Slow? polyhedron comparison

I ran the Polyhedron (http://www.polyhedron.com) set of general
fortran benchmarks as a test of code speed. Some 64bit ran slower
than I expected, which I've not investigated yet. Polyhedron ran all the
ones on the left with an Opteron 2.4G dual cpu, with 333 sdram.
This is similar to our older Sun v2100z's
Polyhedron only tests Linux compilers; I tested Sun Studio11
with patches to late May, on Linux/Solaris, with Lahey as a check.
My Lahey number is within about 6% of their number. I didn't try very
hard to find the best compile options, unlike Polyhedron. I just
used -fast for Studio11.
times in seconds per iteration; larger is slower.
The solaris 10 numbers are derived from a 2.8GHz system,
scaled to compare with 2.4GHz; the dram speed was 400 instead
of 333 so it is not quite apples and apples.
                  from www.polyhedron.com                                 run at UTIG
           ---------------- linux --------------------           ----- linux -----   solaris-10
name     Absoft g95    Intel Lahey NAG    Paths PGI          Lahey sunl32 sunl64 suns32 suns64
AC       10.44 21.25 13.60 24.59 27.37 10.53 14.28        21.48 19.96 40.23 13.60 35.65
AERMOD   38.63 48.19 28.67 48.96 58.67 37.65 35.22        54.00 37.27 47.13 34.58 31.30
AIR      14.46 23.45 12.32 17.66 16.49 15.00 16.79        17.81 14.12 13.02 14.12 12.64
CAPACITA 71.41 111.24 104.03 124.90 119.22 71.02 71.84        130.29 106.54 96.71 103.8 97.06
CHANNEL 17.74 27.06 13.09 17.76 15.93 18.14 15.98        22.67 11.62 11.53 12.72 13.45
DODUC    42.42 52.73 36.60 57.65 57.16 40.91 37.82        51.72 45.21 46.23 42.08 37.76
FATIGUE 6.60   37.83 10.64 20.11 21.79 6.38 11.47        18.45 8.84   8.59   8.80   8.318
GAS_DYN 6.35   22.10 7.84   15.07 21.56 8.73 8.55         19.32 13.97 13.62 13.45 12.81
INDUCT   33.11 49.67 44.24 91.52 60.16 30.56 38.37        81.90 44.82 47.25 44.12 42.39
LINPK    27.81 29.43 27.19 27.03 27.23 27.57 29.02        35.43 23.21 22.92 25.83 26.90
MDBX     20.02 27.10 20.49 25.81 22.33 20.19 21.48        21.96 18.45 18.78 18.58 18.86
NF       26.55 51.07 28.59 42.79 30.69 25.76 30.55        56.89 28.40 28.06 27.32 28.26
PROTEIN 51.67 75.13 49.06 90.58 65.38 51.31 63.65        90.69 64.70 65.03 65.39 65.55
RNFLOW   31.71 55.24 44.40 46.83 44.85 29.80 42.57        43.39 39.84 38.79 34.97 35.82
TEST_FPU 19.33 32.09 20.09 24.30 23.55 19.66 20.94        29.77 18.26 18.42 19.84 19.39
TFFT     8.12   8.65   8.48   8.76   8.40   8.30 8.74         10.84 8.31   7.95   8.30   8.271
GeoMean 21.19 35.84 22.77 33.06 31.37 21.38 24.11        35.04 24.37 25.47 23.64 24.51
   ratio   1      1.69   1.07   1.56   1.48   1.00 1.13         1.65   1.15   1.20   1.11   1.15
The geometric mean ratios suggest that the Sun linux -alpha- compiler may already be close enough to
the best result now available to make it a strong contender.
regards,
mw

What do you expect?! Oracle 9i and Oracle 8i are different particularly by optimizer cost-based(9i) and rule-based(8i) :)
what about both execution plans?

Similar Messages

Why is my refurb MBP 2011 so slow in comparison to an identical model?

