Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, Dec 08, 2007 at 01:07:35 +0000,
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? > I see there is quite a lot of work > in getting ECC testing incorporated into the Linux kernel. > But even if it were there, would it be very valuable? > > I have a feeling that disk errors are far more likely > than RAM errors. > Is that right? A google search should turn up some pages on this. What I read suggests it is worth getting. A bit flips were estimated to be on the order of something like one a month (depending on memory size). And that single bit flips were the most common type of corruption and that can be corrected (not just detected) by ECC memory. Whether or not it is worth it to you depends on what you are using the machine for. -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
Is ECC memory any use?
On Sat, 08 Dec 2007 01:07:35 +0000
Timothy Murphy <tim@birdsnest.maths.tcd.ie> wrote: > > I'm getting memory for a very old (P2B-LS) Asus motherboard, > and I see I can get ECC memory for some 20% more. > > Is there any point in getting this? It depends what for. ECC memory will self correct errors (the ECC framework in the kernel just deals with nice user space reporting of this and ECC unrecoverable errors). If you are doing complex scientific processing or banking where an error isn't acceptable then its essential. If you can tolerate the small change of a bit flip ever year or two then its far less of a concern. It also depends on the amount of RAM - more ram = more change of flips. Disk errors should be different - the odds of an undetected disk error are minute. The odds of catastrophic failure obviously are rather more clear. Alan -- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list |
| All times are GMT. The time now is 04:57 AM. |
VBulletin, Copyright ©2000 - 2013, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.