18 posts • joined Monday 5th October 2009 14:00 GMT
Re: AC Destroyed All Braincells Gordon BTRFS? You must be joking...
Just read 'Ten Reasons Not To Use ZFS'.
2. No support for SELinux ACLs
3. Primarily a Solaris product so it costs money (?)
4. Designed for servers not desktops (WTF?)
5. Linux has Reiser4 (ROTFL)
6. Sun wrote it and they're evil for writing Java
7. No independent benchmarks
8. Microsoft and Sun work together
9. Sun paid SCO for the IP suit
10. No 64-bit Mozilla Java plugin
None of those seem like valid reasons to me. The whole thing seems like rabid fanboyism.
The last sentence - "IMHO any news
from Sun is unwelcome, unless that news is the wholesale GPL
re-licensing of their entire product catalogue."
but (this is hardly a stretch) perhaps that 'second hand' hard drive came from a PC that was nicked from the care home? If these people are as unscrupulous as they appear then it wouldn't be surprising if they were using stolen goods as cheap sources of parts...
Some more maths for you
working set used by the ZFS stuff was 15x greater
So do you think it's fair to say that IOPs/GB can scale by the same amount?
Or do you want to find another strange unit to measure things by?
Sorry, I made a little mistake in my post:
The V7000 I linked to actually only uses 10K drives.
Sadly, the extra 12K IOPS that you'd (theoretically) gain from 15K disks doesn't quite account for the difference.
all those who think you can compare these benchmarks are utterly, utterly wrong. Chris Mellor, equally you should be ashamed of yourself.
Read the summary:
Total ASU (Application Storage Unit) Capacity represents the total storage capacity read
and written in the course of executing the SPC-1 benchmark.
For the ZFS appliance:
Addressable Storage Capacity
For the V7000:
Addressable Storage Capacity
So the working set used by the ZFS server is FIFTEEN TIMES greater.
Perhaps we should be comparing http://www.storageperformance.org/benchmark_results_files/SPC-1/IBM/A00103_IBM_Storwize-V7000/a00103_IBM_Storwize-V7000_2-node_SPC1_executive-summary.pdf
Which is a 2-node Storwize V7000, configured with 240 15K-RPM drives, addressing:
Addressable Storage Capacity
Aha. Now we can compare that to the ZFS box... So what are the figures for this one?
SPC-1 IOPS 53,014.29
SPC-1 Price-Performance $7.52/SPC-1 IOPS™
Total ASU Capacity 24,433.592 GB
Data Protection Level Protected (Mirroring)
Total TSC Price (including three-year maintenance) $389,425.11
Maybe one of the Oracle employees on here should have pointed that one out? Just sayin'
Hmm lots of fail in this comment as well - you should RTFA properly...
Your figures are based upon reverse osmosis providing ALL the water supplied to TW's customers - I don't think at any point the reporter suggested that Thames Water stop drawing water from aquifers and other reservoirs - conversely he suggested that doing so would cause flooding of tube tunnels due to raising of the water table.
Never mind that you have magically turned £40 into £50 for your 'killer bit'. Also did you happen to notice how much TW have spent stopping leaks, digging up roads etc. Scaling that back should comfortably cover the capex for this - and before you moan that leaks are wasteful, the infrastructure is over 100 years old and has been neglected - that's just how it is - and the cost of all that digging is being added to your bill, anyway...
The point of the article was to highlight that it would be possible to use desalinisation plants to make up for low rainfall. Don't you think that regions would benefit from not having to endure drought conditions, standpipes etc?
yes very small
I am yet to come across a Production system which constantly installs, removes and compiles software as its primary purpose.
I would suggest that the 'build server' which you describe is very much an edge case.
That one's Akamai. Shows up when an origin server is struggling under the load and doesn't respond quickly enough
PAEing the price
sure windows can do PAE from 2003 and up but then you have to pay M$ for the Enterprise version of the OS. :(
this even applies to the 2008 version which I wouldn't describe as ancient.
see post above
E.G. a 32-bit OS which will only support up to 3.5GB RAM (Windows, I'm looking at you) so can only have a tiddly disk cache. Never mind that the stupid OS (Windoze again) constantly tries to free RAM by swapping stuff to disk.
Stuff that on a server with >4GB of RAM and you are wasting it.
Virtualise it on a server with >4GB of RAM and even if there's some overhead the hypervisor can use all that extra RAM as a disk cache. Shimples.
zfs would have made this too easy and would have allowed the copy to run at almost network wire-speed (all block-level, you see). you could take a filesystem snapshot and send it over the wire (rsh,ssh) and import it at the other end. I have done this more times than I care to count with filesystems containing up to 30 million files. it also actually defragments the files in the process (currently pretty much the only way you can do that with ZFS)
of course you would have had to copy all the files from the NTFS filesystem to ZFS first though. :facepalm:
I have set up a system using Solaris 10, samba+winbind to store in excess of 100 million files in a multi-user Active Directory environment on ZFS. All kerberized and shared over the network. It even supports NFSv4 ACLs (requires patching samba a bit though).
I will be moving this from one server to another next week with a single command. Nice.
Tomcat's relationship with the hypervisor
I'm struggling to come up with any ways in which TC Server can have any form of relationship with the hypervisor. Given that traditionally, there is a JVM and a (possibly paravirtual) OS in the way, they would surely have to come up with some sort of JVM-pvOS hybrid?
Alternatively, here's an idea - why don't they just ditch the hypervisor and run up some JVMs on Redhat?!?!?!?!
I have got to say that VMWare's purchase of SpringSource ranks right up there as one of the most puzzling ever for me...
@Mr Ed Rowley
He misses the point somewhat - since the collection of credit card details would require the company to be PCI compliant, there should be no storage of the CVV number and the credit card number should be encrypted and stored separately from the keys.
This error would suggest that the app processing the credit card details is the same one that sends out the emails. This would almost certainly result in a PCI non-compliant setup.
I think that Argos might have a visit from the bank's auditors in the near future...