For 2 processor, 4 core, systems AMD did not have a massive performance advantage over Intel's NetBurst Xeons. In spite of Intel's shared bus architechture, for some applications Intel was better then AMD. Even when an AMD system was faster, Intel systems were often considerably cheaper. Blame Dell. AMD's big gains were made in the 4 processor, 8 core, systems.
With the Core 2 server chips, Woodcrest, Intel is the clear winner in 2 and 4 core systems. However Woodcrest can only be used in 2 and 4 core systems. For 8 core, (4 socket/processor) systems Intel still uses the old (failed) NetBurst Xeons. While AMD will lose the 4 core battle, they will maintain their 8 (and 16) core market share. The upshot is that AMD's share of the server market will go down, but not as much as you expect.
children exposed to the more old fashioned fairy tails, stories, and myths where people actually die, grow up to be better adjusted than children who get the more sugar coated "Disney" style stories
A friend of mine recently started coaching baseball for 9 and 10 year olds (little league??). He said one of the most frustrating part of coaching has been that the kids can't handle failure of any kind. He has no idea how he can teach them how to handle loseing.
Purchasing a piece of property is always less expensive than renting.
Incorrect. Purchasing is usually less expensive then renting. However in some markets properties are renting at an effective loss (see sibling). Many property owners are living the property ownership myth and haven't done the math. Purchasing also increases risk (i.e. the roof suddenly needs replaceing).
All property, baring buying in the midst of a housing bubble, appreciates in value.
Most of North America is in the midst of a housing bubble. Babyboomers are already looking to sell their large homes for smaller ones so that they can use the equity to retire (travel, etc). This will soften the housing market. While a total collapse in the next decade is unlikely (but not impossible) buying property as a form of investment is foolish. For most people who do not wish to make property investment a full time job, a simple money market fund is almost certainly a better investment: There is less risk, and the short term returns are often similar.
Even if you only by a townhome or condo, purchasing is a much better choice than renting or living on campus.
No. Purchasing may be a better choice. What is the deffinition of better? Will purchasing save money? What is the risk of purchasing versus an insured investment? Will the potential home owner have the time and money to maintain the property? This is a complex equation with many inputs. For some owning is far and away a better choice. For others renting may be a better choice.
Even in terms of clear dollars renting may be be a financially better option. Saving up for a larger down payment may be better option then buying today. Calculating these numbers to determine the best time to buy is not trivial and requires a number of assumptions and flat out guesses.
Home ownership as an investment is a terrible myth. Paying a mortgage versus paying rent is often financially better. However simple home ownership versus renting is not clear cut. There are many costs associated with owning a home from the manageble, predictable costs such as taxes, to large, unexpected costs such as major repairs. There are quality of life benefits to owning your own home, but quality of life does not translate into money.
Owning and manageing a number of properties as a full time job is a different matter. That can and often does translate into real wealth.
Conroe (Duo-tech) based Xeon chips will be coming out in September.
The Xeon 5100 series chips are Core 2 (Conroe) based. These chips are code named Woodcrest and started shipping in June. See the page 2 of the article. You must be thinking of the older Intel roadmap.
The snapshot and clone features in VMware v5 beat VPC 2004. VMware v5 is a recent product VPC 2004 is not. In general VPC 2004 ran windows guests faster then VMware v4 (four). The difference was marginal. I have not tested the speed of VPC 2004 against VMware v5. VMware runs all non-windows guests faster then VPC v4. (In some cases infinitely faster as VPC had trouble with certain versions of Linux and FreeBSD.) You can run a kernel level debugger such as SoftICE under VPC. SoftICE and Vmware have/had issues. IIRC VPC has no opengl/directx guest support. I doubt that you will ever see that feature in either product. The new virtualization instructions in Intel and AMD processors may change that, but I would not count on it.
I use VMware daily. VMware support other guest operating systems better then VPC. But the big winner is VMware's management features. The snapshot managment, cloneing and templating are wonderfull.
