Disk bottleneck
Hello,
2 weeks ago our my boss told me to check the hardware of our Oracle 8.1.7 database. Cpu and memory looked fairly normal to me.
Now we have 10 disk in a raid10. Split up our oralog (F:), redolog (G:) and database (E:).
Created seperate disk for paging.
But this is what I get.
PhysicalDisk _Total 1 G: 3 F: E:
¾ Disk Write Time 6687,393 879.397 789.015 38350.336
Avg. Disk Bytes/Read 4096.000 0.000 0.000 4096.000
Avq. Disk Bytes/Transfer 6626.102 6148.129 6148.129 4096.000]
Avg. Disk Bytes/Write 6857.944 6148.129 6148.129 0.000
Mij. Disk Queue Ienqth 432.879 8.797 8.523 409.921
Mij. Disk Read Queue Length 31.635 0.003 0.633 26.417
Mij. Disk sec/Transfer 1.580 0.071 0.069 17.822
Disk Reads/sec 22.978 0.000 0.000 22.978
Disk Transfers/sec 273.741 123.883 123.883 22.978
Disk WriteBytes/sec 1719718.749 761648.191 761648.191 0.000I hope someone can give me a clue where to look.
This is not even close to what I'm used to do...
Ps: have statspack logs and pslist logs...where do I need to look for!
Thank you in advance!
Cheers,
P
Jay,
Thank you for your reply. Can I run the sar command in windows?
I tought this was an Unix command.
Here is the statspack output.
Tablespace IO Stats for DB: UNRD Instance: unrd Snaps: 13189 -13190
->ordered by IOs (Reads + Writes) desc
Tablespace
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
UNI_INDEXO
8,948 10 14.3 1.0 17,353 19 13 4.6
UNI_DATAO
8,114 9 12.3 1.5 15,391 17 179 0.3
UNI_ROLLBACK
3 0 193.3 1.0 15,664 17 127 0.0
UNI_TEMP
967 1 3.0 7.9 767 1 2,699 0.1
UNI_DATAC
318 0 5.2 6.6 833 1 29 0.0
TOOLS
131 0 9.5 1.0 671 1 0 0.0
SYSTEM
334 0 7.4 4.0 321 0 2,699 0.1
UNI_INDEXC
82 0 18.3 1.0 278 0 42 0.0
UNI_BO
162 0 6.4 3.5 18 0 0 0.0
File IO Stats for DB: UNRD Instance: unrd Snaps: 13189 -13190
->ordered by Tablespace, File
Tablespace Filename
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
SYSTEM E:\DATABASE\UNRD\DATA\UNRD_SYS_01.DBF
334 0 7.4 4.0 321 0 2,699 0.1
TOOLS E:\DATABASE\UNRD\DATA\TOOLS_01.DBF
131 0 9.5 1.0 671 1 0
UNI_BO E:\DATABASE\UNRD\DATA\UNRD_DATBO_01.DBF
162 0 6.4 3.5 18 0 0
UNI_DATAC E:\DATABASE\UNRD\DATA\UNRD_DATCO_01.DBF
318 0 5.2 6.6 833 1 29 0.0
UNI_DATAO E:\DATABASE\UNRD\DATA\UNRD_DATOP_01.DBF
53 0 16.0 1.0 36 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_02.DBF
73 0 11.4 1.0 72 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_03.DBF
60 0 15.0 1.0 115 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_04.DBF
59 0 15.9 1.0 54 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_05.DBF
205 0 9.0 1.0 188 0 1 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_06.DBF
218 0 11.4 1.0 260 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_07.DBF
170 0 8.1 1.0 380 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_08.DBF
101 0 10.2 1.0 79 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_09.DBF
111 0 11.5 1.0 596 1 20 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_10.DBF
211 0 7.8 1.0 534 1 7 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_11.DBF
152 0 12.6 1.0 92 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_12.DBF
225 0 9.4 1.0 808 1 3 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_13.DBF
363 0 8.8 3.0 468 1 2 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_14.DBF
685 1 5.7 2.2 149 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_15.DBF
648 1 5.6 1.6 420 0 1 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_16.DBF
428 0 8.6 1.1 375 0 19 2.1
E:\DATABASE\UNRD\DATA\UNRD_DATOP_17.DBF
256 0 10.5 1.0 313 0 4 2.5
E:\DATABASE\UNRD\DATA\UNRD_DATOP_18.DBF
322 0 10.0 3.2 340 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_19.DBF
284 0 11.1 3.5 490 1 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_20.DBF
240 0 16.1 4.0 772 1 9 0.0
File IO Stats for DB: UNRD Instance: unrd Snaps: 13189 -13190
->ordered by Tablespace, File
Tablespace Filename
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
UNI_DATAO E:\DATABASE\UNRD\DATA\UNRD_DATOP_21.DBF
202 0 16.6 1.5 603 1 7 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_22.DBF
182 0 11.0 1.0 197 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_23.DBF
265 0 10.8 1.0 255 0 2 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_24.DBF
82 0 17.2 1.0 268 0 2 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_25.DBF
87 0 16.6 1.0 623 1 17 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_26.DBF
90 0 19.4 1.0 341 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_27.DBF
123 0 14.1 1.0 504 1 9 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_28.DBF
94 0 16.8 1.0 327 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_29.DBF
84 0 15.6 1.0 118 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_30.DBF
