I asked that question -- in a blog post on my wife's blog -- after discovering too many personal details about the family out there on her blog. We'd agreed when she started blogging (she's a neophyte for the most part) that details such as names, locations, ages, where we went on vacation, etc. were not allowed, and she agreed.
Hijacking her blog was actually pretty easy, since I'm her local IT Shop -- I host our email server/web services in-house, and of course, her login falls under my.org AD domain (hate to burst everyone's bubble, but I'm more a Windows expert than *nix, though I can play in both). Not only did I make myself an admin for her blog, I made her a non-admin, so she couldn't undo my post (which, by the way, was my first blog post ever -- I suppose that indicates my stance on blogging and personal privacy pretty well). This had the delicious side effect of her not speaking to me for several days. Several whole days! During that time, I added a poll to her blog ("How jaded had [I] become?" was the question), changed the hideous colors/theme and of course, edited out all the stuff that shouldn't have been there in the first place.
She eventually asked nicely for her blog back (there wasn't really much else she could do), and I gave in -- but only after she conceded her own lack of judgment in naming the kids, where we'd moved to, where we went on vacation, how old I am, and other details that have no business on the Internets.
Moral of the story: People who blog, usually say too much. Unless you can hack their blog, good luck stopping it entirely.
My opinion of the story: People who blog, usually shouldn't.
I run my own mail server and have router-side blocks for entire IP blocks, including China. Even so, a *significant* portion of the spam hitting my mail server originates from China. I'm all for universal access, but only if it's used responsibly. Clearly that isn't the case here, so let the pool dry up.
I'd say the same thing if the majority of my spam came from elsewhere. Nigeria, you're next!
I, GEORGE W. BUSH, President of the United States of America, find that, due to the unusual and extraordinary threat to the national security and foreign policy of the United States posed by acts of violence threatening the peace and stability of Iraq and undermining efforts to promote economic reconstruction and political reform in Iraq and to provide humanitarian assistance to the Iraqi people, it is in the interests of the United States to take additional steps with respect to the national emergency declared in Executive Order 13303 of May 22, 2003, and expanded in Executive Order 13315 of August 28, 2003, and relied upon for additional steps taken in Executive Order 13350 of July 29, 2004, and Executive Order 13364 of November 29, 2004.
Since when does the Iraqi conflict pose a "national emergency" for the citizens of the United States? That flawed "notion" is being used by Bush as basis for this new Executive Order.
And for the record, this would appear to conflict with the 4th Amendment, not the 5th Amendment, as the OP incorrectly states.
Helpful links so you can gain your own understanding instead of reading a bunch of off-the-cuff/. commentary:
If your SAN is slower than local disk, all I can say is your SAN needs serious attention. A single, local disk (avg 8ms latency) will never match the throughput of a properly-configured array of disks (avg 1-2ms latency or better) from which you're being fed raw data blocks across a 2Gbit or 4Gbit SAN fabric. Heck, iSCSI using the Microsoft software initiator is faster than local storage in all but a few instances (iSCSI = iSAN, but I can count on way too many fingers how many times people put iSCSI traffic on their production network with everything else...).
High cost or not, I think you need to have a chat with your EMC folks if the latency/throughput is worse than local storage.
(Much of your irritation with SAN seems to be the political environment of your work environment. If you have to beg for 30GB of disk space on a SAN, I guess I can see why you're annoyed, but that doesn't jive with the technical realities.)
Having navigated the career trenches of the technology world for the past 15 years (yeah, I'm still pretty young), the one thing that's always struck me as the downfall of any manager is the inability to make a decision -- quickly, and sometimes without all the information one would like -- and then stick to that decision.
That is not to say you want to work for a manager that makes rash or random decisions and then becomes all-or-nothing about that decision because he says that's the way it is -- obviously, there must be balance between knowing all the details and making a choice so that the team can carry on, whether it's a political, social, technological or other decision.
For that reason, when I interview other prospective managers (I manage a team of 12-15 Technical Support and Escalations Engineers), I almost always toss out this question if I haven't already gotten a good sense of their decision-making abilities:
How many cotton balls would it take to fill up this room?
Within reason, the answer is mostly irrelevant. Someone who still thinks very technically will sit and calculate the dimensions of the room and, just like guessing how many jelly beans are in the contest jar, try to come up with an exact number. Someone who is haphazard will toss out a random number that's usually way high or way low...not that I've ever actually calculated the "right" number for whatever conference room we're in.
