Planning of the CPU
Hello
I recently inherited a Setup.
What we discovered was a guest with 8 cores (dual, quad-core processors). It has eight virtual machines running.
Each of them has been configured with 4 vCPUs. Performance was pretty terrible, as you would expect scheduling issues.
I've reconfigured seven with one vCPU and two vCPU. Performance is significantly improved.
The thing about this issue is it was painfully obvious by looking at the configuration. Looking at the indicators of performance, however, what we saw was the counter which showed the use CPU was down around 20% on the host.
So my question is how do troubleshoot you less obvious examples - where would you diagnose performance due to CPU scheduling problems?
You would with ESXtop or performance counters.
in ESXTOP, you scale the CSTP meter to see if coscheduling is causing a problem. Also watch % RDY to see if you are invading your CPU.
Tags: VMware
Similar Questions
-
Yet another question on the planning of the CPU
Hello!
CPU Scheduling is one of those endless, and although I read a lot about it, I found different answers to the same question. Moreover, this is the last document on this subject and recommended... read http://www.VMware.com/files/PDF/Techpaper/VMware-vSphere-CPU-Sched-perf.PDF .
Is it possible to run 2 virtual machines at the same time in the same logical processor? Very basic question, I'll appreciate any response and if it has links to the official documentation, even better!
Kind regards
elgreco81
Ah, I think I see the confusion. There is no such thing as using a fraction of a nucleus within a single cycle. You CAN use all of a nucleus for a fraction of the total cycles over a period of time, but a physical core is busy or idle. There is no in-between, especially since we are ignoring hyperthreading in this example.
All of a core to aid for a fraction of the total time in a given sampling period, is what gives the illusion of the use of 'part' of a core. Split in fact one heart in this way would require States and the architecture that simply do not exist in the x 86 world.
Note that in the article by Frank Denneman he calls specifically free cycles "shopping on ' Planner. In your example, there will be always 1 free cycle at any time. There are 1000 cycles per second in your original example, and many of these cycles may be free, but there is never more than one at a time.
To say things, otherwise, you cannot use all 1000 MHz in the first millisecond.
The paragraph you cited in the guide to management resources assumes a Jepp 2 GHz. keep in mind that any measure of frequency REQUIRES a time interval. There is no literal split 'given time' is implied here. What they're saying is that if over a certain period of time, say a second, VM1 only consumes 500 cycles and ask no more, then VM2 is welcome to help itself to one of the unused 500 cycles initially entitled to VM1, and cycles of 1000 that VM2 was already entitled.
I would like to know if there is still no ambiguity, or whatever it is that I does not really there.
-
Up-to-date documentation on the use of the CPU and memory
Hi all
I'd like to get the latest information on the operation of management and CPU memory in esx 3.5, I have not looked at the product for a few years but read recall some excellent articles
Someone can point me in the direction of some good technical articles on memory like the blimp and transparent page sharing in addition to the planning of the CPU
Thank you very much
Understand the memory resource management in VMware ESX Server
StarWind Software R & D
-
Satellite L850 - error of cooling and slow down the speed of the CPU even when fan to 80 c
So, I have this problem that my fan stops turning when the system is under load and then turns on, starts running at the speed of 100%, an error which says error with cooling system, laptop computer back to the service.
The fan slows even when the laptop cools completely, it only returns to normal speeds once I closed the laptop.
In the last few days, which has stopped completely. Now the fan won't even middle speeds even though the CPU temperature is 80 C. Due to the bad news of cooling time score about 70 ° C, even at idle.
The laptop is almost a year. I already had this problem after a week of buying it, and they replaced my processor. I have windows 7 x 64, amd 7670 m and 2350 m i3.
I wonder if this may be a BIOS problem or I have to take for the service? How can I stop this from happening? And Yes, my power plan is on high performance.
AFAIK the cooling fan is controlled automatically if you don t have a any influence on it. Temperature sensors are placed on the motherboard and they control the cooling fan activity. That one possibility to change fan cooling activity is to change the option in the option of cooling method (advanced power settings > Toshiba saving settings).
