Logical reads VS physical reads

hello all,
What is the difference between logical reads and physical read ??? I do get the part of logical read (from buffer) and physical read(from disk). And also do know physical reads are bad, but is it true in oracle world..logical reads are bad ??? if so why ??? Could you please explain which on to look for. As i am going thru AWR report and i see segemts by logical/physical read.

The first and foremost difference is that physcial reads are done from the hard disk. And this is always going tobe slwoer than the memory. That's why its said that the physcial ios must be removed. Logical ios are good as they are done from the memory.The theory that logical are also not good is because the logical ios require the access given by latches. So with lots of the gets for the latches put them in to contention and latch contention would make the access to the logical io slower.So its better to do this in less IO even they are logical too.
Lots of logical IOs probably mean that you are accessing unnecessary data which may be not required.
HTH
Aman....

Similar Messages

Logical read Vs Physical Reads

Hi,
I want to know How should we balance the logical reads and physical reads on database.
I generaly assume logical reads is always better but what if too many session accessing the same
objects it will cause latch contention.
If it is physical read then it will take more time to read from disk than reading from SGA.
If I look at statspack report I check the
Logical reads: 59,846.54 840.03
Physical reads: 1,095.91 15.38
Is there any standards for logical reads on database. If there is latch contentions can we go for the
decreasing the SGA to avoid the contentions so that there will balance between the physical reads
and logical reads.

amitbansode wrote:
I want to know How should we balance the logical reads and physical reads on database.There is no balance as it implies some kind of "+perfect ratio+" for logical I/O vs. physical I/O.
A high percentage of physical I/O can be perfectly acceptable and normal and correct for a specific database (e.g. think of a system collecting telemetry data where 90% or more of all I/O is writing new telemetry into the database and the remaining 10% is querying the data, with old data being aged out from the database using partition drops which is negligible I/O).
A high percentage of logical I/O can be indicative of a serious application design problem - where 80GB of database data is read and read again and again and again.. resulting in over a TB of logical I/O. (actually saw this in production database some years ago)
So there is no balance (e.g. not true that physical I/O = BAD and logical I/O = GOOD). No perfect cache hit ratio figure that tells you that the database is doing the right amounts of logical and physical I/O.
And I want to emphasise what Mark said - tuning requires you to identify the performance problem first, before trying to solve it.
It is very dangerous to take one metric, like the I/O cache hit ratio, and attempt to tune that. It alone is meaningless. Just like memory utilisation alone is useless and CPU utilisation alone is useless. It does not by any means point to an actual performance problem. E.g. 100% CPU utilisation can mean hardware has insufficient horses, instead of performance issues related with application design, database setting or kernel configuration.
I often repeat the following mantra here on OTN - a fundamental concept IMO for software engineering:
A solution is only as good as the problem definition.
Identify the problem first - correctly and comprehensively. And then solve it.

Oracle 9i Performance Issue High Physical Reads

Dear All,
I have Oracle 9i Release 9.2.0.5.0 database under HP Unix, I have run the query and got following output. Can any body just have a look and advise what to do in the following situation? We have performance issues.
Many thanks in advance
Buffer Pool Advisory for DB: DBPR Instance: DBPR End Snap: 902
-> Only rows with estimated physical reads >0 are displayed
Size for Size Buffers for Est Physical Estimated
P Estimate (M) Factr Estimate Read Factor Physical Reads
D 416 .1 51,610 4.27 1,185,670,652
D 832 .2 103,220 2.97 825,437,374
D 1,248 .3 154,830 2.03 563,139,985
D 1,664 .4 206,440 1.49 412,550,232
D 2,080 .5 258,050 1.32 366,745,510
D 2,496 .6 309,660 1.23 340,820,773
D 2,912 .7 361,270 1.14 317,544,771
D 3,328 .8 412,880 1.09 301,680,173
D 3,744 .9 464,490 1.04 288,191,418
D 4,096 1.0 508,160 1.00 276,929,627