63 0 18.4 1.0 49 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_31.DBF
87 0 17.7 1.0 274 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_32.DBF
76 0 20.5 1.0 240 0 9 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_33.DBF
79 0 20.9 1.0 186 0 3 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_34.DBF
172 0 16.7 1.0 371 0 1 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_35.DBF
42 0 28.3 1.0 265 0 5 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_36.DBF
30 0 31.0 1.0 15 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_37.DBF
44 0 24.8 1.0 387 0 8 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_38.DBF
121 0 11.2 1.0 833 1 26 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_39.DBF
60 0 32.0 1.0 337 0 17 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_40.DBF
92 0 23.8 1.0 174 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_41.DBF
103 0 19.4 1.0 72 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_42.DBF
116 0 16.6 1.0 164 0 1 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_43.DBF
162 0 15.5 1.0 384 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_44.DBF
172 0 16.1 1.0 780 1 3 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_45.DBF
130 0 18.5 1.0 265 0 1 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_46.DBF
152 0 16.4 1.0 219 0 0
File IO Stats for DB: UNRD Instance: unrd Snaps: 13189 -13190
->ordered by Tablespace, File
Tablespace Filename
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
UNI_DATAO E:\DATABASE\UNRD\DATA\UNRD_DATOP_47.DBF
114 0 17.3 1.0 331 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_48.DBF
94 0 21.9 1.0 47 0 0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_49.DBF
68 0 19.3 1.0 173 0 2 0.0
E:\DATABASE\UNRD\DATA\UNRD_DATOP_50.DBF
64 0 19.7 1.0 48 0 0
UNI_INDEXC E:\DATABASE\UNRD\DATA\UNRD_IDXCO_01.DBF
82 0 18.3 1.0 278 0 42 0.0
UNI_INDEXO E:\DATABASE\UNRD\DATA\UNRD_IDXOP_01.DBF
203 0 17.3 1.0 284 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_02.DBF
196 0 14.0 1.0 353 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_03.DBF
301 0 12.2 1.0 695 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_04.DBF
272 0 14.1 1.0 526 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_05.DBF
219 0 14.2 1.0 500 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_06.DBF
244 0 16.8 1.0 327 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_07.DBF
232 0 12.6 1.0 305 0 1 20.0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_08.DBF
225 0 15.1 1.0 552 1 1 0.0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_09.DBF
166 0 18.7 1.0 292 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_10.DBF
162 0 18.0 1.0 167 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_11.DBF
226 0 12.5 1.0 250 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_12.DBF
204 0 17.3 1.0 339 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_13.DBF
164 0 17.2 1.0 183 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_14.DBF
181 0 14.3 1.0 194 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_15.DBF
478 1 15.0 1.0 2,681 3 2 0.0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_16.DBF
451 0 13.7 1.0 1,180 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_17.DBF
393 0 13.6 1.0 558 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_18.DBF
354 0 13.5 1.0 504 1 2 10.0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_19.DBF
383 0 15.1 1.0 1,006 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_20.DBF
425 0 15.3 1.0 820 1 0
File IO Stats for DB: UNRD Instance: unrd Snaps: 13189 -13190
->ordered by Tablespace, File
Tablespace Filename
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
UNI_INDEXO E:\DATABASE\UNRD\DATA\UNRD_IDXOP_21.DBF
328 0 13.6 1.0 601 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_22.DBF
371 0 12.4 1.0 443 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_23.DBF
546 1 11.6 1.0 694 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_24.DBF
399 0 14.4 1.0 506 1 4 2.5
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_25.DBF
380 0 12.0 1.0 691 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_26.DBF
328 0 11.7 1.0 593 1 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_27.DBF
311 0 15.9 1.0 507 1 2 0.0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_28.DBF
297 0 18.0 1.0 730 1 1 10.0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_29.DBF
280 0 14.7 1.0 351 0 0
E:\DATABASE\UNRD\DATA\UNRD_IDXOP_30.DBF
229 0 14.5 1.0 521 1 0
UNI_ROLLBACK E:\DATABASE\UNRD\DATA\UNRD_RBS_01.DBF
3 0 193.3 1.0 15,664 17 127 0.0
UNI_TEMP E:\DATABASE\UNRD\DATA\UNRD_TMP_01.DBF
967 1 3.0 7.9 767 1 2,699 0.1I do have a couple of permon running as we speak. And I do get some weird number their aswell.