Re:SSL for the first hop isn't enough. (Try SMIME)
on
What's In Your Inbox?
·
· Score: 2, Informative
SMIME and OpenPGP are both standards. What makes you think that a new standard is going to be adopted any faster?
The barrier to wide spread encryption is key management. Key management is hard and there is no perfect solution. A full fledged PKI is would be easiest for most users and could provide sealess encryption. But who manages the PKI? Will a government be able to recover the key with a search warrent? There are some serious privacy problems to overcome. PGP/GPG gives users full control over their keys which is great from a privacy point of view, but the management is hard. Most users don't want to manage their on keys. Worse, many users won't understand how to properly manage keys which will make the system essentially useless. (Many current PGP/GPG users mismanage their keys and they should know better.) PGP/GPG also has key distribution and trust issues.
(Although with only four drives there is no difference.)
There is a difference with only four drives. A stripe of mirrors, RAID 1+0/10, can sustain two failed drives provided the second drive is part of the second pair. The result of the two drive failure is a stripe of two drives. A mirror of stipes, RAID 0+1, can only survive a single drive failure. When a single drive fails the result is a stripe of two drives and a third drive doing nothing. Draw a picture and it will make sense.
This article is terrible and contains no real facts. It is full of buzz words for management.
Go read Schneier. It may seem that most of what he writes is not security related, but it usually it. All forms of security are related. It is important to look at the big security picture and not concentrate on the individual technology pieces.
TFA is a terrible sales pitch (complete with CIO buzzwords) for Fortinet's products.
Last year we were testing one of the smaller Fortinet "firewalls." It was easy to crash the Fortinet box and the protocol/data scanners with a boring network fuzzer. (i.e. we sent bad data at the box) Given time I am sure that we could have exploited the crashes. But, as that was not our job, we moved on to testing better products.
These all in one (adaptive filtering with super duper special proxies) traffic scanning firewalls rely on software that perfectly understands the higher level network protocols. This is not an easy task. Consider the different webbrowser bugs and misfeatures that web pages need to code around. The firewall software needs to understand and allow for these bugs as well. The traffic scanning software is complex with leads to bugs. Some of those bugs will certainly be exploitable.
Anyone relying on the sneakiness of skype is in for a world of hurt. Skype traffic may be hard to detect automatically, but it is almost trivial to detect with a little human analysis.
It is a mistake to call a few hunderd a month "doing well." Certainly that is not a living wage in most (all ??) of the western world.
My cousin does the same (buying used locally, and selling on ebay for a small profit). While he makes a couple hundred a month, he could easily make more money for less work. He considers it a hobby that pays money. It can be argued that the auctions are a form of gambling with a small risk and reward. There is a thrill when a bidding war starts. There is a risk that a buyer won't pay. He finds it fun. I suspect that there are many other ebay sellers just like him and your friends mother.
I mean no disrespect. This form of money making has a few advantages for a stay at home parent. There are few obligations and the work schedule is very flexible. And while the small profit is not much extra money is never a bad thing.
Re:A New Core Class in College?
on
PGP & GPG
·
· Score: 1
Read and understand Schneier's book Secrets and Lies. The problem is described in the preface.
Re:A New Core Class in College?
on
PGP & GPG
·
· Score: 1
In my experience many/. posters think they understand cryptography but don't. (In no way do I mean to imply that you don't understand cryptography. Nor am I a master on the subject.) Many software developers don't understand cryptography either. This includes many developers coding encryption software. There is far more to good cryptography then just getting the algorithm correct. Secure cryptography requires a complete system of software, hardware, and people. Many geek types look at the software and the math behind it. They think that is all that matters and forget about the rest of the system. The software is often the easiest part of the system.
I agree with the gp and suspect that the number of Internet users who understand cryptography at even just a high level is probably in the order of 1e5. Cryptography is hard stuff.
