You should be able to set one up once, dump it to storage somewhere, and just image it onto the target hardware.
You do understand that OS X has all the prerequisites his app needs and ships pre-installed on the hardware, right? Why does he need to prepare his own image? What problem would that solve - it would add work to his roll-out process...
If you we're paying colo fees, the monthly fee for a MacMini would be a fraction of what you would pay for that 1U server. I can colo a MacMini for around $20-40/month, what does it cost to colo your 1U server?
That monthly fee makes a difference to many potential customers, and a 'pimped out' MacMini with n i7 CPU, 16 gig of RAM and two Drives is just a bit more than your 1U $700 box.
A colo provider can put 8 minis in a 1U shelf/tray and split the cost of a 1U server colo across 8 servers, hence monthly colo costs can be 1/8th the price of, say, a colo'd Dell PE1950 (or I liar). Sure, you can achieve the same 'magic' with 8 VMs on a 1U server, but some people really like the idea of dedicated hardware.
Look at the MacMini specs - latest processor, room for two drives (HDs, SSDs, or a mix), and 16 Gig of RAM. Couple those specifications with relatively low power demands and it makes a nice colo box. You can fit six or eight easily on a 1U shelf, more if you put them on their sides. The colo fees for a MacMini is a fraction of the price of a 1U colo server of more conventional design.
The MacMini is 'good enough' in most regards for a general purpose web server.
What was the President's plan for cuts? It was just as vague as the GOP plan.
The sequester was An Obama White House plan, as admitted by none other than Jay Carmey - check YouTube.
The President didn't compromise, he moved the goalposts after 'winning' over the Republicans and getting some $600BN in new tax revenue back in January, 2013. That is in addition to the new tax revenues that were included in the original budget plan that included Sequester.
You really should expand your sources of information...
Why was Obama's first serious discussion about avoiding sequester planned for the morning AFTER the cuts were in-place? Oh yeah, because he spent the last two weeks trying to convince everyone how bad it was going to be and to make sure HIS version of history was spoon-fed to his low-information supporters.
The SS cuts were paid for out of the regular budget - the gov't maintained anticipated revenues and did not accelerate the depletion of the so-called SS trust fund.
Baseline budgeting sets the expectations, and any increase below the baseline budgeted increase is sold as a cut. I've seen it - Democrats used to propose their own budgets with bigger numbers than the Republican budget proposals, then accused Republicans of not caring about (some group) because they wanted to increase spending by a smaller number than the Democrats proposed.
Obama and the Democrats did judt that when the Bush tax cuts were supposed to expire in 2010. When talking about the middle class tax cuts that were about to expire, it was an increase, when talking about the top tax rate cut that were about to expire, it was a return to the previous rate.
The problem is they DID spend the money and built out the infrastructure, they just did it years ago when gigabit Internet wasn't even being dreamed of and they are still paying off their initial investment. To rip out their current infrastructure before it is paid off is very, very expensive. Add to that the reality that very few people will pay over $100 for ISP services, and the decision not to offer it makes a lot of sense. Do you reasonably believe any for-profit company can offer gigabit ISP service for less than $100/month?
No, running the cable to your house is not the "expensive bit" - it is a big up-front cost, which is amortized over the life of the service. Actually providing the service, running the head-end units, the data centers, etc. is the expensive part. Trenching coax to your basement isn't the most "expensive bit."
You won't pay $200 for 100Mb/sec service, would you pay $200/mo for 1 Gb/sec service? Many people would not pay that much for a data plan - most people would pay that much for a data plan, heck very few people would.
If I had the option of gigabit internet in my neighborhood I would jump on it in an instant. So would many other people I suspect.
At what price-point? Current technology precludes any for-profit company offering gigabit service at a price comparable to the price you are currently paying for 10 Mb/sec service. Google is subsidizing gigabit service in Kansas CIty, they won't underwrite the cost of the physical plant nation-wide.
The cable/ISP companies are still paying off their current infrastructure, and to rip it out and replace it with gigabit-capable hardware would be very, very expensive. Could a cable company serve up gigabit residential access on their existing cable plant? Investing in a fiber plant to replace their current copper cable plant makes such an offering too expensive for almost all customers.
What is the point of TWC offering a service no one will pay for?
Many, many posters talk about services that cost $100/month or more as being "too expensive for most people" - how much do you imagine residential gigabit ISP service will cost?
In one small town Google is offering residential gigabit ISP service for a subsidized $70-80/month, what is the adoption rate? How many houses have the ability to get the service but choose not to?
Motorola invested $6BN in starting the service, after bankruptcy it sold all it's assets for $25M.
