gb FC. As far as bandwidth 10>4 even with the overhead of ethernet framing, especially if you can stand the latency of packing two or more FC frames into an ethernet jumbo frames.
As long as a server is within the distance limit of copper, 10GE is about 3-4x cheaper then even 2Gb FC. We've also had a heck of a lot more stability out of our 6500 series switches then we have out of our 9140's and the 9500's are extremely expensive if you have a need for under 3 cards worth of ports.
Any DC running ESX 3.5 on certified hardware can shutdown and powerup host servers on demand, it's a new experimental feature. Depending on your profile you can shutdown during business hours to just essential servers and add capacity for nightly batches or have servers online during the day to handle user demand and scale back for nighttime lulls.
The problem of internet optimization stopped being a research subject years ago and is more of a business problem. Peering relationships and lowest cost routing mean that traffic will often travel a suboptimal route from a networking perspective because it is the best route from a providers financial perspective.
Everything recent from Linksys is IPv6 capable, not sure about other vendors. Also Cisco gear old enough to not support IPv6 is so old you can't even buy support for it anymore, businesses should not be running on that kind of gear anyway.
Well, "recent desktop" in the Microsoft world means Windows 2000+. Standards compliant IPv6 wasn't in the standard Linux kernel until 2.6 so you'd need something of the Fedora Core 2 vintage or newer, that's still 3+ years old. Now the problem is more at the app level since many app vendors never bother to build their app against the IPv6 libs, let alone test IPv6 functionality. Of course if the fed had made it a functional requirement rather than a capability requirement you can bet that all the large software packages would be IPv6 certified by now =)
There are plenty of digital radio protocols that use a very well defined slice of bandwidth, without any more bleed over than traditional AM or FM radio analog broadcasts
Bad example since as a percentage of channel width FM has about the widest guard area of any communications method. Irrespective of that the 2.4Ghz spectrum is ISM so as long as they meet FCC regs for total power and don't bleed too much into non-ISM spectrum they can be as rude as they want to be.
They just need to reallocate some blocks, MIT has a Class A, 4 Class B's and a host of Class C's. That's enough to get most countries online. HP has TWO class A's thanks to the consumption of Compaq/DEC, ham's have a class A as does Xerox and Halliburton. Combined that makes for 100+ million additional IP's to become available if a couple large organizations simply re-ip. Now I know a large scale re-ip can be painful, but they have years to do it if they start now.
Yeah and now everyone will be screwed. The local franchise agreement is the only reason I got cable internet access in a timely manner. Adelphia was in be-acquired mode and wasn't about to put any expenses on the books but our city council put pressure on them to live up to the terms of their franchise agreement and finish the rollout on time. If it weren't for the franchise agreement we would have been screwed and had no broadband for at least a year or two because TW wouldn't have done the rollout right after the acquisition.
Well the theory about limiting current works until there is something wrong with an adjacent high voltage line and current is either transfered or induced powerful enough to cause a spark. This is known to have taken down several airliners over the years. If you were to make an intrinsically safe circuit that was powered by light it would be MUCH more difficult for such a fault to occur. I'm not an electrical engineer but assuming JDS can design such a circuit then it would in fact be more safe than running a conductor from the wiring bay into the fuel tanks.
The Ford Escape Hybrid already gets 34MPG city, with a little work all but the heaviest SUV's and trucks should be able to hit 35MPG.
Damn, it showed up in preview, here's another attempt. They aren't exactly expensive at $3.36 ($2.69 in quantity) for a 1.5ft cable. Hell even a 50' cable is only $12 =) Of course they'd be a hell of a lot more expensive at BB or CC but that's what you get for instant gratification and expensive realestate.
The speed of SATA isn't wasted in theory, SATA allows for a tree topology up to 3 levels deep and up to 128 devices. This has never been a problem in the past because the physical specs didn't allow for SATA to be the host connection so since you were bridging to a separate physical transport you would hide the details of the drives from the host. With SATA 1.2 you now have eSATA which is usable as a host connection. I expect to see cheap storage shelves that are nothing but a SATA->eSATA hub/mux and power distribution box. This would allow you to use one HBA for a few external enclosures just like with SCSI.
If you're worried about your optical cables then get armored ones like these from monoprice.
USB3 is pointless, use eSATA instead. FW3200 will have a use as a transport for HD video between non-PC machines.
Eh, I'm not getting blasted most of the time, my teens are well behind me. On the one occasion that I did get a call when I was not in a condition to be working I plainly said so and reminded my boss I was not on call that week. Doing a password reset after 1-3 beers is not something that has any potential to harm my employer and allows me to actually have a life.
It wasn't quite from the beach but I've done password resets from the park on a nice summer day =)
For a sysadmin continuous access can actually enhance the balance. I don't have to drive in for every little problem because I have integrated lights out on all my servers so only a true hardware failure requires a trip into the office on off hours. Having a smartphone makes it even better because now instead of having to have a laptop with me or doing it from my home workstation I can now do many non-complicated tasks from anywhere (such as at the bar).