Hello dear community.
I purchased my macbook pro early 2011 from the refurb store about four months ago.
When I use a browser and a few (1 - 3) regular programs (office, itunes, acrobat) along the side the OS sometimes starts lagging.
At first I was using chrome and as soon as I had it running for a while or with more than a few tabs i found it was getting really slow and i got the loading ball on the regular. I felt this couldn't really be it as the macbook should be running like a charm. I googled issues and found that in some cases chrome might be bugged and using up all the ram. Reluctantly I switched to Firefox & Safari which I both find to be lesser browsers in terms of usability. As it turns out this doesn't really solve the problem. Running aperture or final cut is still a pain and still takes so much longer than I would think.
2 weeks ago I used a friends identical macbook pro 2011 and it runs a lot smoother than mine.
Could this be related to the refurb purchase or did I do something wrong? My startup is not clogged, I always keep everything clean & tidy, I use progs such as clean my mac etc. pp.
Would love to get some advice on this
thx ahead!

Please read this whole message before doing anything.
This procedure is a diagnostic test. It won’t solve your problem. Don’t be disappointed when you find that nothing has changed after you complete it.
Third-party system modifications are a common cause of usability problems. By a “system modification,” I mean software that affects the operation of other software -- potentially for the worse. The following procedure will help identify which such modifications you've installed. Don’t be alarmed by the complexity of these instructions -- they’re easy to carry out and won’t change anything on your Mac.
These steps are to be taken while booted in “normal” mode, not in safe mode. If you’re now running in safe mode, reboot as usual before continuing.
Below are several lines of text in monospaced type, which are UNIX shell commands. They’re harmless, but they must be entered exactly as given in order to work. If you have doubts about the safety of running these commands, search this site for other discussions in which they’ve been used without any report of ill effects.
Some of the commands will line-wrap or scroll in your browser, but each one is really just a single line, all of which must be selected. You can accomplish this easily by triple-clicking anywhere in the line. The whole line will highlight, and you can then either copy or drag it. The headings “Step 1” and so on are not part of the commands.
Note: If you have more than one user account, Step 2 must be taken as an administrator. Ordinarily that would be the user created automatically when you booted the system for the first time. The other steps should be taken as the user who has the problem, if different. Most personal Macs have only one user, and in that case this paragraph doesn’t apply.
To begin, launch the Terminal application; e.g., by entering the first few letters of its name in a Spotlight search. A text window will open with a line already in it, ending either in a dollar sign (“$”) or a percent sign (“%”). If you get the percent sign, enter “sh” (without the quotes) and press return. You should then get a new line ending in a dollar sign.
Step 1
Copy or drag -- do not type -- the line below into the Terminal window, then press return:
kextstat -kl | awk '!/com\.apple/ {print $6 $7}'
Post the lines of output (if any) that appear below what you just entered (the text, please, not a screenshot.)
Step 2
Repeat with this line:
sudo launchctl list | sed 1d | awk '!/0x|com\.apple/ {print $3}'
This time, you'll be prompted for your login password, which won't be displayed when you type it. You may get a one-time warning not to screw up. You don't need to post the warning.
Step 3
launchctl list | sed 1d | awk '!/0x|com\.apple/ {print $3}'
Step 4
ls -1A /e*/mach* {,/}L*/{Ad,Compon,Ex,Fram,In,Keyb,La,Mail/Bu,P*P,Priv,Qu,Scripti,Servi,Sta}* L*/Fonts 2> /dev/null
Important: If you synchronize with a MobileMe account, your me.com email address may appear in the output of the above command. If so, change it to something like “[email protected]” before posting.
Step 5
osascript -e 'tell application "System Events" to get the name of every login item'
Remember, steps 1-5 are all drag-and-drop or copy-and-paste, whichever you prefer -- no typing, except your password.
You can then quit Terminal.