In my opinion the best solution for you is to contact the nearest Toshiba service provider and ask for help.
They have diagnostic tools that you can check all this. -
How to check the CPU usage and paging using LabVIEW
Hi guys,.
I build an application that is used to check the CPU usage and paging using LabVIEW. How can I do?
any help, suggestions or advice will be greatly appreciated...
Kind regards
Prashant
Hello
If you plan to build your app for Windows, you can use .NET classes. (System.PerformanceCounter), there is a simple example with LabVIEW:
C:\Program NIUninstaller Instruments\LabVIEW 2010\examples\comm\dotnet\SimpleTaskMonitor.llb
Also, you have several screws that you can use to verify information about the processor.
Kind regards
-
Windows XP, SP3, it worked fine. Suddenly Windows/Microsoft Update crashes and uses all the CPU while trying to assess what needs to be updated. The only way to stop him is to kill the process.
A ran Fixit from Microsoft (support.microsoft.com/kb949104), who said that he repaired the window update, but could not fix "problems installing the recent updates. Nothing else was not fixed.
Problems found
Recent updates installation problems
Not fixed
Fix Windows Update
Successful
If I look at the properties of the de...\System32\wuaueng.dll I have the most recent version for XP (7.6.7600.256).
I know that the support for XP will end this year, but I want to Microsoft Update to work down there.
Any help appreciated. Thank you
You are welcome.
How about that. Don't you love it when a plan comes together? Windows updates only or material for pilots also and then update windows is good, but for a more broad ability of other applications Microsoft, then Microsoft Update and Office update is good for a shopping for updates of every Microsoft product you have.
You suppose wrongly, and I thought I answered this before. Technically, you can re-enable the automatic updates but that will require you disabling updates every Monday just before black Tuesday and then manually download and install the cumulative security update newly released and restart before you activate the updates automatic (or notifications only) once more. It is more reliable just disable automatic updates and do not have the headache. Disable the horse of notification that is displayed in the status bar, and that you don't mind.
Please mark this thread as a response. Thank you.
To not argue the point, but XP users required the kind of brilliant 'support' who gave XP since the time of the IE7 to IE8 'false upgrade' users XP need like a hole in the head.
-
I just got my Macbook Pro w / retina delivered today 15 ".
I installed CC After Effects and received this warning message:
Ray-Tracing on GPUS requires a graphics NVIDIA and CUDA 5.0 or later approved.
For now, the ray-tracing will use the CPU.
After installing CUDA 5.0 or later, restart your computer.
For more information, see: http://www.Adobe.com/go/aeraytracedgpu
My mac spec is:
Processor: 2.8 GHz Intel Core i7
Memory: 16 GB 1600 MHz DDR3
Graphics card: AMD Radeon R9 M370X 2048 MB
-
I have absolutely no idea what the implications are for the AMD graphics card instead of the NVIDIA one card, no one knows what effect this will have on my mac when you use CC After Effects? I've never had this warning on my iMac 21.5 "...
Is there something I can do?
Thank you very much, Harry.
The only thing THAT AE really use the GPU is obsolete ray-traced rendering engine. Very few people use the ray-traced rendering engine. I wouldn't worry about this at all.
More info here: features GPU (CUDA, OpenGL) in After Effects and here: http://motionworks.net/after-effects-3d-future/
Now, some third-party plugins (like element, 3D Invigorator, transmutation, etc..) Make use of the GPU, but they will work with the AMD cards and have nothing to do with this warning of CUDA, you want to come. Check that the requirements for all you plan to buy.
You can confidently check the option do not show this message.
-
Is there any fix planned for the slow lingering and pervasive in LR6/cc?
LR6, Win7, 6 GB of ram
Is there any fix planned for the slow lingering and pervasive in LR6/cc? I disabled the GRU. I cleared the cache. My catalogs are updated to LR 5.7.1. Everything is slow. It doesn't matter what it is. Its takes twice as much time to do something. LR will freeze for 5 to 15 seconds at a time without apparent reason. The use cases are to go "Not responding" are numerous and vary widely. I check the CPU during these events Nn sensitive usage and its pretty much slowed down. I have check the disk usage and its also generally inactive.