Small offices don't have less office politics. As soon as you have two people together in one office you have politics. While it may be true that small businesses can't afford the politics, that dosen't mean there aren't any. Small offices still suffer from arbitrary promotions and pay raises, and the boss's incompotent son or daughter "working" durring the summer.
I don't understand the adversion to office politics. Politics are just something else that smart people can hack.
Most data centers of this size will have connections to two different power grids
No they won't. It dosen't make sense. You don't know what you are talking about. The North American power grid is all interconnected. There is no second grid to connect to. Even if there were a second grid, running power lines is insanely expensive. Generators are cheaper.
I don't care how much generator capacity you have, it's most likely not enough to last longer than a couple of days without power.
What does this mean? The generators could effectively run forever. (at any rate far longer then "a couple of day.") The generators used are typically diesel powered. As long as there is fuel in the tank the generator will keep running. It is not hard to keep fuel tanks full. These data centers will have enough generators so that one or more can be offline for maintanance.
Parallels is not in the same league as VMware. The Parallels Workstation/Desktop product is similar to on old version of VMware workstation. More importantly Parallels dosen't have any products for servers. VMware has two powerfull server products and software to properly manage it all.
Power grid reliability is not a big concern. Data centres of this size will have backup generators. Taxes aren't going to be an issue either. These data centres will be given sweatheart tax deals, no interest loans, and other incentives. The states and counties will give out these incentives because the data centres will bring so called "high tech jobs."
We need more guilt free slack. I should not feel guilty for only working my contracted 40 hours a week. I will work the long weeks when required, but that should not be every week. I should also be able to take my coffee and lunch breaks, and have a quick chat around the "water cooler." The alternative to these little breaks is longer sneak breaks taken when no one is looking. These breaks are not as productive as the worker is worried about being caught and can't take 10 minutes to relax.
I was referring to the run-on sentences and poor structure of your post. I do not accept the argument: "it is only a/. post." I expect better from someone who has an MA in Education.
Yet you must have failed English. I have a doctorate in literature in comparison to you and my English is terrible. No wonder the system is in jeopardy: So called Masters of Education can't type a coherent paragraph.
Slashdot Mirror
For 2 processor, 4 core, systems AMD did not have a massive performance advantage over Intel's NetBurst Xeons. In spite of Intel's shared bus architechture, for some applications Intel was better then AMD. Even when an AMD system was faster, Intel systems were often considerably cheaper. Blame Dell. AMD's big gains were made in the 4 processor, 8 core, systems.
With the Core 2 server chips, Woodcrest, Intel is the clear winner in 2 and 4 core systems. However Woodcrest can only be used in 2 and 4 core systems. For 8 core, (4 socket/processor) systems Intel still uses the old (failed) NetBurst Xeons. While AMD will lose the 4 core battle, they will maintain their 8 (and 16) core market share. The upshot is that AMD's share of the server market will go down, but not as much as you expect.
A friend of mine recently started coaching baseball for 9 and 10 year olds (little league??). He said one of the most frustrating part of coaching has been that the kids can't handle failure of any kind. He has no idea how he can teach them how to handle loseing.
Incorrect. Purchasing is usually less expensive then renting. However in some markets properties are renting at an effective loss (see sibling). Many property owners are living the property ownership myth and haven't done the math. Purchasing also increases risk (i.e. the roof suddenly needs replaceing).
Most of North America is in the midst of a housing bubble. Babyboomers are already looking to sell their large homes for smaller ones so that they can use the equity to retire (travel, etc). This will soften the housing market. While a total collapse in the next decade is unlikely (but not impossible) buying property as a form of investment is foolish. For most people who do not wish to make property investment a full time job, a simple money market fund is almost certainly a better investment: There is less risk, and the short term returns are often similar.
No. Purchasing may be a better choice. What is the deffinition of better? Will purchasing save money? What is the risk of purchasing versus an insured investment? Will the potential home owner have the time and money to maintain the property? This is a complex equation with many inputs. For some owning is far and away a better choice. For others renting may be a better choice.