SQl query are running slow in comparison to oracle 8i

Hi everybody,
I have two systems one is Oracle 9i on Windows 2000 and
one is Oracle 8i running on Windows 2000. The database structure is the same. When I run the following qry in
Oracle 9i
SELECT COUNT (a.aint_ac_no) AS total_ac_no, adet_int_rate,
SUM (fun_max_cbal_bal (a.aint_ac_no,
a.aint_subsys_cd,
TO_DATE ('30/01/2009', 'DD/MM/YYYY')
) AS balance
FROM domst120 a, domst730 b
WHERE a.aint_interest_slab_cd = b.adet_srno
AND a.aint_subsys_cd = '137'
AND fun_max_cbal_bal (a.aint_ac_no,
a.aint_subsys_cd,
TO_DATE ('30/01/2009', 'DD/MM/YYYY')
) <> 0
AND a.aint_eff_date =
(SELECT MAX (c.aint_eff_date)
FROM domst120 c
WHERE a.aint_bank_cd = c.aint_bank_cd
AND a.aint_branch_cd = c.aint_branch_cd
AND a.aint_bank_cd = '857'
AND a.aint_branch_cd = '6'
AND a.aint_subsys_cd = c.aint_subsys_cd
AND a.aint_ac_no = c.aint_ac_no)
GROUP BY b.adet_int_rate
ORDER BY b.adet_int_rate
it takes 6 min to complete.
the same qry running in Oracle 8i takes 2 seconds
to complete.
The index structure is same in both systems also the tables
were analyzed with compute statistics.

What do you expect?! Oracle 9i and Oracle 8i are different particularly by optimizer cost-based(9i) and rule-based(8i) :)
what about both execution plans?

WHY DOES MY MACAIR DOWNLOAD SO SLOW IN COMPARISON TO MY IPHONE 6

MY MACBOOK AIR STREAMS AND DOWNLOADS LIKE GARBAGE. I HAVE A VERY FAST INTERNET SERVICE. MY IPHONES HAVE NO PROBLEM WITH STREAMS OR DOWNLOADS.. ANY CLUES?

Regardless of the speed on your iPhones, check your internet download speed > Speedtest.net The Global Broadband Speed Test
Broadband download speeds vary anywhere from 10Mpbs to 125Mpbs or more.
It may not be your Mac. Restart your router then restart your Mac. Try a download.

Slow performance when moving bunch of vector shapes

Hi there,
Photoshop CS6 Beta is soo slow in comparison to Photoshop CS5 when moving lots of vector layers. Check the videoa below please:
CS6 - http://screencast.com/t/wBOzhTySdSf
CS5 - http://screencast.com/t/ab7BDbgNxUYe
- GPU Settings Enabled in both
- GeForce GTX 560, latest drivers
- i7 3,4 GHz, 16GB RAM
Why CS6 is so slow in this situation?
EDIT: I found the Problem comes form the Layer Palete. Take a look how changing the tabs form Layers to Channels affect the moving performance : http://screencast.com/t/Gi01o1GzFPzJ
This have to be fixed, as in more complex files this will be huge problem..
Regards,
SK

Hi,
Nothing peculiar at all. Normal text laters.
Each field have subgroups with info - http://screencast.com/t/ykon4QwdBxC Overall lots of layers there, but in CS5 no problem to move them.
Still can't understand why I can move all of these better when layer palete is not displayed.
Test file - https://docs.google.com/open?id=0B0phn6xiF-cobkc3cUViM0dSV3ktYk5Vbm5Bci1TUQ
Please test it on CS5 (you will notice how smooth it is) and CS6 on both try to switch between layers palete and channels palete and try to move to see the difference. Move whole group named "Form",
Please let me know if I can help with something else to define the problem.
/I'm using Windows 7 x64, Photoshop CS6 beta x64/
Thank you!
Regards,
SK
Message was edited by: Simeon Kartov

Why is PRE 8.0 so slow on my Windows 7 installation?

I've been using Premiere Elements 3.0 on my PC with Vista for some time, and it's acceptably fast. I've now installed Windows 7 (a clean installation), and installed Premiere Elements 8. It is just soooo slow by comparison. If I load a project, which consists of four long MPG clips, it takes around 10 minutes to get to the stage where I can edit - it's 'indexing' and 'conforming' during that time. It doesn't seem to save that information, so the next time I load the project, it indexes and conforms again. I'm now trying to export as an AVI file, and it's promising me something like 20 hours, whereas a similar video in 3.0 under Vista would take only three or so hours. (The whole timeline is about an hour long.) Everything on PRE 8 seems incredibly slow.
My PC is a Dell Dimension 9200, with a Core 2 CPU at 2.4 GHz, with 4GB RAM (obviously only 3GB usable), and a RAID 0 array of two 1 TB hard disks. Even though this is not the fastest, most recent spec, other applications (eg sound editing software) run like greased lightning. For what it's worth, the 'Windows Experience Index' for my PC is 5.5, with the processor/memory/graphics/primary hard disk coming in at 5.8 to 5.9.
Any ideas why Premiere Elements 8 is so slow on my PC? Is this just an endemic problem with this version with this spec PC compared with older versions, or am I unusual - do other people with equivalent spec PCs experience greater speeds? (Although I can see that the interface is different, presumably this version of PRE is doing more or less the same job when exporting the video to AVI as the older version?)