The intent of the question is to determine how quickly that person can make an intelligent and informed decision and move on. Trust me here -- you really don't want to work for a manager that has analysis paralysis. Then again, you don't want to work for a manager that doesn't really care about the details, either. Somewhere in between those two extremes, IMO, is good.
I usually follow that question with another zinger: What is your favorite movie? There's a lot you can divine from knowing what someone's favorite movie is. If it's "Office Space," chances are you'll be working for/with a manager with a sense of humor that won't take themselves very seriously. If it's "Texas Chainsaw Massacre," you might want to drop your standard list of (probably useless*) interviewing questions and delve into some in-depth behavioral interviewing questions... and heaven help you during that discussion.:)
*There's nothing more useless than this question: Where do you see yourself in 5 years? When I hear that in an interview (some of my peers that don't have very good interviewing skills ask it), I cringe visibly.
All I can say is, stay out of the "coffehouses" there for a couple days before posting on/.
SORBS == ass.
As someone who's run my own mail services on a dynamic address for the last 7 years (for reference, the actual IP has changed only 3 times, the most recent when we physically moved), I can assure you that SORBS has the highest false-positive rate of any of the others I've plugged in and tested. The only ones I still use are SpamHaus and SpamCop. Everything else gets killed off by intelligent scoring from SpamAssassin and my own filter rules. And just because I don't have the ability to get a reasonably-priced static address that doesn't live within someone else's IP range is by NO means any indication that my mail server has any problems. As has been pointed out earlier, SORBS model appears be more about blocking entire IP ranges, then make money unblocking them.
DNS block lists are no longer as effective as anti-spam tool as they once were. I get better results so far from greylisting and using a spamtrap. Oh, and tarpitting is always good.:)
...I find it interesting that so few reponses have hit this thread yet. I'm not outraged at the Gracenote nightmare (I got over it), but I would have thought that the cumulative efforts by many folks here would have sparked a bit more interest.
Just do what I do. Block China, Pakistan, Russia and its ex-associated province state territories, India, Singapore, Korea (both parts), etc. from even making a TCP connection. Sorry, it's not that we don't love you guys, but you create a crapload of unnecessary work for the rest of us. Weighing that against the general contribution makes the decision easy.
My life got much easier once I found and/or created an IP-based block list for these and similar countries...
I call 100% Shenanigans. Ok, you said "most ATA RAID..." so I call 99% Shenanigans.
I built a sizeable (by 2003 standards) 4-drive RAID5 array using a cheap P3 motherboard and an Adaptec 2400A RAID controller on Feb. 16, 2003. It's still running swimmingly, serving up my media and user directories, and has never had any glitches -- not even a failed disk. Data corruption? None. If you have data corruption on a RAID5 controller, you need to seriously reconsider your parts list. (Not saying the ATA-2400A is the way to go - it's since been discontinued, and I'm not sure I'd go with the SATA version of Adaptec's RAID5 next time.)
I built it after getting tired of hardware failures -- I was bitten by both the IBM Deathstar drives and Abit's bad capacitors. It's not intended to be 100% available, or 100% infallible. I have a second 300GB single drive that does nightly file copies of critical data off the array, and I do regular "other" backups to DVD, CD-R, etc. RAID5 is not the answer to all your storage problems, but it's served me better than any other system I ever built before or after (so far).
Uptime on the array is over 99.85% over a 3 year period, and that's including 2 days worth of ice-storm induced power outage (January 2004), another half-day of hurricane-inflicted power outage (Isabel, September 2003), and my son disovering the power button more than a few times (grrrr...). Even with all that, it's never flinched once.
It's been my web server, mail server, FTP/TFTP server, print server, Shoutcast server, and audio/video media server (XBMC) for years. Best computer project I ever embarked on, and if I thought my 60K/sec uplink would stand up to the/. onslaught, I'd post links to the project pictures off this very same box.:)
Is it as powerful as a full-blown SCSI array? Definitely not, but it's much quieter, much cheaper (total cost: $320 for the adapter, which was expensive even then; $95 per 80GB WD800JB drive), perfectly adequate (I/O at ~56MB/sec, which is only ~5MB/sec slower than a single SCSI Ultra320 15k RPM drive). And best of all, it's 100% Ghetto Brand quality, so you know it's legit...
