yeah... well... i think i'll put/usr in raid0 and do it that way, so that i have 2 partitions for/storage when one drives crashes i don't lose all data...
i wouldn't bet that through a raid0 i'll get 1% fps more in a game...
The matter with 5400rpm sucks a lot, because i want to put 2x 200gb in my server standin on my room and the server runs 24/7. So i thought about to put just one drive in there but i want/need a raid1 from/home. 7200rpm would produce to much noise and heat so i have to put more fans and that would create more noise, too!
well... mp3 is always backupped... by the internet!:) and I got all my mp3's on my archos so I have a local backup, too! anyway... in production environment raid zero is a good solution when you use it together with raid one:D - okay it's the expensivest solution but maximum performance and maximum data redundancy. on work we have an icp vortex GDT6538RD and 10x 38gb 10k rpm ibm SCA SCSI harddisks and i've tryed a RAID5 and only a performance arround 8mb/sek over ALL harddisks evaluated with iozone with freebsd 4.10-STABLE so I tryed a RAID 1+0 and now write performance is arround 40mb/sek and read performance is arround 80mb/sek (avarange speed!) the raid controler has 32mb RAM (not much...) but the values have depressed me a little bit specally the raid5 rates. some notes about benchmarking: when you benchmark a "solution" you should test on the filesytem running on the harddisks. It doesn't matter what rates you get when you direct access the storage solution because when you use it in real environment you use the filesystem and not direct access. also the benchmarktests should use different record lengths and different data sizes and should make linear read/write re-read/re-write and random "" "" tests (like iozone does). but back to the root: i do aggre to "dont use raid0 for data you don't want to loose!"
well i could took all the nonrelevat data over two or more disks/partitions but i only use this space for movies/mp3 and i hope i'll get a little performance boost by using a striped vinum software raid 0 over 2 harddisks. my new harddisklayout will be:
we don't have to talk about, that a linear raid 0 is "useless" at home nor will have a better performance also i never would buy me some raid-hardware (like a controler) for home.
i see no reason why someone shouldn't use raid0 at home...
well a friend of mine has a 19" raid case and 6 harddrives in it... but all together is 80gb... THATS braindead!
but having 2x 200gb and using raid0 (with vinum for example) for nonrelevant data (mp3s, movies etc.) should be no problem and not worth a discussion.
But i wouldn't store relevant (personal) data on a raid zero.
I currently reading TCP/IP Illustrated Vol.1 (somewhere above Chap. 19) and have begun with Vol.2 a few days ago... (implemetation of IP in FreeBSD havy stuff...:))
Well... if you have read Vol.1 you should not have further questions to a tcpdump or an ethereal or "raw packet binary dump" output.
Stevens explain all fields in the headers and what are possible options/flags and what they do. Also he explain how connections are established and closed and data are delivered. He also gives a short instruction in many protocols like dns, tftp, bootp etc. and a lot of more...
They are a little bit expensive (arround 60 - 70 euro per book) but thouse books are full of information, no unneedet informations or overhead and worth to buy it (at least Vol.1 when you are interested in IP).
just think a little bit about:
100Mbit means (theoretical) ~12MegaByte per second. Let's talk about 8Megabyte per second in this example.
Most movies have a averange duration of 90Min. so you could transfer with 8 Megabyte per second 43.2 Gigabytes in this time.
another oint you should take accocunt: your pc is only so fast, like the slowest part in it (bottlenecks everywhere!).
this means, that when you use 1gbit networking and you use a (e.g!) 5200RPM harddrive you should shoot yourself.
there are many components you have to
so think about, if you NEED it realy and what you will earn if you do so:)
Hi,
for normal (short) analyse I use ```tcpdump'', or ``netstat -I $INTERFACE 1''* to see, who's gonna make traffic (and slows down my ssh connection:)).
For long distance analysing and for raliable traffic data is use nitpicker.
Under windows i would use etherape to analyse packets... it uses pcap under windows, same as tcpdump, and so its okay:)
With open source software you have so much fscking possibilitys to build your software so every binary package lose when you want to use it under a "special" environment when you have to inlcude some exotic things or something like that.
With the FreeBSD ports for example you have the possibility to change the Makefile for compiling options. So you have you OWN package which going to be included as a package into your system and it is simple to use.
When a "port" needs another "port" to be builded, freebsd automaticaly download the source, compile it and install it unlike an RPM where only the dependencies are printed when you don't use YaST or some other dummy-administrator software.
true, but the most people don't think "social" the care only about their capital (capitalism).
no matter if the product is high quality no matter if the people like the product no matter if other people have disatvantages through this product
the only think that realy matters, is to make the MOST money with the LOWEST investigation.
(this attitude is on both sides present (sellers and buyers, but the buyers want high-quality for less money where the sellers want low-quality(cheap) for many money)
It is this way: You buy a book (or cd, or software, whatever...) read it and then you see that the quality of the book is bad but you have paid!
It should be that way: You download the book (for free) read it and THEN you choose, if you pay money (and how much) to the writer.
You can easily do it like the last way, but primaraly its illegal because you have to steal/download the book:)
I think in this way the quality MUST be good, otherwise the people don't pay money. And with the internet it is possible to publicate a book (e.g.) and you dont have to pay (as writer) much money for printing it etc.
Bruce Eckels e.g. do it that way... http://www.bruceeckel.com/
Slashdot Mirror
i would copy the harddisk for example with dd(1) and then i have all the time i need :)
yeah... well... i think i'll put /usr in raid0 and do it that way, so that i have 2 partitions for /storage when one drives crashes i don't lose all data...