I thought it might be VSCO plugins, but it's not. LR5.7.1 is always screaming fast while LR6 is an old slow cow.
Here is an example that I saw that is repeatable:
(1) start an import
(2) choose the photos from an SD card
3) click the + to add them to the catalog
4) appears to enter a catalog name
(5) start to type in the dialog box.
(6) after may entered 8 letters, LR goes "Not responding", 10 seconds(7) after LR starts responding again, emphasis on catalog name field has been lost
There is no reason for a freezing here. The CPU does nothing. The drive does nothing.
There are things like that in every sense.
Another example.
(1) develop the Mode
(2) select a photo
(3) left and right panels are hidden
(4) browse the side panel right of the mouse so that it will pop-out
5) LR freezes for 5 seconds
6( ) when the right panel opens, click on the number of exposure (0) and try to type in 1.25.
7( ) after after graduating 1,2, LR hangs for 10 seconds.
8( ) when LR answers once again, finishing by typing in the 5 other (1.25) and press to enter.
9( ) the image immediately illuminates as expected.
Yay! They have updated LRCC/6 today. The slowness and latency seem to have disappeared! Yay! I can get back to work!
-
explain plan for the same query diff
Hi experts,
Please, help me understand explain the plan. I have tow Server (server and two server). The server are same table, even the type of database, even version Oracle (gr 11 (2), same operating system (linux Redhat 5.5) and same table and index.
but when I explain the plan for the same query on the two server. I got diff--diff to explain the plan. reason it has different, according to my understanding, it should be same. explain please, I share the explain plan and lower indices for the two server.
Server a
SQL > col COLUMN_NAME format a20
SQL > select index_name, column_name, position_colonne from user_ind_columns where table_name = 'LOAN_RUNNING_DETAILS_SOUTH"of order 1.
INDEX_NAME COLUMN_NAME POSITION_COLONNE
------------------------------ -------------------- ---------------
DATE_IND1_S LOANDATETIME 1
IND_MSI_LN_LNS1_S MSISDN 1
IND_MSI_LN_LNS1_S LOANDATETIME 2
IND_MSI_LN_LNS1_S LOANSTATUS 3
LAST_INDEX L_INDX_MSISDN_S 1
MSISDN L_INDX_MSISDN_S 2
SQL > select decode (status, 'N/a', 'Part Hdr', 'Global') ind_type, index_name, NULL nom_partition, status
2 from user_indexes where table_name = 'LOAN_RUNNING_DETAILS_SOUTH '.
3 union
4. Select 'Local' ind_type, index_name, nom_partition, status
5 to user_ind_partitions where index-name in (select index_name in user_indexes where table_name = 'LOAN_RUNNING_DETAILS_SOUTH')
6 order of 1,2,3;
IND_TYPE INDEX_NAME NOM_PARTITION STATUS
-------- ------------------------------ ------------------------------ --------
Global DATE_IND1_S VALID
Global IND_MSI_LN_LNS1_S VALID
Global L_INDX_MSISDN_S VALID
SQL > explain plan for the small circle of MSISDN, TID, of LOAN_RUNNING_DETAILS_SOUTH where LOANDATETIME < = sysdate-2 and LOANDATETIME > sysdate-15 and LOANTYPE = 1;
He explained.
SQL > SQL > set line 200
@?/rdbms/admin/utlxpls.sql
SQL >
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Hash value of plan: 3659874059
------------------------------------------------------------------------------------------------------------------
| ID | Operation | Name | Lines | Bytes | Cost (% CPU). Time | Pstart. Pstop |
------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1448K | 58 M | 21973 (2) | 00:04:24 | | |
|* 1 | FILTER | | | | | | | |
| 2. PARTITION LIST ALL | | 1448K | 58 M | 21973 (2) | 00:04:24 | 1. 11.
|* 3 | TABLE ACCESS FULL | LOAN_RUNNING_DETAILS_SOUTH | 1448K | 58 M | 21973 (2) | 00:04:24 | 1. 11.
