Message boards :
Technical News :
Warner (Mar 11 2009)
Message board moderation
Author | Message |
---|---|
Matt Lebofsky Send message Joined: 1 Mar 99 Posts: 1444 Credit: 957,058 RAC: 0 |
Lots of machine rebooting today as Eric is getting his new hydrogen server online, and I'm finishing work on moving mail servers around. This shouldn't have affected the outside world. During all this Eric gave Jeff and I a quick tutorial on merged file systems. Wacky stuff. Radar wise, I got some lengthy notes from Phil down at Arecibo. Turns out by far most of the radar we see is from the airport, which was news to me, and that's the only thing the hardware blanker checks for. Discussions will continue. Dan, while at Arecibo earlier this week, replaced our non-working raw data drive enclosure with one we've been using up here. It's unclear whether this helped or not. We're learning that SATA drives (and enclosures/backplanes) aren't necessarily meant for excessive hot-swapping, and will fail after N "mating cycles." This may be what we're coming up against. - Matt -- BOINC/SETI@home network/web/science/development person -- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude |
Olli Send message Joined: 25 Apr 07 Posts: 143 Credit: 2,089,162 RAC: 0 |
Please elaborate wacky stuff? |
P. J. Crabtree Send message Joined: 17 Jan 07 Posts: 22 Credit: 1,847,766 RAC: 0 |
I fail after 'N' mating cycles |
Norwich Gadfly Send message Joined: 29 Dec 08 Posts: 100 Credit: 488,414 RAC: 0 |
Would drives which connect via the USB be better ? |
Borgholio Send message Joined: 2 Aug 99 Posts: 654 Credit: 18,623,738 RAC: 45 |
Would drives which connect via the USB be better ? They may be more reliable, but they are also 3 - 10x slower depending on the speed of the sata drives. You will be assimilated...bunghole! |
Norwich Gadfly Send message Joined: 29 Dec 08 Posts: 100 Credit: 488,414 RAC: 0 |
Would speed be such an issue if you had a PC dedicated to transferring data from the USB drives to faster drives on the network ? |
alexnuke Send message Joined: 24 Nov 02 Posts: 3 Credit: 160,483 RAC: 0 |
We're learning that SATA drives (and enclosures/backplanes) aren't necessarily meant for excessive hot-swapping, and will fail after N "mating cycles." Isn't e-SATA designed exactly for this reason, to quote wikipedia ;):
- Alex |
Westsail and *Pyxey* Send message Joined: 26 Jul 99 Posts: 338 Credit: 20,544,999 RAC: 0 |
Not sure on this... I find it curious to know if these connectors are not as robust as thought... I draw connection with CD..In the 80's I was told CD last forever. Yet, I have 1ks of cdrs from mid nineties-today who no longer work.. *puzzled* p.s. about cd's...just to clarify, not format/os issue they have physically degraded. also many not water resistant, lol. Leave in car in rain and ruined cds. Speaking only of cdrs here commercial disks not suffer same issues...so anyway sorry, umm sata connectors? "The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, 'hmm... that's funny...'" -- Isaac Asimov |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 31006 Credit: 53,134,872 RAC: 32 |
Not sure on this... A reporter somewhere got pressed CD's mixed up with CDR's and that is how the urban legend of CDR's lasting any length of time got born. CDR's use a dye. To quoth Kodak, like all dyes they fade over time. As to the connectors, SATA connectors are different than e-SATA connectors. |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
Would speed be such an issue if you had a PC dedicated to transferring data from the USB drives to faster drives on the network ? Wouldn't make a difference. Either way, it would be considerably slower getting the data off the USB drive and onto a machine, any machine. |
PhonAcq Send message Joined: 14 Apr 01 Posts: 1656 Credit: 30,658,217 RAC: 1 |
I think the point was that if USB could fill the buffer faster than the system draws from the buffer, it would be be a viable option. And if the USB approach is more reliable, then it would make sense to switch to it. But this seems too obvious to debate, really. |
Jack Zhang Send message Joined: 2 Jul 06 Posts: 206 Credit: 6,142,449 RAC: 0 |
SATA connectors are almost as brittle as ZIF connectors, bend it up or down too much and the plastic in the middle will snap. What if Fiction was Fact and Fact was Fiction and vice versa? |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14679 Credit: 200,643,578 RAC: 874 |