Even in terms of clear dollars renting may be be a financially better option. Saving up for a larger down payment may be better option then buying today. Calculating these numbers to determine the best time to buy is not trivial and requires a number of assumptions and flat out guesses.
Home ownership as an investment is a terrible myth. Paying a mortgage versus paying rent is often financially better. However simple home ownership versus renting is not clear cut. There are many costs associated with owning a home from the manageble, predictable costs such as taxes, to large, unexpected costs such as major repairs. There are quality of life benefits to owning your own home, but quality of life does not translate into money.
Owning and manageing a number of properties as a full time job is a different matter. That can and often does translate into real wealth.
Millionaires borrow lots of money. But only went it makes sense. Sometimes borrowing money makes money.
That sort of trip will just increase the sense of isolation and the depression. Unless you are an extrovert travelling by yourself is not fun.
The Xeon 5100 series chips are Core 2 (Conroe) based. These chips are code named Woodcrest and started shipping in June. See the page 2 of the article. You must be thinking of the older Intel roadmap.
CVE-2006-1173 is vulnerability in Sendmail. Debian uses Exim by default. Why would CVE-2006-1173 cause you to ditch debian?
The snapshot and clone features in VMware v5 beat VPC 2004. VMware v5 is a recent product VPC 2004 is not. In general VPC 2004 ran windows guests faster then VMware v4 (four). The difference was marginal. I have not tested the speed of VPC 2004 against VMware v5. VMware runs all non-windows guests faster then VPC v4. (In some cases infinitely faster as VPC had trouble with certain versions of Linux and FreeBSD.) You can run a kernel level debugger such as SoftICE under VPC. SoftICE and Vmware have/had issues. IIRC VPC has no opengl/directx guest support. I doubt that you will ever see that feature in either product. The new virtualization instructions in Intel and AMD processors may change that, but I would not count on it.
I use VMware daily. VMware support other guest operating systems better then VPC. But the big winner is VMware's management features. The snapshot managment, cloneing and templating are wonderfull.
SMIME and OpenPGP are both standards. What makes you think that a new standard is going to be adopted any faster?
The barrier to wide spread encryption is key management. Key management is hard and there is no perfect solution. A full fledged PKI is would be easiest for most users and could provide sealess encryption. But who manages the PKI? Will a government be able to recover the key with a search warrent? There are some serious privacy problems to overcome. PGP/GPG gives users full control over their keys which is great from a privacy point of view, but the management is hard. Most users don't want to manage their on keys. Worse, many users won't understand how to properly manage keys which will make the system essentially useless. (Many current PGP/GPG users mismanage their keys and they should know better.) PGP/GPG also has key distribution and trust issues.
(Although with only four drives there is no difference.)
There is a difference with only four drives. A stripe of mirrors, RAID 1+0/10, can sustain two failed drives provided the second drive is part of the second pair. The result of the two drive failure is a stripe of two drives. A mirror of stipes, RAID 0+1, can only survive a single drive failure. When a single drive fails the result is a stripe of two drives and a third drive doing nothing. Draw a picture and it will make sense.
I can't think of any advantage to RAID 0+1.
This article is terrible and contains no real facts. It is full of buzz words for management.
Go read Schneier. It may seem that most of what he writes is not security related, but it usually it. All forms of security are related. It is important to look at the big security picture and not concentrate on the individual technology pieces.
TFA is a terrible sales pitch (complete with CIO buzzwords) for Fortinet's products.
Last year we were testing one of the smaller Fortinet "firewalls." It was easy to crash the Fortinet box and the protocol/data scanners with a boring network fuzzer. (i.e. we sent bad data at the box) Given time I am sure that we could have exploited the crashes. But, as that was not our job, we moved on to testing better products.
These all in one (adaptive filtering with super duper special proxies) traffic scanning firewalls rely on software that perfectly understands the higher level network protocols. This is not an easy task. Consider the different webbrowser bugs and misfeatures that web pages need to code around. The firewall software needs to understand and allow for these bugs as well. The traffic scanning software is complex with leads to bugs. Some of those bugs will certainly be exploitable.