Well, I've got multiprocessing (both of the dual cores) working, by setting the priority of the process to 'Real time'. It was set at 'normal', and seemed to be only running one core flat out at that setting. Upping the process priority to 'Real time' gets both cores close to maximum most of the time. I suspect that might be a Win 7 thing interacting with a PRE8 thing (sorry - very untechnical there!) since v3 seemed to use both cores flat out on the Win 7 Release Candidate.
It's still quite slow compared with v3, and I guess I could live with that. I've tried using MPEG Streamclip to convert the source files from MPG to AVI, and will keep going with that to see how it goes, but...
PRE8 seems to insist on conforming the source files every time I start it up, MPEG or AVI alike, which means it's five or ten minutes before I can use it. I've tried moving the location of the media cache. I've deleted the media cache files and preview files. I've also tried deleting the media cache database - and that appeared to work one time only, in that on one occasion re-opening the project in PRE8 didn't require it to conform everything again, but after that first time it wants to conform it all again. I've looked elsewhere in the forums for advice on this conforming problem, but can't find any other suggestions. When I look at the media cache, there seem to be two copies of each .cfa file - one has a suffix _1 at the end of the file name, which naively I'm asssuming means that PRE8 has decided to conform the media again but instead of overwriting the original .cfa file has created a new file with the suffix - but of course, that could be completely wrong.
The processor is an intel Core 2 6600, running Win 7 32-bit.
I'm close to giving up and uninstalling to return to version 3.0. I was worried that 3.0 might not be stable under Win 7, although it seemed OK in the Release Candidate of Win 7. This version 8 seems quite unstable on my machine, and so going back to version 3 probably couldn't be worse. However, if anyone has more good ideas I'm very willing to try them before giving up, in case it sheds any light on the problem that might help others.
Thanks for your help so far, folks. Much appreciated.
Mark

RoboHelp 7 is Very Slow Compared to 6

Any advice/suggestions on rolling back to robohelp 6 is
welcome. I find the new version clunky and slow in comparison.
Sigh,
STB

Wow, such a lot of great info - thanks everyone. Here's an
update.
Yesterday, I was working remotely. While my RH files are
local, our source control system (MS VSS) is at the office. So I
wondered if the way in which RH7 interacts with VSS was a
contributing factor. (To its credit, RH7 recognized as soon as I
opened my project that it was under source control and correctly
asked me if it could check out all the files, and did so. The
checkout wasn't any faster/slower than normal.)
That initial compile ended up taking 1 hour and 8 minutes! I
have NO idea why. But maybe the conversion (removal of kadov tags,
etc.) takes place then? Today, I came into the office so VSS is
down the hall, I'm on the LAN, etc. Opened RH7, made a couple minor
changes, and compiled. It took less than a minute!!!!
So whatever was going on yesterday, it's either resolved OR
it's something related to VSS and I'll re-encounter it the next
time I'm working remotely.
Now that I'm over the initial shock of the completely
different UI and things seem to be functioning better, I'm a MUCH
happier camper! I read the V7 info on Peter's site, took advantage
of a couple of the handy tutorials on the Adobe site (thanks for
the info and links, Peter), and I might even be PRODUCTIVE today!
kc

Slow imc COM (ActiveX)-libraries inside LabView

Hi
I'm using the imc COM (ActiveX)-libraries inside labview. The key command is the "automation open" block from LabView and selecting the desired ActiveX class.
Here his the definition of one of this function:
Function DChannel.GetDataDouble( StartIndex As Long, Count As Long,
Data As Double() ) As Long
Copies the specified samples in an array of 8Byte real numbers.
Parameters
StartIndex
Index of the 1st sample to copy. The first sample in a channel has an index of 1.
Count
Number of samples to copy.
Data
Array to receive the values. See remarks section.
Return Value
Number of valid samples in <Data>. May be smaller than <count>, when there are not enough samples in the channel (StartIndex + Count -1 > Length).
The problem and also question is why the parameter Data apears as a variant when I use a invoke node for the above method (should be double()??)
When using large Data arrays the libraries are very slow in comparison with VB6. I tested the same functions in VB6 and LabView. In VB6 the COM libraries are very fast! Does anybody know how to improve performance with COM Variants?
Thank you