My boss won't believe it might be a hardware problem otherwise the company who wrote the software will point at us. You know how it goes..so hopefullu I can figure something out...
thx again...really!!!
Message was edited by:
PeterC
Similar Messages
-
Oracle 10g on HP-UX, Terrible Poor Performance!!
Hi All,
I setup an Oracle 10g on HP-UX 11iv1. Server is a HP 9000, 4 CPUs (750
MHZ). It is connected to Disk System 2405 (Virtual Array 7110). Fiber
Channels are connected at 2 GB speed.
I installed a cluster 10g database. First I installed CRS and after
that I installed oracle database. ( I want to test clustered database
with one instance)
I installed every thing line by line as oracle document wrote.
All the things, kernel parameters, patches, are like oracle wrote in
its document.
I installed Golden quality package June 2004.
I increased shmmax to 2.3 G . My SGA is 1.7 G And change some other
parameters as Sandy Gruver wrote in Best Practices for Oracle on HPUX.
I used oracle new storage system called ASM for this case.
When I put the system under the load, I was monitoring the system
carefully.
I started gmp. When we sent some quarries to database (It is not heavy
load, I tested it with a Linux system on proliant ML570 without any
problem), suddenly DISK section in gpm changed to red (critical ) I
read the warning. It said "Disk bottleneck probability = 100%". I
changed the output of disk report to "Report IO by Disk"
"DISK%" was 100% and "RAW IO RT" was about 1000 for two disks ( This
two disks dedicated for ASM). In this situation CPU idle time was 1% or
2% for all the CPUs but load average was about 1. Performance is not
acceptable at all ( In comparison with Oracle that installed on Linux).
Glance reported Disk was in Critical situation.
I think the problem is IO or something about Disks
I used HP Disk System 2405. Fibber channels on both server side and
Disk Array side are configured at 2 Gb and topologies are
PTTOPT_FABRIC.
Is it ok that RAW IO RT about 1000 for each LUN?
Why Disk% in glance/report IO BY Disk/ was 100%?
I found an error in STM logs about I/O.It said:
Entry type: I/O error
Product: Fiber Channel Interface
Logger: td
It logged this error about 12 times during the test.Any comment?
Regards,
HasanSorry, I have not a solution for your problem, but similar things happen on our installation on Solaris 5.8 with Oracle 10:
I have a banking business solution from Windows/SQL Server 2000 to Sun Solaris/ORACLE 10g migrated. In the test environment everything was working fine. On the production system we have very poor DB performance. About 100 times slower than SQL Server 2000!
Environment at Customer Server Side:
Hardware: SUN Fire 4 CPU's, OS: Solaris 5.8, DB Oracle 8 and 10
Data Storage: Em2
DB access thru OCCI [Environment:OBJECT, Connection Pool, Create Connection]
Depending from older applications it's necessary to run ORACLE 8 as well on the same Server. Since we have running the new solution, which is using ORACLE 10, the listener for ORACLE 8 is frequently gone (or by someone killed?). The performance of the whole ORACLE 10 Environment is very poor. As a result of my analyse I figured out that the process to create a connection to the connection pool takes up to 14 seconds. Now I am wondering if it a problem to run different ORACLE versions on the same Server? The Customer has installed/created the new ORACLE 10 DB with the same user account (oracle) as the older version. To run the new solution we have to change the ORACLE environment settings manually. All hints/suggestions to solve this problem are welcome. Thanks in advance.
Anton -
What is better: Export as *.MOV or Export as *.MP4 ?
I have a couple of small DigiCam *.MP4 videos. Some of them I have to rotate 90 deg clockwise, to edit, to re-encode and to save as a new video.
When coming to the "export" step I have the choice between the original *.MP4 (=MPEG-4) format and *.MOV (=Apples Quicktime).
Which one should I take?
I guess both are not lossless. So which one have the minimum losses in quality when re-encoding and saving?
Are there other considerations?
PeterAnytime you work with a potentially destructive process you need to test it and be very confident in it. Do test option II quite well -- I have seen full data exports not actually export all the data. So you need to be very confident it is working as expected. As Glen pointed out some cubes have data loaded high (which is not a best practice as if you load data high it should be allocated back down such that an aggregation will not clobber it).
In regards to Cameron's findings about a restructure working just as fast as a level 0 exp / load / agg this is a case of a disk bottleneck. A restructure process in all versions older than 11.1.2.2 were always single threaded and if you have a well tuned aggregation which does make use of multiple cores then if your disk subsystem isn't a huge bottleneck the
level 0 exp / load / agg will ALWAYS be faster. This being said it's not uncommmon to see poorly architected hardware which Essbase is being ran on which means Cameron's finding may be more the typical environment.