You just gave me mod points for still having 21" CRTs on all my PCs. Even at work, I found some 21" CRTs that had failed white (the Sony tube w/ the resistor that fails). A quick Google search found details on which resistor failed, and what to replace it with. I actually did the replacement while on the phone with a customer, helping them restore a very, very "down" system.
Now it works great, takes up lots of room (less room for papers and crap), and I *know* I am more efficient than my counterparts that got stuck with ^H^H^H^H^H^H^H^H handed 19" LCDs. My 21" CRT will easily do 1600x1200 @100Hz, but the LCDs will only do 1280x1024 @72Hz.
I used to be really against iSCSI, as the native stacks on various OSes just did not deal with it well. By that I mean that a 50 MB/s file transfer would consume almost 100% of a 3ghz CPU. Also, the hard limit on gig-e transfers of 85 MB/s (TCP/IP overhead + iSCSI overhead) was just too low.
Now, that has all changed. Not only can you get TCP/IP Offload Engines for just about every OS (I don't work with Windows, so I don't know what the status of that is). Also, 10 gigabit ethernet has become financially reasonable.
Sorry, but this claim is (I'll be kind and say at least partially) crap. As someone who works for the folks that pretty much invented iSCSI (look it up, no shameless advertising here), I can tell you, there are *VERY* few systems that actually _require_ a TOE (TCP Offload Engine) card.
Even the (gasp!) Microsoft iSCSI initiator, believe it or not, is darn fast (I've done plenty of testing, but since YMMV, I won't make any number claims here), and does not consume anywhere near the CPU you claim. Other software initiators are also (for the most part) plenty fast. We have thousands of customers running multiple mail, database, and CRM services over iSCSI, and unless the backend storage is overloaded (too small raidgroups, full volumes, etc.) or there are underlying network problems, everything works swimmingly.
Now, once you move to 10G Ethernet, I would be concerned that the OS' network stack might not be able to cut it, and you may need a TOE card. But for GigE, I've only seen *1* instance in the last 2 years of supporting customer iSCSI environments, where a TOE card was appropriate and recommended due to the performance data we collected.
Scary-but-strangely-appropriate anecdote: had a customer this month doing a proof-of-concept on a MS Cluster running a very large Exchange environment against our backend storage (from different hosts) over iSCSI. The customer was insistent on running TOE cards for the POC, in this case a pair of Qlogic 4010's. During testing, they noticed that file copies between LUNs across these TOE cards took FOREVER (on the order of hours to copy ~2GB). We looked at all the back-end storage performance -- clean. Looked at the host (the usual stuff like queue depth) -- mostly clean.
It turned out that when copying between the TOE cards (or in and out the same TOE card), the throughput went straight into the crapper. Once they switched to the software initiator, everything was peachy. Here's the scary part: because the "IT guy" had already put his balls on the line and demanded that they needed the 4010s, he wasn't willing to switch to the software iSCSI initiator, probably because he'd lose credibility with his bosses. We looked at newer device drivers for the Qlogic 4010s -- but he wouldn't hear of it. So if you send them e-mail and wonder why it takes so long to respond.... now you know.
Change the voltage. It's easy to take a fan running at the usual 12v and make it run a bit slower (and QUIETER) by mixing voltages. Instead of connecting to the 12v line and ground, cross-connect to the 12v and 5v lines, giving the fan ~7v instead. This is a very easy hybrid approach, only requiring wire strippers and some electrical tape.
If you need something more specific, then a fan controller is in order. Some examples (you can build your own internal to the case, but these look better and are obviously more easily accessible):
You probably don't need anything more than the 3-channel. You can add more than one fan per channel on most decent controllers, as long as you set them high enough to get past initial start-up. If you have a large box like my RAID5 system (2x 120mm, 3x 80mm), you probably need the 4-channel fan controller.
Thinking my new Athlon64 3200+ system would run as hot as the Pentium 3 and P4 lines, I went ahead and bought the XP-120 (http://www.thermalright.com/a_page/main_product_x p120.htm) and a nice, quiet 120mm Panaflo fan (model FBA12G12L1A, the one recommended for use w/ the XP-120) along with all the other parts.
