3Ware 7500 on Linux experiences
on
IDE RAID Examined
·
· Score: 2, Informative
At our business we use the 3Ware 7500 with a 3x 120GB (1200JB) Western Digital RAID5 configuration on Linux RedHat 7.2. The machine is a dual P3/1GHz on a SuperMicro 370DER motherboard. We use the machine as our primary file server/compilation box, so data integrity and fast failure resolution is critical.
The cited benchmark page has excellent information (130 graphs!), and it confirms my first hand experiences of everyday use of the 3Ware 7500. The read times in RAID5 are outstanding, but there are sometimes significant delays on file creation. In addition, it seems that IO is single piped, or serialized on writes at times.
Since the 3Ware 7500 is based on an FPGA (Field Programmable Gate Array, (see http://www.xilinx.com/), with LOTS of extra ECC processing power, the problem with writes is not likely to be related to the "CPU", but rather part of the internal IO structure of the card. I hope it is amenable to correction with a microcode upgrade to the FPGA, but it may be related to the memory architecture of the card.
Overall, I'm pretty happy with the controller, especially the ease of installation and rebuild time. I have high confidence in the data integrity, and the price is good. I also recommend the drive cage that 3Ware sells, even though it is expensive at $200, it's well worth it because it fits 3 drives in the space of 2 5 1/4 inch bays, and it is hot swap proven.
I'd like to see a shoot-out between the 3Ware and some SCSI/Fibre Channel RAID 5 configurations!!!
The most efficient way to leave earth orbit is to build and fuel spacecraft from earth orbit and launch from there. The design of current spacecraft is dominated by the ascent phase needed to clear the Earth's gravitational well.
However, NASA has been acting purely as a jobs program for the last 30 years, lacking vision of any but the purely political kind. The great majority of the NASA budget (some $3.4B a year!) is spent on space shuttle operations, not on research or visionary projects. The exploration missions get buried under the political weight of all those operations workers scattered across 50 states when budgets are developed.
Originally, the shuttle was meant to provide a cost effective means to develop vehicles that would launch from high earth orbit to explore and colonize Mars. Somewhere along the way the feeding frenzy began and NASA became just another sad beaurocracy.
I turned away from aerospace in 1994 after I sat with astronaut Guion Albert at an AIAA dinner, where we heard the NASA director of Aeronautics speak on the future of NASA. His name was Wes Harris and his vision consisted entirely of educating the underserved and enhancing their opportunities. This was the last straw for me and many others who looked to NASA to build the future in space.
Perhaps the recent amateur and commercial efforts in space vehicles like Armadillo Aerospace will give us the long awaited vision and excitement about our future in space travel...
Shuttle introduced one called the SS41G that has an AGP slot and sockets an AMD Athlon XP CPU. I have one running now as my media PC connected to my HDTV and it's running beautifully. I put an Athlon 1400 in it and the CPU temperature is a cool 31C.
One gripe - it does not have a connector for the SPDIF output of a DVD player on the motherboard. This means that SPDIF pass-through will not work, and the Dolby AC3 track must be processed by the CPU instead of just sent directly to the stereo receiver.
BTW - I also have three SS51G machines with Pentium 4 2.53GHz CPUs running as database caching servers (Linux RH7.2). They've run without any hiccups under load for several months now. Great performers, and I think the SB51G should be similar.
Understood from the beginning - this is much like DC-X, which did not feature stabilizing fins.
Also, the instability was probably due to a long list of things, possibly most important being the failure of the flight control computer.
However - some aero stability work would only help the control system do it's job, and the current design just looks highly unstable. Even large angular deviations would be immensely helped by fins, and unless you need huge roll _rates_, which come with commensurately huge control power requirements, why fly without the fins?
BTW - I'm a big fan of this effort, my armchair quarterbacking comes purely from a desire to help. In fact, If I get time I'd like to do some gratuitous CFD work for them...
I concur, this design should be fixed to make it stable for moderate angular deviations. They can still fly it unstable, but only if they have the control power necessary to correct for reasonable deviations from vertical. It looks like the thing went completely unstable due to the small deviation on liftoff.