Regards,
John A. Booth
http://www.metavero.com -
Hi
I have a Hyper-V 2012 R2 server with 192gb RAM running @ 1800mhz, 2 x Xeon E2650 v2, and for VMs, a SATA 7200RPM disk (4tb, no RAID).
I'm running 70 or so VMs, all on, and I'm the only user of the system so no load (of course for things like SQL Server, it still runs timer jobs which access databases, etc). VMs are a mix of SQL Servers, ADFS, AD DS, cert authorities, Sharepoint, System
Center, and lots more.
On all VMs, they are very choppy and jerky. Would this be a disk bottleneck? Memory CPU look good. If it is disk, what perf counters can I use to establish this?
I'm looking to get 2 X 4TB SAS disks (also 7200rpm) in RAID 0 or 1. Would this perform much better?
ThanksHi GSS1
"I'm running 70 or so VMs"
"I'm looking to get 2 X 4TB SAS disks (also 7200rpm) in RAID 0 or 1. Would this perform much better?"
Obviously one sata disk can not supply enough disk I/O for 70 VMs even though with " 2XSAS ".
We can not provide a accurate disk I/O requirement for your environment but I'm sure that the performance of 2XSAS raid 0 must be better than one SATA disk .
(I assume that is a lab evironment , raid 0 do not provide redundancy )
In addition , please refer to the link regarding to measure the disk performance on hyper-v host (it has been posted by
Marc):
http://msdn.microsoft.com/en-us/library/cc768535(v=bts.10).aspx
Best Regards
Elton Ji
We
are trying to better understand customer views on social support experience, so your participation in this
interview project would be greatly appreciated if you have time.
Thanks for helping make community forums a great place. -
PE7: Split screen/picture in picture jerkiness problem?
I'm using Premiere Elements 7, creating a video that displays multiple videos at once. I tried:
Splitting the screen into four boxes of same size (top left, top right, bottom left, bottom right), each filled with its own video
Having one main video with 3 picture-in-picture windows overlaid on top of the main
I followed this instruction: http://www.heron-productions.com/4-splitscreen.html
In both cases, after I rendered work area, I see the video on Track 1 and sometimes Track 2 become very jerky, like skipping backwards and forwards by themselves. Exactly the same issue also when I export into final video. The only way I could solve this issue reliably is by decreasing the number of simultaneous videos/windows to two.
I have also tried this on another machine that has Premiere Pro CS4, resulting exactly the same issue. Very confusing since at least Premiere Pro should be able to do this just fine.
Is this a bug? Or am I doing something wrong? FWIW, all my sources are AVCHD files from a Canon HF100, and the project is a 1080 30i.Thanks for the reply! Yeah, I figured I should've posted at the root forum, instead of in the Tips & Tricks. I'll probably try to repost this there. Anyway...
I don't think this is a hardware issue. It's not about trying to preview in realtime (although the bug shows up there too, after rendering work area). I can wait as long as I need if that's what it takes. But this problem is in the final render result. Yes, I do have an i7 setup for this: Core i7 920, 6GB DDR3 1600, ATI 4830 512MB, and disk load spread out among: 1x 1TB 32mb cache, 1x 750GB 32mb cache, 3x 500GB 16mb cache. They're not in RAID setup, but again, I can live with the current rendering time, which is more CPU-bottlenecked than disk-bottlenecked.
This bug also exists in Premiere Pro CS4 when I tried it on another machine with that app (I myself don't have CS4).
I rendered a short, lower resolution video to show how this bug looks like.
http://cid-48ae90fe02b84e66.skydrive.live.com/self.aspx/Videos/PremiereBug.wmv
Notice the main video in the background. It becomes jerky from time to time when there are more than one PiPs. Take a look around 0:20 where it becomes even worse after the 3rd PiP shows up. -
2008 R2 RDS Allocates Little Memory per Session - Apps Slow
I have an RDS server physical on Dell PowerEdge T410, 16 cores, 32Gb RAM.
With 8-10 RDP sessions connected only 5Gb of physical RAM is used out of the 32Gb available. The apps (Dynamics ERP) are performing sluggishly - at least 50% slower than in native Windows client server on an average desktop. The desktops utilize
2-4Gb each, so I am having hard time understanding how 8 concurrent sessions only uses 5Gb. These session should be faster than on desktops. LAN/WAN bandwidth is not in the picture, because the same slow speed happens on the console.
Note: in addition to RDS, this server also has IIS Role.
Maybe the server is not tuned for RDS?Hi,
Thanks.