Imagine my annoyance/glee when I discovered that, not only do I not need the XP-120/120mm fan, but I am *unable* to get the CPU to go above 100F, even when running at a 145-155%% overclock (I have PC4400 TCCD memory in here) and under heavy load (rendering and recoding video, Half-Life 2, etc.). Try doing that on an Intel P4. On most systems I've seen, at ~125% overclock and even with the best memory, the heat build-up in the P4 core is quite likely to take the system down, unless you're doing watercooling.
I'm now in the process of killing off most of my Intel-based systems and moving to AMD. (Pentium III RAID5 box will stay, only because I'd be tempted to look at moving off the Adaptec 2400A RAID and to a SATA RAID5 solution.) They're clearly cooler (this room has been a cooling problem in spring/summer for years because of it), use about 1/2 to 2/3 the power (measured this as well, so it's not anecdotal), and definitely run quieter due to the lack of high-speed cooling fans.
Anyone want to buy a new/unused XP-120 heatpipe cooler and Panaflo L1A-120?
For accuracy's sake: NetApp storage devices use RAID4 (striping with dedicated parity disk, or dual dedicated parity disks), so they're not RAID5. RAID4 has certain advantages over RAID5 (lower CPU utilization, for one), but the primary benefit of a NetApp storage system is the WAFL (Write-Anywhere File Layout) file system -- instead of having to write data and metadata in different areas of the disk(s), as most vendors do, it's written alongside each other, and any free block is fair game, making them pretty zippy.
For home use, only an older F720, F740 or F760 makes sense due to cost, along with an FC9 disk shelf or two full of 18- or 36-GB Fibre Channel disks... but the noise put off by that is not something you're going to leave under your desk unless you get off on white noise. (A friend of mine actually has a F760 with a few FC9 disk shelves under his house...:)
' If you _where_ going to hire someone to manage your information systems would you see a benefit in them having a specialized education as opposed to 3 or 4 years of experience?"
No, I'd rather they come prepackaged with spelling and grammar built-in.
...than any other manufacturer. They'll follow the money trail. If they can sell more PCs by no allying tightly with Microsoft, so be it. If they can sell more PCs to the home market by appearing to be best buds with Microsoft, well, they'll do that, too.
Wish I had mod points today. This is an insightful response, full of things to chew on (breakroom pun not intended). If you don't want a surprise on your first day, ask to see what you're signing up for!
Really. Pick one you like (Palm or PocketPC/Windows Mobile/whatever it's called now) and just carry it with you. There are dozens of time-tracking apps for both platforms, as well as others (Symbian, Smartphone, etc.).
It might add $100 to the cost. We're talking about an unmanaged, 4- or 5-port GigE switch here. I got my 8-port SMC 8508 (which supports Jumbo Frames, by the way) for around $100, and that was over a year ago. They're now in the $70 ballpark.
Even though I have a 3-year old RAID5 box already (Adaptec 2400A + 4x 160GB WD drives), I would seriously consider buying something like this for the simplicity, but it HAS to support GigE (and support it well), as I use my RAID box heavily for audio/video storage and editing, as well as audio/video streaming.
So what if it runs on Windows? I've been running MDaemon for many years now (at home) and have had a grand total of > 4 spam messages and zero virus-infected attachments since installation. As long as your network is secure, the host box is tightened down, and you properly configure all the niceties (SpamAssassin, RBLs, Bayesian Filtering, etc.), you're good to go.
Contrast that to the 30-50 per day I was getting before through another ISP (Earthlink). I'm surprised how few people even know about MDaemon (www.altn.com), considering how utterly powerful it is.
One of the biggest killers in Windows performance is installing and uninstalling applications. It leads to registry bloat, stray files and directories, and disk fragmentation. You can be reactive and try to fix all that, but it's better to prevent it in the first place.
I've been virtualizing many of my applications inside of VMware and it's helped tremendously. Applications that I don't need all the time, or apps that I only need to use one time, or apps that I'm not sure will do what I need -- they all get installed in a pristine Windows XP VM. If they work, they get reinstalled in my "working VM" and the test VM gets destroyed (thus, stays pristine).
I used to do some of this with multiple PCs, but there's absolutely no need for that anymore, and it's more time-consuming re-imaging a piece of hardware back to its pristine state.
For what it's worth, I have a P3/850 that's running XP that hasn't been reloaded in almost 4 years. It's starting to degrade due to the fact that I violated the above rule a few times, but is still perfectly usable for stuff like e-mail, quick Photoshop edits, instant messaging, word processing.