Hi,
Actually you didnt give the exact problem statement.
Seems to be your database is I/O bound. Ok, do the following one by one:
1. Identify the FTS queries and try to create the optimal indexes (depending on the disk reads factor!!) on the problem queries.
2. To reduce the 276M physical reads, you need to allocate more memory to db_cache_size. try 8GB (initially) and then depending on the buffer advisery you can increase further if you have more memory on the box.
3. as a Next step , configure KEEP and RECYCLE cache to get the benefits of reduced I/O by multiple pools. Allocate objects to the KEEP/RECYCLE pools.
Thanks,

No physical reads, plenty of consistant gets

Hi All,
Oracle v11.2 on Linux.
Please have a look at the query I ran and the output. The SQL_ID is of a SELECT command.
What does this situation represents, where there are zero physical reads and plenty of consistant gets ?
For consistant gets, we do read undo information (correct ??), if that undo is read from disk, will that be a "physical read"? i.e. if we read disks for consistant gets, will that be counted under physical reads or not ?
How can I describe the exact data retrival of the command here? Is it the case of "everything it needs is found in the buffer cache" ?
select a.sid, a.value , B.NAME , s.sql_id
from v$sesstat a, v$statname b, v$session s
where A.STATISTIC# = B.STATISTIC#
and b.name in ( 'redo size','physical read bytes','physical reads cache','consistent gets' )
and a.sid = s.sid
and a.sid=1018
order by a.sid;
       SID      VALUE NAME                                                             SQL_ID
      1018    7281396 consistent gets                                                  434u36htuz0s9
      1018          0 physical reads cache                                             434u36htuz0s9
      1018          0 physical read bytes                                              434u36htuz0s9
      1018       4448 redo size                                                        434u36htuz0s9
4 rows selected.Thanks in advance.

>
There are no physical reads, so weather its doing consistant-gets or not, can I say, all data required for the SELECT was in buffer cache ?
>
The data for those system views is cached in memory so Oracle does not read the disk (except at startup) to gather the information.
Some static information, like dictionary objects, is stored stored on disk in the system tablespace but this data is read when the database is mounted and stored in memory structures.
Other dynamic information, like session info, is only stored in memory structures (similar to C arrays) and Oracle can query these as if they were tables.
So no, the data was not in the buffer cache. It was already in system memory.

Understanding Physical Reads

Hello everyone,
I'm trying to understand about physical reads statistics.
we have a few statistics related to this.
They are physical reads, physical reads cache, physical reads cache prefetch, physical reads direct, physical reads direct (lob), physical reads direct temporary tablespace, and physical reads prefetch warmup.
From all those statistics, I thought that physical reads should be like the total of the rest of the other statistics.
However, when I looked at the actual AWR report I collected from a production database here, it seems physical reads value is only similar to the value of physical reads cache + physical reads direct.
I wonder about the rest of the statistics related to this physical reads, I thought that physical reads statistics is like the grand total of all those physical reads, however that doesn't seems to be the case. How do I interpret these statistcs?
Appreciate if you can explain how these statistics can be used when understanding a database performance.
Thank you,
Adhika

Hello everyone,
I'm trying to understand about physical reads statistics.
we have a few statistics related to this.
They are physical reads, physical reads cache, physical reads cache prefetch, physical reads direct, physical reads direct (lob), physical reads direct temporary tablespace, and physical reads prefetch warmup.
From all those statistics, I thought that physical reads should be like the total of the rest of the other statistics.
However, when I looked at the actual AWR report I collected from a production database here, it seems physical reads value is only similar to the value of physical reads cache + physical reads direct.
I wonder about the rest of the statistics related to this physical reads, I thought that physical reads statistics is like the grand total of all those physical reads, however that doesn't seems to be the case. How do I interpret these statistcs?
Appreciate if you can explain how these statistics can be used when understanding a database performance.
Thank you,
Adhika