------------------------------------------------------------------------------------------------------------------
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Information of predicates (identified by the operation identity card):
---------------------------------------------------
1 - filter(SYSDATE@!-2>SYSDATE@!-15)
3 - filter("LOANTYPE"=1 AND "LOANDATETIME">SYSDATE@!-15 AND "LOANDATETIME"<=SYSDATE@!-2)
16 selected lines.
Second server
SQL > select index_name, column_name, position_colonne from user_ind_columns where table_name = 'LOAN_RUNNING_DETAILS_SOUTH"of order 1.
INDEX_NAME COLUMN_NAME POSITION_COLONNE
------------------------------ -------------------- ---------------
DATE_IND1_S LOANDATETIME 1
IND_MSI_LN_LNS1_S MSISDN 1
IND_MSI_LN_LNS1_S LOANDATETIME 2
IND_MSI_LN_LNS1_S LOANSTATUS 3
LAST_INDEX L_INDX_MSISDN_S 1
MSISDN L_INDX_MSISDN_S 2
SQL > select decode (status, 'N/a', 'Part Hdr', 'Global') ind_type, index_name, NULL nom_partition, status
2 from user_indexes where table_name = 'LOAN_RUNNING_DETAILS_SOUTH '.
Union
3 4 Select 'Local' ind_type, index_name, nom_partition, status
5 to user_ind_partitions where index-name in (select index_name in user_indexes where table_name = 'LOAN_RUNNING_DETAILS_SOUTH')
6 order of 1,2,3;
IND_TYPE INDEX_NAME NOM_PARTITION STATUS
-------- ------------------------------ ------------------------------ --------
Global DATE_IND1_S VALID
Global IND_MSI_LN_LNS1_S VALID
Global L_INDX_MSISDN_S VALID
SQL > explain plan for the small circle of MSISDN, TID, of LOAN_RUNNING_DETAILS_SOUTH where LOANDATETIME < = sysdate-2 and LOANDATETIME > sysdate-15 and LOANTYPE = 1;
SQL > set line 200
@?/rdbms/admin/utlxpls.sql
SQL >
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Hash value of plan: 1161680601
----------------------------------------------------------------------------------------------------------------------------------
| ID | Operation | Name | Lines | Bytes | Cost (% CPU). Time | Pstart. Pstop |
----------------------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 2. 84. 5 (0) | 00:00:01 | | |
|* 1 | FILTER | | | | | | | |
|* 2 | TABLE ACCESS BY INDEX ROWID | LOAN_RUNNING_DETAILS_SOUTH | 2. 84. 5 (0) | 00:00:01 | ROWID | ROWID |
|* 3 | INDEX RANGE SCAN | DATE_IND1_S | 2. | 3 (0) | 00:00:01 | | |
----------------------------------------------------------------------------------------------------------------------------------
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Information of predicates (identified by the operation identity card):
---------------------------------------------------
1 - filter(SYSDATE@!-2>SYSDATE@!-15)
2 - filter ("LOANTYPE" = 1)
3 - access("LOANDATETIME">SYSDATE@!-15 AND "LOANDATETIME"<=SYSDATE@!-2)
17 selected lines.
Reg,
Hard
Hi , HemantKChitale,
I also update statistics manual as you say, but not see 'TABLE ACCESS FULL' good result
What should I do? my need of production tuning, but I cannot able tune this...
SQL > exec dbms_stats.gather_table_stats (-online 'ttt' ownname, tabname => 'LOAN_RUNNING_DETAILS_SOUTH', cascade => TRUE, estimate_percent => NULL, method_opt => 'for all columns size 254', => of degree 4);
PL/SQL procedure successfully completed.
