Wait until everyone leaves on holiday some unusually hot 4th of July morning. The earlybirds are fine, but those leaving later have empty "tanks" because ConEd sucked out all their battery power to run all of the air conditioners.
The YF-12was a high altitude and high speed interceptor. It fired Air Intercept Missiles (AIM-7's)which are already aircraft in their own right. It did not drop free fall munitions at high speed.
Thanks for the reply. My puzzlement is that when measuring the pure iSCSI performance (with a ramdisk on the target) performance was great, regardless of the network -- avg I/O delays of 1mS with one IOMeter worker thread and 4mS with four worker threads... and a few hundred IOPS. This is all random I/O and the total throughput in MB/s is fairly low and not important for my app.
When the iSCSI target was disconnected from the ramdisk and connected to the real drives, the avg I/O delays were considerably worse than just adding the pure iSCSI overhead (1-4 mS) and the local performance on the target machine (7 mS). Instead of about 8-11 mS I get an avg I/O delay of 22-26 mS.
This is a simple "SAN" with one initiator and one target, no MPIO, etc. For pure random I/O there was no performance difference between a dedicated HP ProCurve, consumer 100MBit, consumer 1GBit, and consumer 802.11 wireless! There was a major performance difference between the networks with sequential I/O and large blocks of course.
I spent last weekend running a number of performance tests (using IOMeter) against a couple of iSCSI targets on a oldish server and there's a performance issue that I just don't understand. The initiator is MS's software on W2K3 R2. I hooked up five 10k spindles and a dedicated RAID box (RAID0) to the server's U160 SCSI HBA. Running IOMeter locally on the server, performance was pretty good for the hardware, about what I'd expect -- with four threads & 4k chunks & 100% random & 50% R/W, an avg req time of 7 mS and a few hundred IOPS. Using the Starwind target with a small RAM disk to measure the raw end-to-end iSCSI performance the numbers were very good (avg req time of 1-4 mS), regardless of the network.
But When I "marry" the storage to the iSCSI target software, instead of getting req delays of A+B = 7+4 = 11 mS, I see delays of 22 mS. With the same target hardware running Openfiler the avg delay was 26 mS.
Very interestingly, the speed of the network had almost no impact on performance unless I was testing with large chunks and sequential reads. With random I/O, I got essentially the same performance using 802.11g wireless as I did using 1GBit hardwired! The RAID had 128MB of writeback cache and was using cached I/O. The drives were using command tag queuing.
Why oh why is there such an added "penalty" latency when combining the storage with the iSCSI target application? I was testing with consumer network gear at home, but I got about the same numbers using a dedicated HP commercial switch at work. Also roughly the same numbers when using an LSI RAID card in the server instead of the external RAID. I want to build a moderate performance iSCSI SAN for use with a couple of Exchange 2007 boxes, but I'm not going to bother if I have to take a huge penalty hit for using iSCSI instead of cobbling up some local storage.
For 95% of what I listen to, the difference is minimal or nonexistant, but for the other perhaps 5% of my music library
What software and technique do you use to convert your CDs to mp3? I use EAC to grab the.wav files and LAME to encode them (256kbps CBR). Using headphones there are only a couple of tracks that I think I can tell the difference between the.wav and.mp3. Interestingly, one of the tracks is a decent but not great quality recording of a live rock concert and the applause doesn't sound exactly the same. I wonder if the original recording was processed with a weird effect that makes the resulting wav file hard to encode?
MS is like the poor cat who has been caught numerous times trying to get into the birdcage. The cat knows better, and knows that it's going to get caught again, but it just can't help it: it's the nature of a cat. MS is the chastised cat, creeping ever so slowly and stealthily toward the birdcage. The cat is completely out in the open and can be seen by everyone, but hopes that no one will notice if it moves very, very slowly. The cat's plan is simple. Embrace the Tweety-Bird, Extend its claws, and Extinguish the life of the creature. Psssst. Hey, cat. The whole world is watching.
i can tell you the difference between 320kbps mp3 and a cd
That's interesting. I'm not trolling, but I am interested in the methodology of your A/B listening test. Are you comparing the accurately ripped raw.wav from your CD with the carefully encoded 320kbps.mp3 of that file, played through the same quality soundcard into your sound system? And normalized down from 99% or 100% levels if needed?