Oracle 11g R2 - AWR Section UnOptimized Read Reqs / Optimized Read Reqs

Hello guys,
using Oracle 11g R2 more and more i have checked out the new AWR and its sections.
I have found some section lke this
SQL ordered by Physical Reads (UnOptimized)DB/Inst: SID/sid Snaps: 20296-202
-> UnOptimized Read Reqs = Physical Read Reqts - Optimized Read Reqs
-> %Opt   - Optimized Reads as percentage of SQL Read Requests
-> %Total - UnOptimized Read Reqs as a percentage of Total UnOptimized Read Reqs
-> Total Physical Read Requests:         151,508
-> Captured SQL account for   25.3% of Total
-> Total UnOptimized Read Requests:         151,508
-> Captured SQL account for   25.3% of Total
-> Total Optimized Read Requests:               1
-> Captured SQL account for    0.0% of TotalWhat the heck is "Optimized Read Reqs" and "UnOptimized Read Reqs"? These terms are used very often right now in AWR of Oracle 11g R2.
Does anyone know what this term means and how it is defined? Don't find any information on web and documentation.
Thanks guys!

Hello,
If my guess is close than "Buffer Hit" (Instance Efficiency Percentages) could be as low as 0% from this report.No it isn't ... check it here:
Instance Efficiency Percentages (Target 100%)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
            Buffer Nowait %:   99.96       Redo NoWait %: 100.00
            Buffer Hit   %:   99.50    In-memory Sort %: 100.00
            Library Hit   %:   99.09        Soft Parse %:   89.55
         Execute to Parse %:   95.93         Latch Hit %:   98.93
Parse CPU to Parse Elapsd %:   68.68     % Non-Parse CPU:   98.81Sometimes i wonder why oracle introduces such new terms / measurements without documenting it.
Regards

Find physical reads and logical reads ?

Hi,
how will find out physical reads and logical reads ?

Well I would suggest you read the report from statspack/awr.They as suggested by Amit,have a load profile section.That would be helpful for you in finding the details of this thing. Also which version you are? If you are in 10g than the EM is able to give you a compare period report where you can compare 2 different days's periods information and can check which particular part has changed.
In addition to this , look for the information of the metrics in the documentation.As the physical read and logical reads are statistics which are happening in the system.So from 10g onwards, oracle is keeping a track in the deflectionin the statistics. So if you see that than it willbe easy for you to manage and monitor it.
I shall try to findthe name of some views related to it and post.
Aman....

Physical reads, writes and logical reads,writes

hi there,
is there any way to find the physical reads and writes, logical reads and writes
other than statspack report?

It depends which version of Oracle are you using.
If you are using 10g, you can get very comprehensive information from the dictionary views.
If you are on <=9i, then, use, v$sysstat and look for 'physical reads', 'logical reads'.
The values dervied from v$sysstat are cummulative values. i.e. values counted since the instance started.
Jaffar

Physical read & logical read

Hi
What is ment by Physical-Read and Logical-Read in oracle?Can anybody explain me clearly.
Aqueel.

Physical read, reading from the disk and logical reads mean reading it from the memory.
When a session(query) required information (data blocks), it will first look into the buffer cache (SGA), if it find the required data block, it will read it from the memory, i.e. logical reads, but, if it doesn't find required data block in the memory, then it read from the disk(datafile), physical read.
Generally, people motive to reduce physical read and read as much as possible from the memory, I was also int he same wrong impression until I came across a good paper by tuning expert Cary Milsap, 'why we should focus on LIO instead of PIO'. also, search http://asktom.oracle.com 'logical reads', there are good discussion on this point.
Jaffar