SQL > explain plan for the small circle of MSISDN, TID, of LOAN_RUNNING_DETAILS_SOUTH where LOANDATETIME<=sysdate-2 and="" loandatetime="">sysdate-15 and LOANTYPE = 1;
He explained.
SQL > set line 200
@?/rdbms/admin/utlxpls.sql
SQL >
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Hash value of plan: 3659874059
------------------------------------------------------------------------------------------------------------------
| ID | Operation | Name | Lines | Bytes | Cost (% CPU). Time | Pstart. Pstop |
------------------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1874K | 75 M | 19626 (2) | 00:03:56 | | |
|* 1 | FILTER | | | | | | | |
| 2. PARTITION LIST ALL | | 1874K | 75 M | 19626 (2) | 00:03:56 | 1. 11.
|* 3 | TABLE ACCESS FULL | LOAN_RUNNING_DETAILS_SOUTH | 1874K | 75 M | 19626 (2) | 00:03:56 | 1. 11.
------------------------------------------------------------------------------------------------------------------
PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Information of predicates (identified by the operation identity card):
---------------------------------------------------
1 - filter(SYSDATE@!-2>SYSDATE@!-15)
3 - filter("LOANDATETIME">SYSDATE@!-15 AND "LOANTYPE"=1 AND "LOANDATETIME")<>
16 selected lines.
-
Help to understand the plan of the explain command
Hello world
I have a question:
INSERT INTO ICM_UPSEL_1_ALL
Select a.customer_no, a.pr_code_bbl, pr_code_pmm, a.score, a.price_diff, a.flag (b.pr_code_pmm) min
ICM_UPSEL_MIN_PRDIFF_1 a, icm_pre b
where a.customer_no = b.customer_no
and a.pr_code_bbl = b.pr_code_bbl
and a.score = b.score
and a.flag = b.flag
and a.price_diff = b.price_diff
and b.price_diff > 0
and b.score > = 0.5
and b.flag = 1
and b.price_diff > 0 and b.price_diff < = 10
Group of a.customer_no, a.pr_code_bbl, a.score, a.price_diff, a.flag
This query runs in a pl/sql procedure. The plan of the explain for this query command is:
Hash value of plan: 3124235498
------------------------------------------------------------------------------------------------------------------
| ID | Operation | Name | Lines | Bytes | Cost (% CPU). Time |
------------------------------------------------------------------------------------------------------------------
| 0 | INSERT STATEMENT. | 1. 143. 4 (25) | 00:00:01 |
| 1. LOAD TABLE CLASSIC | ICM_UPSEL_1_ALL | | | | |
| 2. HASH GROUP BY. | 1. 143. 4 (25) | 00:00:01 |
| 3. NESTED LOOPS | | | | | |
| 4. NESTED LOOPS | | 1. 143. 3 (0) | 00:00:01 |
|* 5 | TABLE ACCESS BY INDEX ROWID | ICM_UPSEL_MIN_PRDIFF_1 | 1. 65. 1 (0) | 00:00:01 |
|* 6 | INDEX RANGE SCAN | MPD_1_FLAG_IDX | 1. | 1 (0) | 00:00:01 |
|* 7 | INDEX RANGE SCAN | ICM_PRE_FLAG_IDX | 1. | 2 (0) | 00:00:01 |
|* 8 | TABLE ACCESS BY INDEX ROWID | ICM_PRE | 1. 78. 2 (0) | 00:00:01 |
------------------------------------------------------------------------------------------------------------------
Information of predicates (identified by the operation identity card):
---------------------------------------------------
5 - filter("A".") PRICE_DIFF"> 0 AND 'A '. "SCORE" > = 0.5 AND 'A' "." " (PRICE_DIFF"< = 10)
6 - access("A".") FLAG "= 1)
7 - access("B".") FLAG "= 1)
8 - filter("B".") PRICE_DIFF"< = 10 AND 'B '. "SCORE" > = 0.5 AND 'B' "." " PRICE_DIFF"> 0 AND
"A"." CUSTOMER_NO '=' B '. "' CUSTOMER_NO ' AND 'A '. "" PR_CODE_BBL "=" B ". "' PR_CODE_BBL ' AND 'A '. "SCORE" = 'B' "." " SCORE '.