I can hear a difference between some CDs played through my 2nd generation player and the.wav files played through the soundcard in my jukebox PC, because the D/A in my soundcard is different (and better) than the D/A in my CD player -- especially for high hats and some piano. This is on a decent late '80's sound system (Onkyo Grand Integra and Infinity Kappa 8s). So far*, I cannot hear any difference between the.wav files and carefully encoded 256kbps CBR.mp3 files, even on 1812 or Adagio for Strings. FWIW, the sound card is a Turtle Beach Santa Cruz (NO SCO jokes, please).
I love the local school-run dance/pop FM station but I can't stand to listen during the day when they take music that's already compressed a lot and compress it to death and boost the high end into noise...
* There is one CD that almost sounds just a tiny shade different. It's a live rock concert recorded on who-knows-what equipment back in '80-something (Robin Trower - In Concert [King Biscuit Flower Hour Presents]) and I usually listen to the.wav files of those songs.
Yes, and the tech could have a very detrimental impact in certain segments of the garment industry. It's hard enough as it is to get those damn Casimir goats herded together for shearing...
Over the years, various brands and models of drives have had less than stellar reliability. Toshiba MK-134, Quantum "Bigfoot" Furball, IBM DeathStar, and many other turkeys (it'll die once I type this, but my 45GB PATA DeathStar has been running great since 2000 or 2001. Yes, all data on it is replicated to two other physically separated drives).
I've never managed to get a WD drive to work in my main PC at home, though we've used various WD drives at work. At the moment my newer PC drives are all Seagate and Maxtor, and the 160GB perpendicular drive in my notebook is a Hitachi.
You can get a discount on IBM-branded notebooks? Go to Lenovo.com, click on products, and drill down to the T60p under "ThinkPad T Series mobile workstations".
It officially supports Novell Suse Linux Enterprise Desktop 10, which is arguably less evil than MS Vista. And of course it only comes with "DOS entitlement" so you have to install Linux yourself.
> you have to agree to anything that gets put in there or risk losing the right to use that code?
No. You can use it all you like. You just can't distribute (or promulgate) it without accepting the license.
If it's Vx-or-later you can still distribute it under Vx, even if Vy is the current license. You just can't change the license from Vx-or-later to Vx-only when you distribute it, if you are not the author. If it's a combo of Vx-or-later and Vy, then of course you can't distribute it as Vx-only.
caveat: I'm basing this reply on my knowledge of current common laptop hard drive prices. I have no idea what drives cost for Macbooks. When you terminate your employment, remove your hard drive and give it to the company. Tell them to shred it or recycle it or whatever they want. It's got to be less expensive than paying for a security consultant, yes?
Slashdot Mirror
Wait until everyone leaves on holiday some unusually hot 4th of July morning. The earlybirds are fine, but those leaving later have empty "tanks" because ConEd sucked out all their battery power to run all of the air conditioners.
Thanks for the reply. My puzzlement is that when measuring the pure iSCSI performance (with a ramdisk on the target) performance was great, regardless of the network -- avg I/O delays of 1mS with one IOMeter worker thread and 4mS with four worker threads... and a few hundred IOPS. This is all random I/O and the total throughput in MB/s is fairly low and not important for my app.
When the iSCSI target was disconnected from the ramdisk and connected to the real drives, the avg I/O delays were considerably worse than just adding the pure iSCSI overhead (1-4 mS) and the local performance on the target machine (7 mS). Instead of about 8-11 mS I get an avg I/O delay of 22-26 mS.
This is a simple "SAN" with one initiator and one target, no MPIO, etc. For pure random I/O there was no performance difference between a dedicated HP ProCurve, consumer 100MBit, consumer 1GBit, and consumer 802.11 wireless! There was a major performance difference between the networks with sequential I/O and large blocks of course.