Consistent gets and physical reads

Hi all,
I am tuning a DM SQL query, by comparing execution plans with STAR TRANSFORMATION enabled or disabled. I got the following results:
STAR TRANSFORMATION ON
74889 consistent gets
254365 physical reads
STAR TRANSFORMATION OFF
1945892 consistent gets
168028 physical reads
I thought a physical read would be counted as a logical read as well, because the data block would be read from disk (1 physical IO), placed in the buffer cache and then read from there (1 more logical IO or consistent get).
So, one physical IO does not cause a logical IO?
Thanks!
Edited by: user10634835 on 12-Jul-2011 08:40

But shouldn't consistent gets be >= physical reads (Since, as per my understanding, 1 PIO causes at least 1 LIO)? In this case it is not.
74889 consistent gets
254365 physical readsJust clarifying for my knowledge.
regards

No of physical reads of buffers from disk on PROD and Clone has huge differ

Please help !
I am having a problem interpreting a tkprof report.
I am executing the same statement below in PROD and Clone of PROD.
What I do not understand why the tkprof report shows huge difference in disk reads for same executions from PROD and Clone of PROD. I.e. first execution disk reads = 757187 , second = 6.
I expected to see Similar disk reads in the second execution also.The data it fetches is same in both the Instances.
Statment
SELECT NVL(SUM(RG_OTH.DEBIT),0),NVL(SUM(RG_OTH.CREDIT),0)
FROM
JA_IN_PLA PLA,JAI_RG_OTHERS RG_OTH,HR_ALL_ORGANIZATION_UNITS HOU,
HR_LOCATIONS LOC,JA_IN_HR_ORGANIZATION_UNITS HR_ORG WHERE 1 = 1 AND
PLA.ORGANIZATION_ID = :b1 AND TRUNC(PLA.CREATION_DATE) >= :b2 AND
TRUNC(PLA.CREATION_DATE) <= :b3 AND HOU.ORGANIZATION_ID =
PLA.ORGANIZATION_ID AND LOC.LOCATION_ID = PLA.LOCATION_ID AND
HR_ORG.ORGANIZATION_ID = PLA.ORGANIZATION_ID AND HR_ORG.LOCATION_ID =
PLA.LOCATION_ID AND TRANSACTION_ID = 19 AND INVENTORY_ITEM_ID != 0 AND
NVL(ROUNDING_ID,1) != -1 AND RG_OTH.SOURCE_REGISTER_ID = PLA.REGISTER_ID
AND RG_OTH.SOURCE_REGISTER = 'PLA' AND RG_OTH.TAX_TYPE IN (
'EXCISE_EDUCATION_CESS' )
Tkproff from PROD.
call count cpu elapsed disk query current rows
Parse 1 0.01 0.01 0 0 0 0
Execute 1 0.04 0.03 0 0 0 0
Fetch 1 973.41 1724.95 757187 789442 0 1
total 3 973.46 1725.01 757187 789442 0 1
Tkproff from CLONE of PROD.
call count cpu elapsed disk query current rows
Parse 1 0.06 0.05 1 789 0 0
Execute 1 0.00 0.00 0 0 0 0
Fetch 1 0.73 0.72 6 30594 0 1
total 3 0.79 0.78 7 31383 0 1