AND 'A '. "" PRICE_DIFF "=" B ". ("' PRICE_DIFF")
This query runs for 10 hours now and still no results. Could someone please help me where goes wrong... Why is - it take as long to insert data...
The number of rows in ICM_UPSEL_MIN_PRDIFF_1: 84 858
The number of rows in icm_pre: 455,500,944
All columns are indexed.
My version of the database is 11.1.0.6 running on windows server 2003 R2.
Thanks in advance
This plan seems much more suitable for me: instead of doing > 80 K of the loops of the CBO now decides to make a unique HASH JOIN. I would try to run the operation with the dynamic_sampling flag. If the system statistics are not completely misleading the operation now only takes a few minutes (and no 10: 00 >).
Concerning
Martin
-
Tuning SQL - do not understand the plan of the explain command.
Hi all
I use 11g R2 and I have 2 questions about the setting of a query.
My relevant parameters for the optimizer are:BANNER -------------------------------------------------------------------------------- Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production PL/SQL Release 11.2.0.3.0 - Production CORE 11.2.0.3.0 Production TNS for 64-bit Windows: Version 11.2.0.3.0 - Production NLSRTL Version 11.2.0.3.0 - Production
The query I want to run is quite simple. It returns a few exceptions with a filter.NAME TYPE VALUE ------------------------------------ ----------- ----------------------- -- optimizer_capture_sql_plan_baselines boolean FALSE optimizer_dynamic_sampling integer 2 optimizer_features_enable string 11.2.0.3 optimizer_index_caching integer 0 optimizer_index_cost_adj integer 100 optimizer_mode string ALL_ROWS optimizer_secure_view_merging boolean TRUE optimizer_use_invisible_indexes boolean FALSE optimizer_use_pending_statistics boolean FALSE optimizer_use_sql_plan_baselines boolean TRUE
The execution plan is the following:SELECT ERO.DVC_EVT_ID, E.DVC_EVT_DTTM FROM D1_DVC_EVT E, D1_DVC_EVT_REL_OBJ ERO WHERE ERO.MAINT_OBJ_CD = 'D1-DEVICE' AND ERO.PK_VALUE1 = :H1 AND ERO.DVC_EVT_ID = E.DVC_EVT_ID AND E.DVC_EVT_TYPE_CD IN ('END-GSMLOWLEVEL-EXCP-SEV-1', 'STR-GSMLOWLEVEL-EXCP-SEV-1') ORDER BY E.DVC_EVT_DTTM DESC;
So, as you can see, rank 8, I have a TABLE ACCESS BY GLOBAL INDEX ROWID. But what I am failling to see is how Oracle display an ACCESS BY GLOBAL INDEX ROWID TABLE without using any index. As the thought that Oracle has always been a ROWID due to an index.Plan hash value: 3627978539 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | Id | Operation | Name | Starts | E-Rows |E-Bytes| Cost (%CPU)| Pstart| Pstop | A-Rows | A-Time | Buffers | Reads | OMem | 1Mem | Used-Mem | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 1 | | | 7131 (100)| | | 1181 |00:00:17.17 | 8627 | 2978 | | | | | 1 | SORT ORDER BY | | 1 | 3137 | 275K| 7131 (1)| | | 1181 |00:00:17.17 | 8627 | 2978 | 80896 | 80896 |71680 (0)| | 2 | NESTED LOOPS | | 1 | | | | | | 1181 |00:00:17.16 | 8627 | 2978 | | | | | 3 | NESTED LOOPS | | 1 | 3137 | 275K| 7130 (1)| | | 2058 |00:00:08.09 | 6709 | 1376 | | | | | 4 | TABLE ACCESS BY INDEX ROWID | D1_DVC_EVT_REL_OBJ | 1 | 3137 | 125K| 845 (1)| | | 2058 |00:00:04.37 | 820 | 799 | | | | |* 5 | INDEX RANGE SCAN | D1T404S0 | 1 | 3137 | | 42 (0)| | | 2058 |00:00:00.08 | 27 | 23 | | | | | 