1. Do you have any antivirus/antimalware/security/etc. on this server? Have you tried completely removing it and then testing?
2. Have you applied the latest Dell SUU bios/firmware/drivers/etc. for your server? Even if you just received the server often it will not have the latest SUU updates installed.
3. Do you notice slowness when you are the only user logged on (at the console), with no programs running? For example, you are the only logged on, you should be able to open Control Panel, Notepad, Task Manager, other basic stuff, and they should
come up almost instantly.
4. Have you checked to see where potential bottlenecks are by using Performance Monitor? For example, you may have a disk bottleneck, or perhaps there is some rogue process that is maxing out the cpus.
5. Related to #4, what is the disk configuration of this server? Multiple 15K RPM drives in RAID 10, or ?
-TP -
My video in the timeline freezes. It continues to play the audio, and will pick back up... but why is my video freezing in certain parts of the timeline.
If I export to avi. it plays just fine.
I just upgraded my processor, because I thought my dual core 2 processor 2.6 wasn't enough...but that didn't solve the problem.
Computer:
Intel Quad Core 2
2.83 ghz 3.50 gb
NVIDIA GeForce 880GT
Asus P5N-E SLI motherboardThat is a very good plan. It could well be that the freezing could be caused by the hard disk bottleneck. Your new CPU may no longer be the major bottleneck, even with AVCHD material. Did you have a look at these guides:
How to get the best from a PC? Some guides...
Maybe it will give you some ideas. -
Should Hyperthreading Be Enabled? - i7 920
I set up a new computer running CS4 and I'm wondering if hyperthreading is going to offer any advantage? I thought the reason that intel stopped using hyperthreading on the quad-cores was because they already had 4 physical cores. I've got an i7 920 OCed to 3.7 and I was told that disabling hyperthreading would help keep the temps down.
Mostly this question pertains to Premiere, Encore and Media Encoder for my purposes.
thanks for any input.A computers performance is a balance between the most important ingredients: CPU, memory, disk and sometimes video.
If you see relatively low CPU usage, that means that the other components are busy. It may be a memory bottleneck, it may be a disk bottleneck and very, very occasionally a video bottleneck.
Often this happens when significant amounts of data have to be processed by memory and then stored on disk. Often it is simply disk thruput that is bottlenecking the system.
When I encode a large timeline on my mediocre system, I get results like this:
This means that my system in reasonable balance, but memory or disks are holding back performance, otherwise the CPU load would have been more than the 70-80% I saw. Quite simply the CPU has to wait for the other components. Now, in this case this happened with a lot of resident programs active like Perfect Disk, SnagIt, SEP, and Mozilla with around 10 tabs open , so that may account for the lower CPU load.
This was a pretty long SD timeline (05:21:48:23) encoded to MPEG2-DVD with quality set to 5 and AC3 sound, which encoded in around 20 minutes. This can of course cause some disk trhoughput problems and memory congestion.
If you want to know how your system is configured and tuned, run the http://ppbm4.com benchmark and send the results to Bill (and possibly send them to me by PM) and you can see how your system performs in relation to other systems. -
So I've been reading a few topics but it never actually addresses what I'm talking about (except for one topic but hardware differs so dunno).
When I open a few apps, GNOME 3 starts getting slow and unresponsive. I can't have C::B (or QtCreator), Chromium and Clementine or it starts getting REALLY slow. I like Gnome 3 and I really, really don't want to switch. There's about 20% of my CPU devoted to Gnome, I don't think that's normal.
Any ideas why it should be consuming so much? Where should I look? What can I configure or trim down to reduce consumption?
I have an Intel Core i3-350M processor (2.2GHz), 4 GB RAM, and an Intel HD Graphics card (that might be a problem from what I'm reading). It's a HP dv6-3225dx laptop.
Thanks a lot for any help in advance.bbarcher wrote:
Pawbla wrote:
PowerTop has given me problems in the past, but I'll give it a shot when I can switch DE and it starts getting slow.
I have... many partitions. First two are Windows, sda3 is /boot, sda5 is swap, sda6 is / and sda7 is /home.
parted -l:
You could use top but powertop gives much better insight. With powertop running you'll be halfway there. Without info from powertop I doubt there will be an effective way to help you. As for the gnome monitor. For the sake of clarity IMO it would be best to shut it down for every case you check and every readouts you make drom top/powertop.
You could also do a quick boot into live media GNOME using e.g. Manjaro.
I really don't think your issue is caused by disk bottleneck. Just wanted to check if you haven't squeezed your linux into a 10GB partition at the end of the disk
Yeah I normally use top. However the system usually becomes unresponsive during this issue.