I'm not sure you can do anything about the speed as long as you use ActiveX. At least that was the answer I got when I asked a similar question here. It seems that there is a lot of data copying between memory spaces that goes on between ActiveX and LabVIEW and, if there is much data involved, it can take a while. Search these forums for the terms safearray or "safe array" or marshal data and you will find more info on what you are up against.

TCP on PDA with GPRS slow

Hi
I finded solution to activate GPRS connection form LabView for PDA (I used Pocket pc) with a dll call.
I have a problem: if I launch a PDA TCP server on PDA and I use a pc like a client (like example on labview) the transmission is very slow i transmit only casual number and client read and plot this number .
Is a problem of GPRS connection ??
Any suggestion?
thank you very much
best regards
elisa

gulm wrote:
Hi
I finded solution to activate GPRS connection form LabView for PDA (I used Pocket pc) with a dll call.
I have a problem: if I launch a PDA TCP server on PDA and I use a
pc like a client (like example on labview) the transmission is very
slow i transmit only casual number and client read and plot this number
Is a problem of GPRS connection ??
What type of GPRS link do you have? The usual speed is
14.4 kB unless you have some newer UMTS systems, so I would expect the
data transmission to be VERY slow in comparison to a normal network
connection.
Rolf Kalbermatter
Rolf Kalbermatter
CIT Engineering Netherlands
a division of Test & Measurement Solutions

Publishing is very slow.

Hi,
Well, everything is in the title. Even when I added nothing, when I republish, it takes like 5 minutes. Before, it took 5 seconds.
I publish on MobileMe, so is it a general problem or is it just me ?
Sincerely.

The iDisk is very slow when it comes to syncing. But for uploading the files to the site's location iWeb uses the WebDAV protocol and is faster. Using the Finder to upload anything is very slow by comparison. Try the following:
1 - go to the System/MobileMe preference pane and log out of MMe.
2 - login with a bogus username and password. This clears the caches.
3 - login with your correct username and password.
4 - launch iWeb and try again.
Do you use the File->Publish Site Changes menu option to publish rather than the Publish button at the bottom?
MMe is frequently very slow. It varies quite a bit in this respect.

Firefox loads pages much slower than Chrome

Recently switched back to using Firefox as I wasn't finding the extensions I like for Chrome only to discover that Firefox is running significantly slower by comparison. Tried uninstalling/reinstalling and it hasn't helped.

A possible cause is security software (firewall,anti-virus) that blocks or restricts Firefox or the plugin-container process without informing you, possibly after detecting changes (update) to the Firefox program.
Remove all rules for Firefox and the plugin-container from the permissions list in the firewall and let your firewall ask again for permission to get full unrestricted access to internet for Firefox and the plugin-container process and the updater process.
See:
*https://support.mozilla.org/kb/Server+not+found
*https://support.mozilla.org/kb/Firewalls
Create a new profile as a test to check if your current profile is causing the problems.
See "Creating a profile":
*https://support.mozilla.org/kb/profile-manager-create-and-remove-firefox-profiles
*http://kb.mozillazine.org/Standard_diagnostic_-_Firefox#Profile_issues
There may be extensions and plugins installed by default in a new profile, so check that in "Tools > Add-ons > Extensions & Plugins" in case there are still problems.
If the new profile works then you can transfer some files from the old profile to that new profile, but be careful not to copy corrupted files.
*http://kb.mozillazine.org/Transferring_data_to_a_new_profile_-_Firefox

Slow Intel iMacs

Hi we have recently installed two new intel iMacs on to our system and they seem to be quite slow in comparison with our existing G5 Imac.
I realise there will be a problem in certain applications until they get native versions, but our slow problems seem to be across the board, and particularly at times in the Finder.
Is there anything we can do to up the performance of these machines a little?
Thanks
Jez
iMac G5 Mac OS X (10.4.7)