6 | PARTITION RANGE ITERATOR | | 2058 | 1 | | 1 (0)| KEY | KEY | 2058 |00:00:03.69 | 5889 | 577 | | | | |* 7 | INDEX UNIQUE SCAN | D1T400P0 | 2058 | 1 | | 1 (0)| KEY | KEY | 2058 |00:00:03.66 | 5889 | 577 | | | | |* 8 | TABLE ACCESS BY GLOBAL INDEX ROWID| D1_DVC_EVT | 2058 | 1 | 49 | 2 (0)| ROWID | ROWID | 1181 |00:00:09.05 | 1918 | 1602 | | | | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Peeked Binds (identified by position): -------------------------------------- 1 - (VARCHAR2(30), CSID=178): '271792300706' Predicate Information (identified by operation id): --------------------------------------------------- 5 - access("ERO"."PK_VALUE1"=:H1 AND "ERO"."MAINT_OBJ_CD"='D1-DEVICE') filter("ERO"."MAINT_OBJ_CD"='D1-DEVICE') 7 - access("ERO"."DVC_EVT_ID"="E"."DVC_EVT_ID") 8 - filter(("E"."DVC_EVT_TYPE_CD"='END-GSMLOWLEVEL-EXCP-SEV-1' OR "E"."DVC_EVT_TYPE_CD"='STR-GSMLOWLEVEL-EXCP-SEV-1'))
I also have an index on the DVC_EVT_TYPE_CD column in my table (line 8 of the predicate information section)
And finally, what would be your suggestions to improve the performance of this query...
Thank you very much.If I read content (RowSource) statistical information and plan properly, then access the table in step 4 on D1_DVC_EVT_REL_OBJ needs dry 4.37 and accesses 820 buffers. If you had an index D1_DVC_EVT_REL_OBJ (MAINT_OBJ_CD, PK_VALUE1, DVC_EVT_ID) I guess this table access could be avoided. To avoid the access of the table on D1_DVC_EVT an index on D1_DVC_EVT (DVC_EVT_ID, DVC_EVT_TYPE_CD, DVC_EVT_DTTM) should be sufficient.
I think that these indices would improve the performance of the query, but of course, they would have a negative impact on the performance of DML on the table and they could influence other queries on the table.
-
Hi all
What is the cost of the CPU in the execution plan?
Thank you
JohnHello
I like one of the best explanations and demonstration was given in Charles hooper blog. I have the same explained, but it is clean of charles himself.
refer to the link... http://hoopercharles.wordpress.com/2010/02/19/What-is-the-meaning-of-the-CPU-column-in-an-EXPLAIN-plan/I like the most.
HTH
-Pavan Kumar N
-
The CPU used by the resource manager.
Hello
in 10g R2, I created a Plan for DBMS resources manager:
Now, SCOTT runs the following (example from 24 through the database manager):EXEC DBMS_RESOURCE_MANAGER.CREATE_PENDING_AREA(); EXEC DBMS_RESOURCE_MANAGER.CREATE_SIMPLE_PLAN(SIMPLE_PLAN => 'simple_plan1',CONSUMER_GROUP1 => 'mygroup1', GROUP1_CPU => 80,CONSUMER_GROUP2 => 'mygroup2', GROUP2_CPU => 20); ALTER SYSTEM SET RESOURCE_MANAGER_PLAN ='simple_plan1'; EXEC DBMS_RESOURCE_MANAGER.SWITCH_CONSUMER_GROUP_FOR_USER ('SCOTT', 'MYGROUP2'); EXEC DBMS_RESOURCE_MANAGER_PRIVS.GRANT_SWITCH_CONSUMER_GROUP ('SCOTT', 'MYGROUP2', TRUE); EXEC DBMS_RESOURCE_MANAGER.SUBMIT_PENDING_AREA; EXEC DBMS_RESOURCE_MANAGER.CLEAR_PENDING_AREA;
And CPU use goes to 100%. Why? According to its consumer group should not only have 20% of CPU?DECLARE m NUMBER; BEGIN FOR i IN 1..100000 LOOP FOR j IN 1..100000 LOOP /* * The following query does a cartesian product without * a predicate and takes up significant CPU time. */ select count(*) into m from v$sysstat, v$system_event; END LOOP; END LOOP; END; END; /
No explanation?