/home. 7200rpm would produce to much noise and heat so i have to put more fans and that would create more noise, too!
i wouldn't bet that through a raid0 i'll get 1% fps more in a game...
The matter with 5400rpm sucks a lot, because i want to put 2x 200gb in my server standin on my room and the server runs 24/7. So i thought about to put just one drive in there but i want/need a raid1 from
- mirko
well... :) and I got all my mp3's on my archos so I have a local backup, too! :D - okay it's the expensivest solution but maximum performance and maximum data redundancy.
mp3 is always backupped... by the internet!
anyway... in production environment raid zero is a good solution when you use it together with raid one
on work we have an icp vortex GDT6538RD and 10x 38gb 10k rpm ibm SCA SCSI harddisks and i've tryed a RAID5 and only a performance arround 8mb/sek over ALL harddisks evaluated with iozone with freebsd 4.10-STABLE so I tryed a RAID 1+0 and now write performance is arround 40mb/sek and read performance is arround 80mb/sek (avarange speed!) the raid controler has 32mb RAM (not much...) but the values have depressed me a little bit specally the raid5 rates.
some notes about benchmarking:
when you benchmark a "solution" you should test on the filesytem running on the harddisks. It doesn't matter what rates you get when you direct access the storage solution because when you use it in real environment you use the filesystem and not direct access.
also the benchmarktests should use different record lengths and different data sizes and should make linear read/write re-read/re-write and random "" "" tests (like iozone does).
but back to the root: i do aggre to "dont use raid0 for data you don't want to loose!"
well i could took all the nonrelevat data over two or more disks/partitions but i only use this space for movies/mp3 and i hope i'll get a little performance boost by using a striped vinum software raid 0 over 2 harddisks.
/usr | raid 1 | raid 0 | /var | raid 1 | raid 0 | /home /storage
my new harddisklayout will be:
disk 1: | / |
disk 2: | swap |
raid :
we don't have to talk about, that a linear raid 0 is "useless" at home nor will have a better performance also i never would buy me some raid-hardware (like a controler) for home.
i see no reason why someone shouldn't use raid0 at home...
well a friend of mine has a 19" raid case and 6 harddrives in it... but all together is 80gb... THATS braindead!
but having 2x 200gb and using raid0 (with vinum for example) for nonrelevant data (mp3s, movies etc.) should be no problem and not worth a discussion.
But i wouldn't store relevant (personal) data on a raid zero.
I currently reading TCP/IP Illustrated Vol.1 (somewhere above Chap. 19) and have begun with Vol.2 a few days ago... (implemetation of IP in FreeBSD havy stuff... :))
Well... if you have read Vol.1 you should not have further questions to a tcpdump or an ethereal or "raw packet binary dump" output.
Stevens explain all fields in the headers and what are possible options/flags and what they do. Also he explain how connections are established and closed and data are delivered. He also gives a short instruction in many protocols like dns, tftp, bootp etc. and a lot of more...
They are a little bit expensive (arround 60 - 70 euro per book) but thouse books are full of information, no unneedet informations or overhead and worth to buy it (at least Vol.1 when you are interested in IP).
all in one... he explains the internet :)
just think a little bit about:
100Mbit means (theoretical) ~12MegaByte per second. Let's talk about 8Megabyte per second in this example.
Most movies have a averange duration of 90Min. so you could transfer with 8 Megabyte per second 43.2 Gigabytes in this time.
another oint you should take accocunt: your pc is only so fast, like the slowest part in it (bottlenecks everywhere!).
this means, that when you use 1gbit networking and you use a (e.g!) 5200RPM harddrive you should shoot yourself.
there are many components you have to so think about, if you NEED it realy and what you will earn if you do so :)
for normal (short) analyse I use ```tcpdump'', or ``netstat -I $INTERFACE 1''* to see, who's gonna make traffic (and slows down my ssh connection
For long distance analysing and for raliable traffic data is use nitpicker.
Under windows i would use etherape to analyse packets... it uses pcap under windows, same as tcpdump, and so its okay :)
Greetings
Mirko
* your `uname -s' should look like this:
With open source software you have so much fscking possibilitys to build your software so every binary package lose when you want to use it under a "special" environment when you have to inlcude some exotic things or something like that.
With the FreeBSD ports for example you have the possibility to change the Makefile for compiling options. So you have you OWN package which going to be included as a package into your system and it is simple to use.
When a "port" needs another "port" to be builded, freebsd automaticaly download the source, compile it and install it unlike an RPM where only the dependencies are printed when you don't use YaST or some other dummy-administrator software.
true, but the most people don't think "social" the care only about their capital (capitalism).
no matter if the product is high quality
no matter if the people like the product
no matter if other people have disatvantages through this product
the only think that realy matters, is to make the MOST money with the LOWEST investigation.
(this attitude is on both sides present (sellers and buyers, but the buyers want high-quality for less money where the sellers want low-quality(cheap) for many money)
I think its the age of "free fnformations"
:)
It is this way:
You buy a book (or cd, or software, whatever...) read it and then you see that the quality of the book is bad but you have paid!
It should be that way:
You download the book (for free) read it and THEN you choose, if you pay money (and how much) to the writer.
You can easily do it like the last way, but primaraly its illegal because you have to steal/download the book
I think in this way the quality MUST be good, otherwise the people don't pay money. And with the internet it is possible to publicate a book (e.g.) and you dont have to pay (as writer) much money for printing it etc.
Bruce Eckels e.g. do it that way... http://www.bruceeckel.com/