Presuming that they did plan enough control power to correct for these instabilities, the other alternative could be that one corner nozzle was damaged on lift-off.
Four large fins off of the bottom flange should be enough to remove the instability I think. They have to be really large for that squat a body though. Why not lengthen the rocket guys?
Slashdot Mirror
At our business we use the 3Ware 7500 with a 3x 120GB (1200JB) Western Digital RAID5 configuration on Linux RedHat 7.2. The machine is a dual P3/1GHz on a SuperMicro 370DER motherboard. We use the machine as our primary file server/compilation box, so data integrity and fast failure resolution is critical.
The cited benchmark page has excellent information (130 graphs!), and it confirms my first hand experiences of everyday use of the 3Ware 7500. The read times in RAID5 are outstanding, but there are sometimes significant delays on file creation. In addition, it seems that IO is single piped, or serialized on writes at times.
Since the 3Ware 7500 is based on an FPGA (Field Programmable Gate Array, (see http://www.xilinx.com/), with LOTS of extra ECC processing power, the problem with writes is not likely to be related to the "CPU", but rather part of the internal IO structure of the card. I hope it is amenable to correction with a microcode upgrade to the FPGA, but it may be related to the memory architecture of the card.
Overall, I'm pretty happy with the controller, especially the ease of installation and rebuild time. I have high confidence in the data integrity, and the price is good. I also recommend the drive cage that 3Ware sells, even though it is expensive at $200, it's well worth it because it fits 3 drives in the space of 2 5 1/4 inch bays, and it is hot swap proven.
I'd like to see a shoot-out between the 3Ware and some SCSI/Fibre Channel RAID 5 configurations!!!
Comments?
However, NASA has been acting purely as a jobs program for the last 30 years, lacking vision of any but the purely political kind. The great majority of the NASA budget (some $3.4B a year!) is spent on space shuttle operations, not on research or visionary projects. The exploration missions get buried under the political weight of all those operations workers scattered across 50 states when budgets are developed.
Originally, the shuttle was meant to provide a cost effective means to develop vehicles that would launch from high earth orbit to explore and colonize Mars. Somewhere along the way the feeding frenzy began and NASA became just another sad beaurocracy.
I turned away from aerospace in 1994 after I sat with astronaut Guion Albert at an AIAA dinner, where we heard the NASA director of Aeronautics speak on the future of NASA. His name was Wes Harris and his vision consisted entirely of educating the underserved and enhancing their opportunities. This was the last straw for me and many others who looked to NASA to build the future in space.
Perhaps the recent amateur and commercial efforts in space vehicles like Armadillo Aerospace will give us the long awaited vision and excitement about our future in space travel...
One gripe - it does not have a connector for the SPDIF output of a DVD player on the motherboard. This means that SPDIF pass-through will not work, and the Dolby AC3 track must be processed by the CPU instead of just sent directly to the stereo receiver.
BTW - I also have three SS51G machines with Pentium 4 2.53GHz CPUs running as database caching servers (Linux RH7.2). They've run without any hiccups under load for several months now. Great performers, and I think the SB51G should be similar.
Kudos to Shuttle!
Also, the instability was probably due to a long list of things, possibly most important being the failure of the flight control computer.
However - some aero stability work would only help the control system do it's job, and the current design just looks highly unstable. Even large angular deviations would be immensely helped by fins, and unless you need huge roll _rates_, which come with commensurately huge control power requirements, why fly without the fins?
BTW - I'm a big fan of this effort, my armchair quarterbacking comes purely from a desire to help. In fact, If I get time I'd like to do some gratuitous CFD work for them...
I concur, this design should be fixed to make it stable for moderate angular deviations. They can still fly it unstable, but only if they have the control power necessary to correct for reasonable deviations from vertical. It looks like the thing went completely unstable due to the small deviation on liftoff.
Presuming that they did plan enough control power to correct for these instabilities, the other alternative could be that one corner nozzle was damaged on lift-off.
Four large fins off of the bottom flange should be enough to remove the instability I think. They have to be really large for that squat a body though. Why not lengthen the rocket guys?