Thank you.Hello
user522961 wrote:
And CPU use goes to 100%. Why? According to its consumer group should not only have 20% of CPU?A quote from docs explained:
The currently active resource plan does not have the limits of the allowance until the CPU usage is 100%. If the CPU usage is less than 100%, the database is not related to the CPU and there is therefore no need to enforce limits to ensure that all the sessions to get their allocation of designated resources.
In addition, when the limits are applied, allowance unused by consumer groups can be used by other groups of consumers. In the previous example, if the SALES group does not use all of his allowance, the Resource Manager allows the group to MARKET or DEVELOP it group to use the unused allocation. -
I have an iMac end of 2012, and lately he worked very slowly. A few days ago, I checked the activity monitor to see "Installer" take a lot of CPU. I leave and rejected it as an error, but sometimes it starts to do it again. I just kept smoking them until I found several installers taking up half of the space of the CPU. I did some research and I read that the Genio virus, but there is no house icon in the upper right corner and I just can't find any files for it.
Can download some adware or malware. A simple, fast and safe way to remove adware and malicious software is to download and to use Malwarebytes Anti-Malware for Mac. There has been developed by one of the most respected contributors and the top of the page in these forums and recommended by almost all of the top contributors here.
If you prefer to do it manually, follow these instructions in the site theSafeMac.com (author of Malwarebytes): adware removal Guide
If you want to check to see what adware/malware, you may have before you run Malwarebytes and then download and run Etrecheck. Copy and paste the results in your response. Etrecheck is a diagnostic tool that has been developed by one of the most respected users here (and recommended by Apple Support personnel) of the CSA to provide an overview of the system and help identify the most obvious culprits that can adversely affect the performance of a Mac.
-
Running process crushing the CPU
Hey guys,.
I installed Sierra on my Macbook Pro 13 "2015 today and had a problem where the process of"photoanalysisd"ran all the time. I used between 120-150% of the CPU. When I force quit, he came right away. After a few hours, I went back to el Capitan. If there are others out there who have had the same experience and a possible solution would be greatly appreciated. Thank you.
I think the "photoanalysisd" process of facial recognition / scene on your photo library. I've seen this as well as a few other processes using higher than normal amounts of CPU. A few hours later, everything is back to normal once all the new features have done their thing. Depending on how many files, the speed of your system (CPU and storage read/write), and all what you use, this background process will end eventually. It took a few hours for me to see that I had the minimal use of the processor idling. I hope this helps.
Maybe you are looking for
-
How to reinstall Satellite M100 under license of Windows Vista Edition basic family?
How to get a recovery of files for Toshiba M100 under license of Windows Vista Home Basic?
-
Can I Add Bluetooth in Satellite P100-194?
Can I Add Bluetooth to my P100-194? No extern Stick. Or infrared? I would like to connect my wireless UMTS-Mobile.
-
Have a w 2014 15 "MBP 500 MB SSD. Can I switch to a 1 TB SSD? Apple will do this or this is a job that can be done, but not supported by Apple? Thank you!
-
Search for HDD compatible upgrade the Satellite Pro L10
Hi ive just bought a satellite pro L10, who needs a new drive hard and im new to computers laptops so im wondering if there is a particular model of hard drive I need and there at - it those who recommend or avoid. any advice will be greatly appiecia
-
How will I know if my phone is Bluetooth compatible?
I want to use the Bluetooth on my laptop, but I would first find if it is already installed. How can I find and how do I use if it is installed?