After posting I used powertop to check a few things that were taking just a bit more resources than I'd like and I adjusted them to avoid this. I haven't had this issue happen again since (except if you count while running my own coding mistakes), so I'll be blaming it on Deluge I guess. I thought the thing was light! It was hoarding quite a bit of resources. That's why it doesn't happen all the time I guess, I just never thought Deluge would be to blame. I haven't had gnome-shell taking 20% again - if it ever starts happening again I'll post about it. Dunno if that could be caused by deluge too or if it has to be something from gnome-shell.
Sorry for wasting everyone's time!
Last edited by Pawbla (2014-03-06 17:38:12) -
hello
Im Planing for Production Installation ( IS-retail ) with i5/OS(V5r4m0) .
i am having FC5786 Storage Device. i am Planning to keep Journals on ASP2.
we have following droughts
1. RAID 1 and hot swamping( standby )is Possible with above device?
2. Is there sizing proportion for ASP1 & ASP2.
3. How i can monitor Individuals hand disk bottleneck's ?
regards
sachinHi Sachin,
1. RAID 1 and hot swamping( standby )is Possible with above device?
=> don't know this term ...
2. Is there sizing proportion for ASP1 & ASP2.
=> more or less everything into ASP1 and 1 or 2 mirrored disks (=> 2 or 4 disks) into ASP 2 - As the disks will have a size of 70 or 140GB this would be sufficient space in ASP 2
3. How i can monitor Individuals hand disk bottleneck's ?
=> WRKDSKSTS
Regards
Volker Gueldenpfennig, consolut international ag
http://www.consolut.de - http://www.4soi.de - http://www.easymarketplace.de -
Will RAC's performance bottleneck be the shared disk storage ?
Hi All
I'm studying RAC and I'm concerned about RAC's I/O performance bottleneck.
If I have 10 nodes and they use the same storage disk to hold database, then
they will do I/Os to the disk simultaneously.
Maybe we got more latency ...
Will that be a performance problem?
How does RAC solve this kind of problem?
Thanks.J.Laurence wrote:
I see FC can solve the problem with bandwidth(throughput),There are a couple of layers in the I/O subsystem for RAC.
There is CacheFusion as already mentioned. Why read a data block from disk when another node has it in is buffer cache and can provide that instead (over the Interconnect communication layer).
Then there is the actual pipes between the server nodes and the storage system. Fibre is slow and not what the latest RAC architecture (such as Exadata) uses.
Traditionally, you pop a HBA card into the server that provides you with 2 fibre channel pipes to the storage switch. These usually run at 2Gb/s and the I/O driver can load balance and fail over. So it in theory can scale to 4Gb/s and provide redundancy should one one fail.
Exadata and more "+modern+" RAC systems use HCA cards running Infiniband (IB). This provides scalability of up to 40Gb/s. Also dual port, which means that you have 2 cables running into the storage switch.
IB supports a protocol called RDMA (Remote Direct Memory Access). This essentially allow memory to be "+shared+" across the IB fabric layer - and is used to read data blocks from the storage array's buffer cache into the local Oracle RAC instance's buffer cache.
Port to port latency for a properly configured IB layer running QDR (4 speed) can be lower than 70ns.
And this does not stop there. You can of course add a huge memory cache in the storage array (which is essentially a server with a bunch of disks). Current x86-64 motherboard technology supports up to 512GB RAM.
Exadata takes it even further as special ASM software on the storage node reconstructs data blocks on the fly to supply the RAC instance with only relevant data. This reduces the data volume to push from the storage node to the database node.
So fibre channels in this sense is a bit dated. As is GigE.
But what about the hard drive's reading & writing I/O? Not a problem as the storage array deals with that. A RAC instance that writes a data block, writes it into storage buffer cache.. where the storage array s/w manages that cache and will do the physical write to disk.
Of course, it will stripe heavily and will have 24+ disk controllers available to write that data block.. so do not think of I/O latency ito of the actual speed of a single disk. -
Is my OSX app waiting on (bottlenecked by) disk?
I have an older Macbook Pro and my Lightroom application is slow to come up. I think it may be waiting on the USB attached disk, but I'd like to observe the aggregate and real-time waits it's doing on disk reads/writes (such as a time-based graph of the size of the disk op wait queues). I've tried the following:
Activity Monitor - only shows aggregate bytes read/written
iosnoop and fs_usage - only shows traces of individual disk ops
Disk Utility - only shows space usage
Disk Inventory X - only show space usage
The Activity Monitor does show the real-time CPU usage, and that's only 20% for my Lightroom application, so I don't think it's limited by the CPU.
I have to believe that such a monitoring tool exists, but Google search can not seem to find it using the keywords I've tried. Any insights would be greatly appreciated.
- Thanks in advance, JimAn external disk can be a bottleneck if you install the unneeded software they all seem to have when new. You don't need anything extra to use an external drive on a Mac and, if you use the drive maker's software package, you are likely to see slowdowns.