wins wrote:
SELECT NVL(SUM(RG_OTH.DEBIT),0),NVL(SUM(RG_OTH.CREDIT),0)
FROM
JA_IN_PLA PLA,JAI_RG_OTHERS RG_OTH,HR_ALL_ORGANIZATION_UNITS HOU,
HR_LOCATIONS LOC,JA_IN_HR_ORGANIZATION_UNITS HR_ORG WHERE 1 = 1 AND
PLA.ORGANIZATION_ID = :b1 AND TRUNC(PLA.CREATION_DATE) >= :b2 AND
TRUNC(PLA.CREATION_DATE) <= :b3 AND HOU.ORGANIZATION_ID =
PLA.ORGANIZATION_ID AND LOC.LOCATION_ID = PLA.LOCATION_ID AND
HR_ORG.ORGANIZATION_ID = PLA.ORGANIZATION_ID AND HR_ORG.LOCATION_ID =
PLA.LOCATION_ID AND TRANSACTION_ID = 19 AND INVENTORY_ITEM_ID != 0 AND
NVL(ROUNDING_ID,1) != -1 AND RG_OTH.SOURCE_REGISTER_ID = PLA.REGISTER_ID
AND RG_OTH.SOURCE_REGISTER = 'PLA' AND RG_OTH.TAX_TYPE IN (
'EXCISE_EDUCATION_CESS' )
Tkproff from PROD.
call count cpu elapsed disk query current rows
Parse 1 0.01 0.01 0 0 0 0
Execute 1 0.04 0.03 0 0 0 0
Fetch 1 973.41 1724.95 757187 789442 0 1
total 3 973.46 1725.01 757187 789442 0 1
Tkproff from CLONE of PROD.
call count cpu elapsed disk query current rows
Parse 1 0.06 0.05 1 789 0 0
Execute 1 0.00 0.00 0 0 0 0
Fetch 1 0.73 0.72 6 30594 0 1
total 3 0.79 0.78 7 31383 0 1
As others have pointed out, this looks like a change in execution plan. At a guess, the PROD database is doing a tablescan somewhere (physical reads is approximately equal to logical reads is a pointer).
There are many reasons why this could happen, continuing with the guesswork - we note that the statement was re-optimised (hard parsed) in the clone, but used an existing cursor in production (logical I/O = 0 in prod, 789 in clone). So the plan you got in the clone was the one that Oracle thought best for the inputs you had supplied. The plan in prod could have been produced as the response to a completely different set of input; since you have two bind variables applied to a range based predicated and one bind variable with equality it's possible that you have a histogram that makes a difference on the equality column, alternatively the range of dates used to generate the original plan was very different from the range used in the test.
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
Author: Oracle Core

Corruption Parameters Increase High Physical Read for one query

Hi Oracle Experts,
Here is what I am currently facing in my non-prod environment:
We are testing out corruption parameters in non-prod environment and doing a perf test for them and we found that one SELECT query has seen significant decreases in performance; What I mean by that is after adding corruption parameters query is executing 40 sec compare to 18 sec and less. Also, this query is widely use by user in Prod environment, so performance degradation will create serious impact on their work.
I have generated AWR Diff for baseline and perf test following are parameter we have set for corruption test:
db_block_checking (Baseline) LOW (Perftest) MEDIUM
db_block_checksum (Baseline) TRUE (Perftest) FULL
Load Profile from AWR Diff are below:
Load Profile
1st Per Sec 2nd Per Sec %Diff 1st Per Txn 2nd Per Txn %Diff
Redo size: 758,356.06 752,760.94 -0.74 67,161.76 66,631.71 -0.79
Logical reads: 104,637.62 108,677.76 3.86 9,266.95 9,619.77 3.81
Block changes: 1,578.11 1,560.15 -1.14 139.76 139.76 0.00
Physical reads: 103.78 544.41 424.58 9.19 48.19 424.37
Physical writes: 108.94 107.13 -1.66 9.65 9.48 -1.76
User calls: 3,477.02 3,497.26 0.58 307.93 309.57 0.53
Parses: 948.36 949.61 0.13 83.99 84.06 0.08
Hard parses: 0.79 0.54 -31.65 0.07 0.05 -28.57
Sorts: 121.48 120.32 -0.95 10.76 10.65 -1.02
Logons: 0.36 0.27 -25.00 0.03 0.02 -33.33
Executes: 1,575.55 1,591.40 1.01 139.53 140.87 0.96
Transactions: 11.29 11.30 0.09
If we gather stats for tables involved in the query than it performs well in fact with in 5 Sec, but I believe in prod we can't gather state very often.
Questions:
1) How to remedy this situation and have query perform well along with corruption parameters?
2) Does corruption parameters have impact on SELECT query too, I believe it will have impact on INSERT and UPDATE.
3) Any reference to Doc will be highly appreciated.