Google is subsidising the 1Gb/sec service at levels no company hoping to turn a profit on the venture would ever do... There's a reason they picked a small town to host this "showcase" ISP service.
Precisely. If I recall correctly, Comcasst offers 100 MB connectivity in my area for around $300 per month.
Because they didn't build out thier physical plant for every household to subscribe to that level of service - they scaled their network for lower bandwidth.
Google's 1 GB fiber connectivity is somewhere between $70 and $80 per month.
And is being offered below the cost of providing the service (subsidised) - that is not a sustainable business model for a for-profit company.
Do I want 100 MB, or even 1 GB? Oh, hell yeah. But can I pay more than around $100 per month for an überfast connection? Unfortunately, no. It's not lack of desire.
Offering a service people want is a no-brainer, offering a service people want but are unwilling to pay for is a non-starter. Motorola learned this with their "Iridium" Satellite phone service...
There is no way you can trust that a degree means a person knows their stuff.
My degree was assembled from coursework taken at various colleges, the degree program that collected them all together and granted me my BA required that all degree candidates hold a one hour discussion with an independent professor on a single topic related to the degree sought. The conversation is recorded, and it is a pass/fail "exam".
I think a similar "sanity test" could scale pretty well for an online college.
As a serious proposal, I don't see how this really changes anything.
Do these 50 equally populous states assign their electoral votes "winner take all" or proportionally?
If "winner take all" then very little will change, since that is how most current states award their electoral votes, and since each state has a number of electoral votes based on it's population the outcome wouldn't likely change.
If proportionally, then that would be a shift from the current model, but would the actual result really change? In the last election, President Obama won re-election with 51% of the popular vote, but won by a larger margin electorally. I would be very interested in seeing someone run this "50 Equal States" map against the county-by-county results from the 2012 election and see how the electoral results would change. (Obviously the popular vote wouldn't change.)
Rural areas don't have undue weight - how many rural states does it take to equal one OH, NY, FL, TX or CA? Electorally those states are monsters that decide who will be President - the rural areas do not have undue weight.
Slashdot Mirror
You do understand that OS X has all the prerequisites his app needs and ships pre-installed on the hardware, right? Why does he need to prepare his own image? What problem would that solve - it would add work to his roll-out process...
Perhaps, but they are a bit pricey IMHO, and as implemented currently have more limited storage options than the MacMini...
Colo fees for a MacMini range as low as $25/month when you supply the MacMini...
Four cores pegged is 100% CPU utilization.
If you we're paying colo fees, the monthly fee for a MacMini would be a fraction of what you would pay for that 1U server. I can colo a MacMini for around $20-40/month, what does it cost to colo your 1U server?
That monthly fee makes a difference to many potential customers, and a 'pimped out' MacMini with n i7 CPU, 16 gig of RAM and two Drives is just a bit more than your 1U $700 box.
Why people insist on using an HTPC oriented machine that is clearly not marketed at server-esque operations is beyond me.
apple markets them with OS X Server OS as a server solution.
Dell is very consistent with server internals and BUSINESS desktops/laptops.
Consumer products from Dell are a different matter....
A colo provider can put 8 minis in a 1U shelf/tray and split the cost of a 1U server colo across 8 servers, hence monthly colo costs can be 1/8th the price of, say, a colo'd Dell PE1950 (or I liar). Sure, you can achieve the same 'magic' with 8 VMs on a 1U server, but some people really like the idea of dedicated hardware.
Low power consumption and form factor together make for very affordable colo fees, a fraction of the price for a more conventional 1U server.
Look at the MacMini specs - latest processor, room for two drives (HDs, SSDs, or a mix), and 16 Gig of RAM. Couple those specifications with relatively low power demands and it makes a nice colo box. You can fit six or eight easily on a 1U shelf, more if you put them on their sides. The colo fees for a MacMini is a fraction of the price of a 1U colo server of more conventional design.
The MacMini is 'good enough' in most regards for a general purpose web server.
What was the President's plan for cuts? It was just as vague as the GOP plan.
The sequester was An Obama White House plan, as admitted by none other than Jay Carmey - check YouTube.
The President didn't compromise, he moved the goalposts after 'winning' over the Republicans and getting some $600BN in new tax revenue back in January, 2013. That is in addition to the new tax revenues that were included in the original budget plan that included Sequester.
You really should expand your sources of information...
Why was Obama's first serious discussion about avoiding sequester planned for the morning AFTER the cuts were in-place? Oh yeah, because he spent the last two weeks trying to convince everyone how bad it was going to be and to make sure HIS version of history was spoon-fed to his low-information supporters.