Also, USB2 is functionally slower than FireWire 400, the slowest FireWire implementation. USB tends to write in bursts, not streams, and that slows tit when you do a real-world test,
Another factor can be the drive's formatting as shipped. I have a WD external drive that never behaved right due to a proprietary formatting scheme, even though the drive was sold as a "Mac Edition." I completely erased the drive and reformatted it for Mac using Disk Utility. It immediately went from being sluggish to speedy.
The reformatting only works us you have no reason to share the drive with a Windows computer, -
2.5 GB Ram, no application - Disk usage bottleneck - extrem slow response
Hi everybody,
I'm experiencing some trouble. Last weekend, i updated my itunes library with bpm count for DJing via BPMer. The Librarys size is about 100 GB. Usually i have 50 - 75 GB free space depending on the size of some Logic Pro projects.
While updating, Time Machine went crazy with tons of gigs of updates but after some time my internal (not time machine) drive was getting full.
So i started to delete some old files but the left space didnt grow up.
After some time i recognized i had over 100 GB free space, why?
From this time on my disk is going crazy by doing whatever (thats my big question), my Ram is used with up to 2.5 gig without displaying any application using this much space and the free disk space is changing in minutes between 135 and 100 GB.
I tried some maintainace skipts, PRAM reset, Permissions repair, disk repair, virus/ trojan scan without finding the reason.
The real problem is i can see in the istat pro monitoring, that the disk is going crazy, but du in terminal doesnt give me hint on what application it is. The same thing is true for the ram which is used, but no application is listed with more than 200MB.
Thanks for your help!Defragmenting using iDefrag solved the problem…
The mac only maintains files smaller than 1 GB, iDefrag did the rest… -
We run a small network in our office consisting mainly of Intel iMacs we purchased several months ago. The server computer is a brand new Intel Mac Pro server with 4 500 gig drives raided together, two 2.66GHz dual core intel processors, all the other bells and whistles we could think of, etc. We added in two fiber-optic switches: one in the server room, one in the office. The switches connect to each other and the server with fiber optic cables, but the iMacs connect with ethernet cables.
We had the iMacs for a while but we just recently got the new server and upgraded our old 100 base switches. Afterwards we wanted to test out the data transfer speeds, as we plan to back up to the server frequently. We were dismayed to find that transfer rates capped out at 60 MegaBytes/sec according to the Activity Monitor's Network Activity tab. In fact, it would range mostly from 40-45MB/sec. None of us here have much experience with networking, but that seemed a tad too slow. My basic math tells me that a byte is 8 bits, and from that a Gigabit network should transfer data at 120 MegaBytes/sec, which is three times the speeds we were actually seeing.
We sent data both to and from the server in order to test this. Thinking it was perhaps a problem with the ethernet itself, we grabbed an external FireWire hard drive and transferred data from one of the iMacs directly to it and noticed exactly the same transfer rates. We plugged two iMacs directly into each other and transferred at the same rate, ~45MB/sec.
Well this was highly frustrating. All Macs supposedly ship with Gigabit ethernet since, what, 2002? Earlier? Why are our speeds so slow? We thought the hard drives might be slow, so we got info on the drives and googled them for their tech specs. The iMacs' Western Digital drives are capable of much much faster speeds according to everything we've read.
We started reading anything we could online that addressed this issue. Some information read the problem might be that the optical DVD-R/CD-R drive was only capable of slower transfer speeds. Since the optical drive and the hard drive are on the same bus it would slow down the hard drive's maximum transfer rate, much like having a Gigabit hub with a 10 base computer plugged into it would slow the entire network down (which is of course why we use switches instead of hubs). Is there any truth to this? If this is, in fact, the case, can we bypass this bottleneck somehow? I'm not talking about opening the computer and manually disabling the optical drive, as that's a waste of a perfectly good DVD-R/CD-R drive.
Also, if this is true, why in blue blazes is Apple flaunting Gigabit ethernet if the computer can only take advantage of 1/3 the speeds Gigabit ethernet has to offer?! I'll happily provide any more information that's relevant to the problem at hand.
Thank you
Mac OS X (10.4.8)Here's the specs for the default Hard Drive that Apple put in here:
http://www.westerndigital.com/en/products/Products.asp?DriveID=137
From that page
Buffer To Host (Serial ATA) 300 MB/s (Max)
Buffer To Disk 748 Mbits/s (Max)
I have no idea what signaling overhead for my data transfer protocol means.
The ethernet cords are no longer than 20 feet, are all Cat5E, and are well shielded. There is little to no ElectroMagnetic interference in any area they run. The longest cable is the Fiber Optic one, which runs about... oh... 40 yards? These were all installed by professionals.