If u didn't update them, then remember that updating firmware (whatever that is) must be done manually. Meaning, you have to search your internet browser for the firmware update file(s), and double-click everything yourself, instead of using Mac OS X's "Software Update."
What apps in particular are slower? Check to see if some1 accidentally replaced the native versions with the PPC versions by opening the app, and looking at its type in Activity Monitor, found somewhere in Macintosh HD>Applications>Utility.
Maybe you have the best of the best of iMac G5s upped with maximum RAM and everything else, comparing to the weakest iMac Intel available. Maybe this is a reason?
So, to reiterate, intels' are probably slower b/c of
a) un-native apps
b) not enough RAM
I have 512MB RAM, everything runs great (except when multi-tasking while burning DVDs), but that's comparing to a G4 466Mhz digital audio tower, which is an extremely huge difference.
Hope this helped
peace

Visual spikes and slow running.

Hello, I'm experiencing a problem on my iMac.
The system seems to operate a little slow in comparison to 1 month ago when everything was running smoothly.
My main concern is that the visuals have spikes. Meaning, the animations of everything (system icons popping, running video, applications) are freezing for a bit (less than a second) while enabled or active.
I am a new mac user so i don't really know what i should do.
I'm using a 3.2 GHz Intel Core i5 with 8 GB ddr3 memory, on OS X 10.8.5.
Thanks in advance.

after having repaired laptops for years in the past, a tiny drop that isnt seen gets somewhere........checks out fine.....but the residue later causes a fault.
Its not their fault, I used to have to deal with that quite a bit,.......someone would come back upset that its "still bad", but in fact what happens is the creeping havoc of a spill pops its head up somewhere ELSE which before "checked out just fine"
logical, Ive seen it often myself firsthand........ yet frustrating .
This is why all spills are havoc, not merely due to damage, but it 'creeps'. You can fix it, but find out later a drop "fell" somewhere now its got a NEW problem.

Slow internet on one device

Why is my internet slow on just my Mac Mini(desktop)?
Here is my setup: I've got Time Warner Oceanic Cable internet running to my Airport Extreme, which in turn provides a wifi connection to my Mac Mini (about 6months old), two iPads, two iPhones (when both are in the house), an old Macbook Pro, as well as a wired connection to my Vizio TV and my gaming console. For some reason, my internet connection on my desktop is dreadfully slpw. I've done side by side comparisons where my iPad will be 15mbps download and my desktop will get between .5-1.5mbps. All of my other devices seem to get good speed, although this depends on time of day. Regardless, the desktop is nearly always slow by comparison.
I've tried changing channels on the router, but that doesn't really do anything. The desktop has full wifi signal, as does the iPad, even though they are a solid 40 30 feet away in another room.
Any thoughts?

WiFi signal and link speed can be different..
Post the screenshot showing the link.. hold down the option key while you click the wifi symbol.
Remember wifi is shared.. if other devices are using the bandwidth, the mini cannot use it.. nothing to do with signal or link speed.
Use the wireless diagnostics.. what does that tell you??
Open the utilities and do a wireless survey of the area.
About Wireless Diagnostics - Apple Support
You can then do performance.

Slow unresponsive post 8.1.1 update to iPad 2

please be gentle, first time poster.
i Have loaded iOS 8.1 and my iPad 2 now takes around 10 seconds to load an Internet page via Safari, all applications are extremely slow in comparison to ios7.
now loaded to 8.1.1 and still no change, I feel cheated by apple and pressurised into buying a new iPad to gain the same running speed as previous iOS 7 which I've worked with for two years plus with no complaints at all, so much so I have purchased an iMac 27 inch machine..... Please help and tell me there is a fix without the need to throw away my iPad 2 and needlessly spend hundreds on a new iPad ?

I Am beginning to lose faith in Apple products. They don't seem to care about our problems. I also downloaded the new operating system on my Mac and it also is giving me problems I didn't have previously. The iPad has lots more problems now than before I updated to 8.1.1. Maybe we should consider non Apple products for our next purchase. I am really frustrated by apples lack of Customer service for us. They created the problem. They should fix it!

Slow? polyhedron comparison

Similar Messages

Maybe you are looking for