The SS cuts were paid for out of the regular budget - the gov't maintained anticipated revenues and did not accelerate the depletion of the so-called SS trust fund.
Baseline budgeting sets the expectations, and any increase below the baseline budgeted increase is sold as a cut. I've seen it - Democrats used to propose their own budgets with bigger numbers than the Republican budget proposals, then accused Republicans of not caring about (some group) because they wanted to increase spending by a smaller number than the Democrats proposed.
Obama and the Democrats did judt that when the Bush tax cuts were supposed to expire in 2010. When talking about the middle class tax cuts that were about to expire, it was an increase, when talking about the top tax rate cut that were about to expire, it was a return to the previous rate.
The problem is they DID spend the money and built out the infrastructure, they just did it years ago when gigabit Internet wasn't even being dreamed of and they are still paying off their initial investment. To rip out their current infrastructure before it is paid off is very, very expensive. Add to that the reality that very few people will pay over $100 for ISP services, and the decision not to offer it makes a lot of sense. Do you reasonably believe any for-profit company can offer gigabit ISP service for less than $100/month?
No, running the cable to your house is not the "expensive bit" - it is a big up-front cost, which is amortized over the life of the service. Actually providing the service, running the head-end units, the data centers, etc. is the expensive part. Trenching coax to your basement isn't the most "expensive bit."
You won't pay $200 for 100Mb/sec service, would you pay $200/mo for 1 Gb/sec service? Many people would not pay that much for a data plan - most people would pay that much for a data plan, heck very few people would.
At what price-point? Current technology precludes any for-profit company offering gigabit service at a price comparable to the price you are currently paying for 10 Mb/sec service. Google is subsidizing gigabit service in Kansas CIty, they won't underwrite the cost of the physical plant nation-wide.
The cable/ISP companies are still paying off their current infrastructure, and to rip it out and replace it with gigabit-capable hardware would be very, very expensive. Could a cable company serve up gigabit residential access on their existing cable plant? Investing in a fiber plant to replace their current copper cable plant makes such an offering too expensive for almost all customers.
What is the point of TWC offering a service no one will pay for?
Many, many posters talk about services that cost $100/month or more as being "too expensive for most people" - how much do you imagine residential gigabit ISP service will cost?
In one small town Google is offering residential gigabit ISP service for a subsidized $70-80/month, what is the adoption rate? How many houses have the ability to get the service but choose not to?
Yeah - Motorola called it "Iridium"...
Motorola invested $6BN in starting the service, after bankruptcy it sold all it's assets for $25M.
Google is subsidising the 1Gb/sec service at levels no company hoping to turn a profit on the venture would ever do... There's a reason they picked a small town to host this "showcase" ISP service.
You can have "neighbor" IP addresses with redundancy (and diversity)...
Because they didn't build out thier physical plant for every household to subscribe to that level of service - they scaled their network for lower bandwidth.
And is being offered below the cost of providing the service (subsidised) - that is not a sustainable business model for a for-profit company.
Offering a service people want is a no-brainer, offering a service people want but are unwilling to pay for is a non-starter. Motorola learned this with their "Iridium" Satellite phone service...
My degree was assembled from coursework taken at various colleges, the degree program that collected them all together and granted me my BA required that all degree candidates hold a one hour discussion with an independent professor on a single topic related to the degree sought. The conversation is recorded, and it is a pass/fail "exam".
I think a similar "sanity test" could scale pretty well for an online college.
It allows motivated students to study on their own schedule.
It also allows unmotivated students to accumulate massive amounts of debt...
Was this deployment large enough to qualify for Software Assurance with it's Win7 downgrade option?
I assume the school has Education Assurance, which also allows for Win7 downgrade.
And as an art project, it's interesting.
As a serious proposal, I don't see how this really changes anything.
Do these 50 equally populous states assign their electoral votes "winner take all" or proportionally?
If "winner take all" then very little will change, since that is how most current states award their electoral votes, and since each state has a number of electoral votes based on it's population the outcome wouldn't likely change.
If proportionally, then that would be a shift from the current model, but would the actual result really change? In the last election, President Obama won re-election with 51% of the popular vote, but won by a larger margin electorally. I would be very interested in seeing someone run this "50 Equal States" map against the county-by-county results from the 2012 election and see how the electoral results would change. (Obviously the popular vote wouldn't change.)
Rural areas don't have undue weight - how many rural states does it take to equal one OH, NY, FL, TX or CA? Electorally those states are monsters that decide who will be President - the rural areas do not have undue weight.