Would speed be such an issue if you had a PC dedicated to transferring data from the USB drives to faster drives on the network ? Is there any such thing as a USB drive, or just a SATA/PATA drive in a USB enclosure? [That's what I use for data recovery from drives which won't boot any more]. Remember we're talking 500GB and up, so it's not like a memory stick. From what Matt's posted in the past (remember that bit of pink packing foam?), they ship just the bare drives to and from Arecibo: if they had to ship drives in enclosures, they'd need a bigger box, just to start with.... |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
I think the point was that if USB could fill the buffer faster than the system draws from the buffer, it would be be a viable option. And if the USB approach is more reliable, then it would make sense to switch to it. But this seems too obvious to debate, really. Yeah, I caught that point. My point is that we don't know that it will fill the buffer faster faster than the system can drain it, or how they do the transfer in the first place (it may not be direct to buffer), but that any which way you look at it, it will be slower. If they fill direct to buffer, we don't know how fast the server draws from this buffer. If they fill to a cache before the direct buffer, it could still be fast enough, but that it would take longer to fill this cache, and hence someone would have to find more multitasked options to fill their time while waiting for the drive to empty. I'm not saying that it isn't worth trying, it may very well be a more reliable option that should be looked at, but it will be a slower one regardless. But I guess everything has to be a debate, or anyone stating an observational fact must be an obstructionist and trying to shoot down a good suggestion. |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
Would speed be such an issue if you had a PC dedicated to transferring data from the USB drives to faster drives on the network ? Ummm... we need to keep in mind the two separate connections: the Universal Serial Bus and the Serial/Parallel "AT" Attachment interface. Yes, the "USB drives" are typically just SATA/PATA drives in an enclosure, but the enclosure utilizes the Universal Serial Bus connection to transfer data, which, due to overhead and design, is slower than raw SATA/PATA, so regardless if the drive inside is a SATA/PATA, it is using a slower bus to move the data around while inside that enclosure. So yes, it would still be slower to transfer off the USB than it would by using eSATA (which does not have the overhead of USB since it is designed for HDDs and not a "universal" connection interface for all peripherals) or even Firewire "B" (IEEE 1394B). [Edit] The other questions are: does the speed difference really matter? Which is more important: speed or reliability? I cannot answer those questions. I can only say that I know for a fact that USB is slower than eSATA, Firewire & raw SATA/PATA because it was not designed as an interface for transferring large amounts of data, especially off an external mass storage device. |
Norwich Gadfly Send message Joined: 29 Dec 08 Posts: 100 Credit: 488,414 RAC: 0 |
You could copy from several drives simultaneously to offset the slower transfer rate, so that in a given elapsed time the amount of data transferred would be the same. The number of drives required would increase if the longer connect time caused them to miss the deadline for shipping for the next cycle. |
Westsail and *Pyxey* Send message Joined: 26 Jul 99 Posts: 338 Credit: 20,544,999 RAC: 0 |
How about an IEEE 1394 raid cluster with it's own ps+case? Then can ship whole enclosure an no need for hot swapping. Just musing outloud here.. "The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, 'hmm... that's funny...'" -- Isaac Asimov |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
If the issue is the connector, then the solution has to include not cycling the existing SATA connector. In other words, whatever is used to convert SATA to USB has to be permanently attached. If you go to native USB drives, the connector on the drive (whatever it is) has to be rated for the relevant number of cycles. ... or all you've really done is move the problem. |
Cosmic_Ocean Send message Joined: 23 Dec 00 Posts: 3027 Credit: 13,516,867 RAC: 13 |
I just got an external 5-bay hotswap enclosure. uses SATA and has a built-in port multiplier connected to the host system with eSATA. Works really well for me. Has its own 300w PSU and is about the size of a SFF box. Linux laptop: record uptime: 1511d 20h 19m (ended due to the power brick giving-up) |
Dr. C.E.T.I. Send message Joined: 29 Feb 00 Posts: 16019 Credit: 794,685 RAC: 0 |
. . . Thanks for the Updates Matt - Accolades to Each of you over there @ Berkeley ps - Warner eh - as in Bros. ;) BOINC Wiki . . . Science Status Page . . . |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.