Slashdot Mirror
I asked that question -- in a blog post on my wife's blog -- after discovering too many personal details about the family out there on her blog. We'd agreed when she started blogging (she's a neophyte for the most part) that details such as names, locations, ages, where we went on vacation, etc. were not allowed, and she agreed.
Hijacking her blog was actually pretty easy, since I'm her local IT Shop -- I host our email server/web services in-house, and of course, her login falls under my .org AD domain (hate to burst everyone's bubble, but I'm more a Windows expert than *nix, though I can play in both). Not only did I make myself an admin for her blog, I made her a non-admin, so she couldn't undo my post (which, by the way, was my first blog post ever -- I suppose that indicates my stance on blogging and personal privacy pretty well). This had the delicious side effect of her not speaking to me for several days. Several whole days! During that time, I added a poll to her blog ("How jaded had [I] become?" was the question), changed the hideous colors/theme and of course, edited out all the stuff that shouldn't have been there in the first place.
She eventually asked nicely for her blog back (there wasn't really much else she could do), and I gave in -- but only after she conceded her own lack of judgment in naming the kids, where we'd moved to, where we went on vacation, how old I am, and other details that have no business on the Internets.
Moral of the story: People who blog, usually say too much. Unless you can hack their blog, good luck stopping it entirely.
My opinion of the story: People who blog, usually shouldn't.
Discuss amongst yourselves. :)
I run my own mail server and have router-side blocks for entire IP blocks, including China. Even so, a *significant* portion of the spam hitting my mail server originates from China. I'm all for universal access, but only if it's used responsibly. Clearly that isn't the case here, so let the pool dry up. I'd say the same thing if the majority of my spam came from elsewhere. Nigeria, you're next!
Yeah, I see nothing on that blog about that now.... funny. I bet it got changed immediately.
Since when does the Iraqi conflict pose a "national emergency" for the citizens of the United States? That flawed "notion" is being used by Bush as basis for this new Executive Order.
And for the record, this would appear to conflict with the 4th Amendment, not the 5th Amendment, as the OP incorrectly states.
Helpful links so you can gain your own understanding instead of reading a bunch of off-the-cuff /. commentary:
http://www.whitehouse.gov/news/releases/2003/05/2http://www.whitehouse.gov/news/releases/2007/07/2
If your SAN is slower than local disk, all I can say is your SAN needs serious attention. A single, local disk (avg 8ms latency) will never match the throughput of a properly-configured array of disks (avg 1-2ms latency or better) from which you're being fed raw data blocks across a 2Gbit or 4Gbit SAN fabric. Heck, iSCSI using the Microsoft software initiator is faster than local storage in all but a few instances (iSCSI = iSAN, but I can count on way too many fingers how many times people put iSCSI traffic on their production network with everything else...).
High cost or not, I think you need to have a chat with your EMC folks if the latency/throughput is worse than local storage.
(Much of your irritation with SAN seems to be the political environment of your work environment. If you have to beg for 30GB of disk space on a SAN, I guess I can see why you're annoyed, but that doesn't jive with the technical realities.)
Having navigated the career trenches of the technology world for the past 15 years (yeah, I'm still pretty young), the one thing that's always struck me as the downfall of any manager is the inability to make a decision -- quickly, and sometimes without all the information one would like -- and then stick to that decision.
...not that I've ever actually calculated the "right" number for whatever conference room we're in.
:)
That is not to say you want to work for a manager that makes rash or random decisions and then becomes all-or-nothing about that decision because he says that's the way it is -- obviously, there must be balance between knowing all the details and making a choice so that the team can carry on, whether it's a political, social, technological or other decision.
For that reason, when I interview other prospective managers (I manage a team of 12-15 Technical Support and Escalations Engineers), I almost always toss out this question if I haven't already gotten a good sense of their decision-making abilities:
How many cotton balls would it take to fill up this room?
Within reason, the answer is mostly irrelevant. Someone who still thinks very technically will sit and calculate the dimensions of the room and, just like guessing how many jelly beans are in the contest jar, try to come up with an exact number. Someone who is haphazard will toss out a random number that's usually way high or way low
The intent of the question is to determine how quickly that person can make an intelligent and informed decision and move on. Trust me here -- you really don't want to work for a manager that has analysis paralysis. Then again, you don't want to work for a manager that doesn't really care about the details, either. Somewhere in between those two extremes, IMO, is good.