I feel it's important to stress the fact that we tested an isolated direct computer to computer transfer with a single 6 foot Cat5E Ethernet cable and still experienced the same speeds that we experienced over the network. We tested multiple computers, multiple cables, multiple files. -
Fastest possible disk access in os9
hi there
as the topic says im interested in achieving the best possible disk speeds in os9 for use with old media programs.
i have resurrected an old g4 450mhz sawtooth that im using to run some old music production programs
im pretty sure that the disk speed is the biggest bottleneck on this machine.
Ive been looking into the following ACARD devices ive provided links to newegg
simple for the images + reference
ACARD AEC-6280M (4 CH IDE | NON RAID)
http://www.newegg.ca/Product/Product.aspx?Item=N82E16816123110R
ACARD AEC-6290M (2 CH SATA150 | NONRAID)
http://www.newegg.ca/Product/Product.aspx?Item=N82E16816123109R
ACARD AEC-6880M(4CH IDE RAID0/1)
http://www.newegg.ca/Product/Product.aspx?Item=N82E16816123007R
ACARD AEC-6885M (4CH IDE RAID0/1/0+1)
http://www.newegg.ca/Product/Product.aspx?Item=N82E16816123008R
ACARD AEC-6890M (2CH SATA150 RAID0/1)
http://www.newegg.ca/Product/Product.aspx?Item=N82E16816123111R
features booting into os9
my question is can anyone reccomend to me any other cards that i might find cheaply + available
that provide the following criteria:
a) bootable into os9
b) supporting SATA hard drive (so i can buy new hard drives in the future IDE are no longer available for purchase at my local shop)
c) hardware raid
d) supporting HFS volume visible to both os9 + osx
e) hardware raid dip-switchable from RAID0/RAID1 (RAID0+1 would be the ultimate)
f) PCI interface and compatible with all G3/G4 and also pci/pci-x based G5's
the 6890M seems like the perfect option for me but i have found that these items are now unavailable + sold out.
is there any other product that i can try to find under a diferent brand name ? i have read that acard did cards for sonnet/miglia/SIIG
that were rebranded... Im hoping someones already done their homework on all this stuff and can point me in the rigth direction
so i can get to using my programs + os to do what i love to do and stop worrying about technical headaches..
again obv this isnt my only computer but i like to use os9 for some programs + also its unique GUI environment!
fingers crossed@brody
i actually own one i have it on the desk sitting next to me right now. its a silicon images controller with custom firmware to provide the mac compatibility... works great... but only provides 2x sata150 ports.. which is great in and of itself.. but doesnt provide raid to os9.. doesnt alllow you to combine the speed of two drives together undre os9... to achieve a speed comparible of sata II (sata2 or sata300)
@jon
yes i imagined that card would be the best bang for buck but again that card provides the ports for ide drives and doesnt provide the combination of more then one drive unless u were to use it under osx.. and then u could acheive this with softare raid from disk utility but wouldnt allow me to use two drives as one volume under os9
ive actually been in touch with acard and they said they are out of stock completely on the AEC-6890M which is the card that provides RAID with 2x Sata drives for os9. im looking for a similar product to this.. that has the raid controlled by the card and not the OS.. configurable between RAID0, RAID1 and possibly even RAID0+1(which would require 4 drives tho, instead of just 2)
ive just got a response back from a calssified ad i posted to but had almost given up on after not receiving a response. the guy was on vacation;) anyways hes got a AEC-6880M which is the raid capable IDE Card.. for a good price.. the only problem is.. the availability of IDE hard drives.. they are all dying and i dont think they are making IDE ATA133 drives anymore.. ? does anyone know?
Maybe you are looking for
-
How to match a series of midi events to the tempo?
hello there, i've got a series of midi events that i'd like to adjust to the tempo -there are only quarter notes. i thought to use flex time but it seems to be working with audio tracks only. any idea? as well, i'd like to fix all notes to a same len
-
Converting bytes to unsigned bytes
int u=0; int i=32; byte [] data1 = new byte[1024]; int[] unsignedValue=new int[1024]; while(u<100){ unsignedValue[u] =data1[i] & 0xff; u++; i++; } from in
-
How to make J2EE engine auto load The JSP Pages
After I modify some JSP pages, If I just copy them to the folder in the Server, The pages didn't update. How to make it auto update?
-
LabView TabControl Refnum bug?..
LabView 6.0.2 is getting crashed if to create new control (not a VI!), then to create Control Refnum and to drag&drop TabControl into it. Immediate memory fault follows together with LV crash. However if to assign VI Server Class explicitely by right
-
I am a beginner with using java with databases, so help would be appreciated. I am tring to convert one of my ResultSet columns into an String[] array, but I keep on recieving the following sql error: "Feature not implemented" I am using a MySql data