Anyone relying on the sneakiness of skype is in for a world of hurt. Skype traffic may be hard to detect automatically, but it is almost trivial to detect with a little human analysis.
It is a mistake to call a few hunderd a month "doing well." Certainly that is not a living wage in most (all ??) of the western world.
My cousin does the same (buying used locally, and selling on ebay for a small profit). While he makes a couple hundred a month, he could easily make more money for less work. He considers it a hobby that pays money. It can be argued that the auctions are a form of gambling with a small risk and reward. There is a thrill when a bidding war starts. There is a risk that a buyer won't pay. He finds it fun. I suspect that there are many other ebay sellers just like him and your friends mother.
I mean no disrespect. This form of money making has a few advantages for a stay at home parent. There are few obligations and the work schedule is very flexible. And while the small profit is not much extra money is never a bad thing.
Read and understand Schneier's book Secrets and Lies. The problem is described in the preface.
In my experience many /. posters think they understand cryptography but don't. (In no way do I mean to imply that you don't understand cryptography. Nor am I a master on the subject.) Many software developers don't understand cryptography either. This includes many developers coding encryption software. There is far more to good cryptography then just getting the algorithm correct. Secure cryptography requires a complete system of software, hardware, and people. Many geek types look at the software and the math behind it. They think that is all that matters and forget about the rest of the system. The software is often the easiest part of the system.
I agree with the gp and suspect that the number of Internet users who understand cryptography at even just a high level is probably in the order of 1e5. Cryptography is hard stuff.
exactly.
Small offices don't have less office politics. As soon as you have two people together in one office you have politics. While it may be true that small businesses can't afford the politics, that dosen't mean there aren't any. Small offices still suffer from arbitrary promotions and pay raises, and the boss's incompotent son or daughter "working" durring the summer.
I don't understand the adversion to office politics. Politics are just something else that smart people can hack.
Most data centers of this size will have connections to two different power grids
No they won't. It dosen't make sense. You don't know what you are talking about. The North American power grid is all interconnected. There is no second grid to connect to. Even if there were a second grid, running power lines is insanely expensive. Generators are cheaper.
I don't care how much generator capacity you have, it's most likely not enough to last longer than a couple of days without power.
What does this mean? The generators could effectively run forever. (at any rate far longer then "a couple of day.") The generators used are typically diesel powered. As long as there is fuel in the tank the generator will keep running. It is not hard to keep fuel tanks full. These data centers will have enough generators so that one or more can be offline for maintanance.
Parallels is not in the same league as VMware. The Parallels Workstation/Desktop product is similar to on old version of VMware workstation. More importantly Parallels dosen't have any products for servers. VMware has two powerfull server products and software to properly manage it all.
Power grid reliability is not a big concern. Data centres of this size will have backup generators. Taxes aren't going to be an issue either. These data centres will be given sweatheart tax deals, no interest loans, and other incentives. The states and counties will give out these incentives because the data centres will bring so called "high tech jobs."
The manufacturers could build an extra long card with multiple GPUs, but does the ATX spec allow for it?
We need more guilt free slack. I should not feel guilty for only working my contracted 40 hours a week. I will work the long weeks when required, but that should not be every week. I should also be able to take my coffee and lunch breaks, and have a quick chat around the "water cooler." The alternative to these little breaks is longer sneak breaks taken when no one is looking. These breaks are not as productive as the worker is worried about being caught and can't take 10 minutes to relax.
I was referring to the run-on sentences and poor structure of your post. I do not accept the argument: "it is only a /. post." I expect better from someone who has an MA in Education.
I have an MA in Education
Yet you must have failed English. I have a doctorate in literature in comparison to you and my English is terrible. No wonder the system is in jeopardy: So called Masters of Education can't type a coherent paragraph.