I usually follow that question with another zinger: What is your favorite movie? There's a lot you can divine from knowing what someone's favorite movie is. If it's "Office Space," chances are you'll be working for/with a manager with a sense of humor that won't take themselves very seriously. If it's "Texas Chainsaw Massacre," you might want to drop your standard list of (probably useless*) interviewing questions and delve into some in-depth behavioral interviewing questions... and heaven help you during that discussion.
*There's nothing more useless than this question: Where do you see yourself in 5 years? When I hear that in an interview (some of my peers that don't have very good interviewing skills ask it), I cringe visibly.
All I can say is, stay out of the "coffehouses" there for a couple days before posting on /.
:)
SORBS == ass.
As someone who's run my own mail services on a dynamic address for the last 7 years (for reference, the actual IP has changed only 3 times, the most recent when we physically moved), I can assure you that SORBS has the highest false-positive rate of any of the others I've plugged in and tested. The only ones I still use are SpamHaus and SpamCop. Everything else gets killed off by intelligent scoring from SpamAssassin and my own filter rules. And just because I don't have the ability to get a reasonably-priced static address that doesn't live within someone else's IP range is by NO means any indication that my mail server has any problems. As has been pointed out earlier, SORBS model appears be more about blocking entire IP ranges, then make money unblocking them.
DNS block lists are no longer as effective as anti-spam tool as they once were. I get better results so far from greylisting and using a spamtrap. Oh, and tarpitting is always good.
...I find it interesting that so few reponses have hit this thread yet. I'm not outraged at the Gracenote nightmare (I got over it), but I would have thought that the cumulative efforts by many folks here would have sparked a bit more interest.
Just do what I do. Block China, Pakistan, Russia and its ex-associated province state territories, India, Singapore, Korea (both parts), etc. from even making a TCP connection. Sorry, it's not that we don't love you guys, but you create a crapload of unnecessary work for the rest of us. Weighing that against the general contribution makes the decision easy.
My life got much easier once I found and/or created an IP-based block list for these and similar countries...
70 hours! Holy crap. Even my ATA RAID5 array wouldn't take that long (one failed 80GB drive in a 4x80GB RAID5 array -- rebuild took ~20 hours).
That's just more fodder toward buying a used filer (NetApp) running RAID4 and WAFL...
(Warning: Shameless work plug and great stock tip above.)
I call 100% Shenanigans. Ok, you said "most ATA RAID..." so I call 99% Shenanigans.
/. onslaught, I'd post links to the project pictures off this very same box. :)
I built a sizeable (by 2003 standards) 4-drive RAID5 array using a cheap P3 motherboard and an Adaptec 2400A RAID controller on Feb. 16, 2003. It's still running swimmingly, serving up my media and user directories, and has never had any glitches -- not even a failed disk. Data corruption? None. If you have data corruption on a RAID5 controller, you need to seriously reconsider your parts list. (Not saying the ATA-2400A is the way to go - it's since been discontinued, and I'm not sure I'd go with the SATA version of Adaptec's RAID5 next time.)
I built it after getting tired of hardware failures -- I was bitten by both the IBM Deathstar drives and Abit's bad capacitors. It's not intended to be 100% available, or 100% infallible. I have a second 300GB single drive that does nightly file copies of critical data off the array, and I do regular "other" backups to DVD, CD-R, etc. RAID5 is not the answer to all your storage problems, but it's served me better than any other system I ever built before or after (so far).
Uptime on the array is over 99.85% over a 3 year period, and that's including 2 days worth of ice-storm induced power outage (January 2004), another half-day of hurricane-inflicted power outage (Isabel, September 2003), and my son disovering the power button more than a few times (grrrr...). Even with all that, it's never flinched once.
It's been my web server, mail server, FTP/TFTP server, print server, Shoutcast server, and audio/video media server (XBMC) for years. Best computer project I ever embarked on, and if I thought my 60K/sec uplink would stand up to the
Is it as powerful as a full-blown SCSI array? Definitely not, but it's much quieter, much cheaper (total cost: $320 for the adapter, which was expensive even then; $95 per 80GB WD800JB drive), perfectly adequate (I/O at ~56MB/sec, which is only ~5MB/sec slower than a single SCSI Ultra320 15k RPM drive). And best of all, it's 100% Ghetto Brand quality, so you know it's legit...