1) How to remedy this situation and have query perform well along with corruption parameters?Use faster CPU. Checksum is a thing that needs to be computed.
2) Does corruption parameters have impact on SELECT query too, I believe it will have impact on INSERT and UPDATE.According to docs corruption parameters do not have impact on SELECT queries.
However, I believe, checksum has to be recomputed after delayed block cleanout that may be done by SELECT query after big update. ( http://jonathanlewis.wordpress.com/2009/06/16/clean-it-up/ )
3) Any reference to Doc will be highly appreciated.
Why is it difficult for you to find the docs yourself?

Is it possible to have a 10g (R1) LOGICAL standby db in read-ONLY mode?

From what I've gathered reading the 10g (R 1) docs, the logical standby turns read-only when applying sql, and then in read-write mode otherwise. I have to meet a requirement that the remote database be strictly in read-only mode the entire time (so as to not risk writing to it), as well as being as up-to-date as possible off the primary. (Hence I decided on the logical vs. the physical) Does anyone pls. know if I can keep the logical read-only all the time?
Much appreciated,
Sophie

Oh, Ok. I guess i interpreted it differently. The direct quote from the docs: "Although the logical standby database is opened in read/write mode, its target tables for the regenerated SQL are available only for read-only operations." So I guess what it means, as you suggest, is the entire set of tables from the primary, and not just those few that happened to be updated at that particular time frame, are kept in read-only the entire time...

Reducing Physical Reads

Hi
My oracle version is 10.2.0.4
Is there any way to reduce the physical reads apart from tuning the query and index creation.
Can I have my whole table arranged in blocks sequentially one after the other so that my search becomes simple.
Is there any option for that like Coallesce /deallocate unused space/ Compact.

littleboy wrote:
Is there any way to reduce the physical reads apart from tuning the query and index creation.Incorrectly phrased. By reducing PIO (physical I/O) you can imply that you want to increase LIO (logical I/O) as this is faster and will thus increase performance.
That is not tuning. That is hacking of a terrible kind.
In fact, a high percentage LIO is indicative of an application design problem.
The correctly phrased question is "how to reduce I/O?" - as less I/O means less work. And less work does not equate to only less PIOs. It means less I/O (of all kinds). Period.
Can I have my whole table arranged in blocks sequentially one after the other so that my search becomes simple.
Is there any option for that like Coallesce /deallocate unused space/ Compact.From the questions you have asked the past few days, I get the feeling that you are looking for magical silver bullet solutions to performance. A knob to turn somewhere in Oracle, a switch to throw to enable some special behaviour.
That is, and never was, performance tuning. Performance starts with the design of the system. It continues with the architecture used and implemented. And remains with every single line of code written.
You do not pop Oracle's hood and rummage around in the engine, muck about with the fuel injectors, in order to get it to go faster. You design the application to use Oracle correctly. You implemented Oracle correctly. That is where performance start. Not with popping the hood.
Messing with space management to make Oracle go faster? Messing with undocumented parameters? Changing process priorities? Supersizing this and that? That is a FAIL as far as the correct software engineering approach to performance goes.

Physical reads and autotrace

Hii all
In http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:880343948514 tom gives an example table
has a 320 block and the query "select * from table"
autotrace output shows physical reads=46 the statics physical reads= number of block read from disk but table has 320 block how this could be ?
Best regards..

As Oracle has a buffer cache, Oracle will place blocks in buffer cache as soon as it needs it.
When it needs those records subsequently, there will be a consistent get only.
Logical I/O is predictable. Physical I/O depends on the size of the cache.
There will never be a 1 on 1 relationship between size of a table and physical reads or you must disable the buffer cache.
Clarifications about Asktom articles are also best asked at Asktom.
Sybrand Bakker
Senior Oracle DBA
Experts: those who did read documentation.

Logical reads VS physical reads

Similar Messages

Maybe you are looking for