I spent last weekend running a number of performance tests (using IOMeter) against a couple of iSCSI targets on a oldish server and there's a performance issue that I just don't understand. The initiator is MS's software on W2K3 R2. I hooked up five 10k spindles and a dedicated RAID box (RAID0) to the server's U160 SCSI HBA. Running IOMeter locally on the server, performance was pretty good for the hardware, about what I'd expect -- with four threads & 4k chunks & 100% random & 50% R/W, an avg req time of 7 mS and a few hundred IOPS. Using the Starwind target with a small RAM disk to measure the raw end-to-end iSCSI performance the numbers were very good (avg req time of 1-4 mS), regardless of the network.
But When I "marry" the storage to the iSCSI target software, instead of getting req delays of A+B = 7+4 = 11 mS, I see delays of 22 mS. With the same target hardware running Openfiler the avg delay was 26 mS.
Very interestingly, the speed of the network had almost no impact on performance unless I was testing with large chunks and sequential reads. With random I/O, I got essentially the same performance using 802.11g wireless as I did using 1GBit hardwired! The RAID had 128MB of writeback cache and was using cached I/O. The drives were using command tag queuing.
Why oh why is there such an added "penalty" latency when combining the storage with the iSCSI target application? I was testing with consumer network gear at home, but I got about the same numbers using a dedicated HP commercial switch at work. Also roughly the same numbers when using an LSI RAID card in the server instead of the external RAID. I want to build a moderate performance iSCSI SAN for use with a couple of Exchange 2007 boxes, but I'm not going to bother if I have to take a huge penalty hit for using iSCSI instead of cobbling up some local storage.
What software and technique do you use to convert your CDs to mp3? I use EAC to grab the
MS is like the poor cat who has been caught numerous times trying to get into the birdcage. The cat knows better, and knows that it's going to get caught again, but it just can't help it: it's the nature of a cat. MS is the chastised cat, creeping ever so slowly and stealthily toward the birdcage. The cat is completely out in the open and can be seen by everyone, but hopes that no one will notice if it moves very, very slowly. The cat's plan is simple. Embrace the Tweety-Bird, Extend its claws, and Extinguish the life of the creature. Psssst. Hey, cat. The whole world is watching.
That's interesting. I'm not trolling, but I am interested in the methodology of your A/B listening test. Are you comparing the accurately ripped raw
I can hear a difference between some CDs played through my 2nd generation player and the
I love the local school-run dance/pop FM station but I can't stand to listen during the day when they take music that's already compressed a lot and compress it to death and boost the high end into noise...
* There is one CD that almost sounds just a tiny shade different. It's a live rock concert recorded on who-knows-what equipment back in '80-something (Robin Trower - In Concert [King Biscuit Flower Hour Presents]) and I usually listen to the
Over the years, various brands and models of drives have had less than stellar reliability. Toshiba MK-134, Quantum "Bigfoot" Furball, IBM DeathStar, and many other turkeys (it'll die once I type this, but my 45GB PATA DeathStar has been running great since 2000 or 2001. Yes, all data on it is replicated to two other physically separated drives).
I've never managed to get a WD drive to work in my main PC at home, though we've used various WD drives at work. At the moment my newer PC drives are all Seagate and Maxtor, and the 160GB perpendicular drive in my notebook is a Hitachi.
You can get a discount on IBM-branded notebooks? Go to Lenovo.com, click on products, and drill down to the T60p under "ThinkPad T Series mobile workstations".
It officially supports Novell Suse Linux Enterprise Desktop 10, which is arguably less evil than MS Vista. And of course it only comes with "DOS entitlement" so you have to install Linux yourself.
But, still, it isn't Vista.
> you have to agree to anything that gets put in there or risk losing the right to use that code?
No. You can use it all you like. You just can't distribute (or promulgate) it without accepting the license.
If it's Vx-or-later you can still distribute it under Vx, even if Vy is the current license. You just can't change the license from Vx-or-later to Vx-only when you distribute it, if you are not the author. If it's a combo of Vx-or-later and Vy, then of course you can't distribute it as Vx-only.
caveat: I'm basing this reply on my knowledge of current common laptop hard drive prices. I have no idea what drives cost for Macbooks. When you terminate your employment, remove your hard drive and give it to the company. Tell them to shred it or recycle it or whatever they want. It's got to be less expensive than paying for a security consultant, yes?