You just gave me mod points for still having 21" CRTs on all my PCs. Even at work, I found some 21" CRTs that had failed white (the Sony tube w/ the resistor that fails). A quick Google search found details on which resistor failed, and what to replace it with. I actually did the replacement while on the phone with a customer, helping them restore a very, very "down" system.
:)
Now it works great, takes up lots of room (less room for papers and crap), and I *know* I am more efficient than my counterparts that got stuck with ^H^H^H^H^H^H^H^H handed 19" LCDs. My 21" CRT will easily do 1600x1200 @100Hz, but the LCDs will only do 1280x1024 @72Hz.
I win.
Sorry, but this claim is (I'll be kind and say at least partially) crap. As someone who works for the folks that pretty much invented iSCSI (look it up, no shameless advertising here), I can tell you, there are *VERY* few systems that actually _require_ a TOE (TCP Offload Engine) card.
Even the (gasp!) Microsoft iSCSI initiator, believe it or not, is darn fast (I've done plenty of testing, but since YMMV, I won't make any number claims here), and does not consume anywhere near the CPU you claim. Other software initiators are also (for the most part) plenty fast. We have thousands of customers running multiple mail, database, and CRM services over iSCSI, and unless the backend storage is overloaded (too small raidgroups, full volumes, etc.) or there are underlying network problems, everything works swimmingly.
Now, once you move to 10G Ethernet, I would be concerned that the OS' network stack might not be able to cut it, and you may need a TOE card. But for GigE, I've only seen *1* instance in the last 2 years of supporting customer iSCSI environments, where a TOE card was appropriate and recommended due to the performance data we collected.
Scary-but-strangely-appropriate anecdote: had a customer this month doing a proof-of-concept on a MS Cluster running a very large Exchange environment against our backend storage (from different hosts) over iSCSI. The customer was insistent on running TOE cards for the POC, in this case a pair of Qlogic 4010's. During testing, they noticed that file copies between LUNs across these TOE cards took FOREVER (on the order of hours to copy ~2GB). We looked at all the back-end storage performance -- clean. Looked at the host (the usual stuff like queue depth) -- mostly clean.
It turned out that when copying between the TOE cards (or in and out the same TOE card), the throughput went straight into the crapper. Once they switched to the software initiator, everything was peachy. Here's the scary part: because the "IT guy" had already put his balls on the line and demanded that they needed the 4010s, he wasn't willing to switch to the software iSCSI initiator, probably because he'd lose credibility with his bosses. We looked at newer device drivers for the Qlogic 4010s -- but he wouldn't hear of it. So if you send them e-mail and wonder why it takes so long to respond.... now you know.
(The customer is NOT always right!)
Change the voltage. It's easy to take a fan running at the usual 12v and make it run a bit slower (and QUIETER) by mixing voltages. Instead of connecting to the 12v line and ground, cross-connect to the 12v and 5v lines, giving the fan ~7v instead. This is a very easy hybrid approach, only requiring wire strippers and some electrical tape.
= connector.shtml
2 E168351182172 E168139995062 E16813999504
A brief, but good explanation is here: http://www.heatsink-guide.com/content.php?content
If you need something more specific, then a fan controller is in order. Some examples (you can build your own internal to the case, but these look better and are obviously more easily accessible):
1 channel: http://www.newegg.com/Product/Product.asp?Item=N8
3 channel: http://www.newegg.com/Product/Product.asp?Item=N8
4 channel: http://www.newegg.com/Product/Product.asp?Item=N8
You probably don't need anything more than the 3-channel. You can add more than one fan per channel on most decent controllers, as long as you set them high enough to get past initial start-up. If you have a large box like my RAID5 system (2x 120mm, 3x 80mm), you probably need the 4-channel fan controller.
Thinking my new Athlon64 3200+ system would run as hot as the Pentium 3 and P4 lines, I went ahead and bought the XP-120 (http://www.thermalright.com/a_page/main_product_x p120.htm) and a nice, quiet 120mm Panaflo fan (model FBA12G12L1A, the one recommended for use w/ the XP-120) along with all the other parts.
Imagine my annoyance/glee when I discovered that, not only do I not need the XP-120/120mm fan, but I am *unable* to get the CPU to go above 100F, even when running at a 145-155%% overclock (I have PC4400 TCCD memory in here) and under heavy load (rendering and recoding video, Half-Life 2, etc.). Try doing that on an Intel P4. On most systems I've seen, at ~125% overclock and even with the best memory, the heat build-up in the P4 core is quite likely to take the system down, unless you're doing watercooling.
I'm now in the process of killing off most of my Intel-based systems and moving to AMD. (Pentium III RAID5 box will stay, only because I'd be tempted to look at moving off the Adaptec 2400A RAID and to a SATA RAID5 solution.) They're clearly cooler (this room has been a cooling problem in spring/summer for years because of it), use about 1/2 to 2/3 the power (measured this as well, so it's not anecdotal), and definitely run quieter due to the lack of high-speed cooling fans.
Anyone want to buy a new/unused XP-120 heatpipe cooler and Panaflo L1A-120?
[Full disclosure - I am a NetApp employee.]
:)
For accuracy's sake: NetApp storage devices use RAID4 (striping with dedicated parity disk, or dual dedicated parity disks), so they're not RAID5. RAID4 has certain advantages over RAID5 (lower CPU utilization, for one), but the primary benefit of a NetApp storage system is the WAFL (Write-Anywhere File Layout) file system -- instead of having to write data and metadata in different areas of the disk(s), as most vendors do, it's written alongside each other, and any free block is fair game, making them pretty zippy.
For home use, only an older F720, F740 or F760 makes sense due to cost, along with an FC9 disk shelf or two full of 18- or 36-GB Fibre Channel disks... but the noise put off by that is not something you're going to leave under your desk unless you get off on white noise. (A friend of mine actually has a F760 with a few FC9 disk shelves under his house...
1) Which one will you be happiest at based on your career path, goals and objectives, ability to achieve those goals, etc.?
2) Which one will your wife support you the most in?
If the answers between #1 and #2 differ..... well, you have a different question to post to AskSlashdot, don't you?
Wouldn't know. Never been dumb enough to buy a Dell. :)
...than any other manufacturer. They'll follow the money trail. If they can sell more PCs by no allying tightly with Microsoft, so be it. If they can sell more PCs to the home market by appearing to be best buds with Microsoft, well, they'll do that, too.
Nothing to see here.
Wish I had mod points today. This is an insightful response, full of things to chew on (breakroom pun not intended). If you don't want a surprise on your first day, ask to see what you're signing up for!
Dude, buy a PDA.
Really. Pick one you like (Palm or PocketPC/Windows Mobile/whatever it's called now) and just carry it with you. There are dozens of time-tracking apps for both platforms, as well as others (Symbian, Smartphone, etc.).
It might add $100 to the cost. We're talking about an unmanaged, 4- or 5-port GigE switch here. I got my 8-port SMC 8508 (which supports Jumbo Frames, by the way) for around $100, and that was over a year ago. They're now in the $70 ballpark.
Even though I have a 3-year old RAID5 box already (Adaptec 2400A + 4x 160GB WD drives), I would seriously consider buying something like this for the simplicity, but it HAS to support GigE (and support it well), as I use my RAID box heavily for audio/video storage and editing, as well as audio/video streaming.
So what if it runs on Windows? I've been running MDaemon for many years now (at home) and have had a grand total of > 4 spam messages and zero virus-infected attachments since installation. As long as your network is secure, the host box is tightened down, and you properly configure all the niceties (SpamAssassin, RBLs, Bayesian Filtering, etc.), you're good to go.
Contrast that to the 30-50 per day I was getting before through another ISP (Earthlink). I'm surprised how few people even know about MDaemon (www.altn.com), considering how utterly powerful it is.
One of the biggest killers in Windows performance is installing and uninstalling applications. It leads to registry bloat, stray files and directories, and disk fragmentation. You can be reactive and try to fix all that, but it's better to prevent it in the first place.
I've been virtualizing many of my applications inside of VMware and it's helped tremendously. Applications that I don't need all the time, or apps that I only need to use one time, or apps that I'm not sure will do what I need -- they all get installed in a pristine Windows XP VM. If they work, they get reinstalled in my "working VM" and the test VM gets destroyed (thus, stays pristine).
I used to do some of this with multiple PCs, but there's absolutely no need for that anymore, and it's more time-consuming re-imaging a piece of hardware back to its pristine state.
For what it's worth, I have a P3/850 that's running XP that hasn't been reloaded in almost 4 years. It's starting to degrade due to the fact that I violated the above rule a few times, but is still perfectly usable for stuff like e-mail, quick Photoshop edits, instant messaging, word processing.