Why would oracle 9i drivers faster than oracle 10g drivers against a 10g?

I'm skeptical of the claim but we have a system at work and tests have been done that apparently is showing that the older oracle 9i thin jdbc driver is performing a fetch faster than the 10g driver. This for a query that is currently doing a full table scan.
Is there a default setting in 10g vs 9i that can explain why the perceived query performance is faster with the older thin driver?

steffi2 wrote:
What was observed was that when they started using the old Oracle 8.1.7 8i client jar against this 10g data the actual execution plan changed dramatically to use indexes where was previously it was not doing so and it was doing a full tablescan.
Why would the introduction of the old 8i jar have this affect?Maybe the test is flawed. For example one test was run with the network was loaded while the other wasn't. Or different connection parameters.
That said I believe somewhere the claim has been made that Oracle drivers changed from one API to another somewhat recently. Thus that could be the source.
Or maybe something to do with hints.

Similar Messages

  • Why would Tx Required be faster than TxNotSupported?

    I have a stateless session bean that exposes a business method which calls another stateless session bean's worker method many times. Both beans use container-managed transactions and are deployed in a WL8.1 server.
              All methods of both beans were originally configured using the "*" wildcard with a trans-attribute of Required. Since this business method is a report-ish method that doesn't change data in any way, I thought I'd speed it up a bit by removing the requirement for a transaction. I did this by adding additional container-transaction tags for the business method and the worker method that use trans-attributes of NotSupported and Supports respectively.
              But instead of running faster, the business method takes about 3 times longer than when transactions were required.
              Something is obviously amiss but I'm not sure where to look. Any suggestions?
              Thanks.

    Ed MacDonald wrote:
              > I have a stateless session bean that exposes a business method which calls another stateless session bean's worker method many times. Both beans use container-managed transactions and are deployed in a WL8.1 server.
              >
              > All methods of both beans were originally configured using the "*" wildcard with a trans-attribute of Required. Since this business method is a report-ish method that doesn't change data in any way, I thought I'd speed it up a bit by removing the requirement for a transaction. I did this by adding additional container-transaction tags for the business method and the worker method that use trans-attributes of NotSupported and Supports respectively.
              >
              > But instead of running faster, the business method takes about 3 times longer than when transactions were required.
              >
              > Something is obviously amiss but I'm not sure where to look. Any suggestions?
              >
              > Thanks.
              Hi. If the DBMS work is done within a transaction, it is most likely that a single pool
              connection is reserved and transparently used for all JDBC in the transaction. However,
              if they are all non-transactional, all the beans will obtain their own connection, and
              if you have the pool set to test-on-reserve, this will happen for each connection reserve.
              Joe

  • I have explation guys..why apple dual qure is faster than other quad qure devices?and appl 1g ram .is faster 2g of other devises?...i wish to get the reason

    i have explation guys..why apple dual qure is faster than other quad qure devices?and appl 1g ram .is faster 2g of other devises?...i wish to get the reason

    It is the OS. have you ever heard of UNIX? If you installed linux on your mac it work run just at fast... windows is very resource intensive. Also apple make sure the drivers are correct and run properly while windows could care less. Someone else can give you more reasons.

  • Why is my MacBook downloading faster than my new Imac?

    I have a two year old Macbook; 2.4 GHz Intel Core 2 Duo, 2GB 667 MHz DDR2 SDRAM with about 30GB left of memory. I also just purchased, from apple, a 27" Imac; 2.8 GHz Intel Core i7 with 4 GB 1067 MHz DDR3. and have just under 850GB left. The Macbook is running on 10.5.8 and the Imac is on 10.6.3.
    My issue is that when i download/stream video to my Imac it takes a substantial amount of time. I love to watch movies, tv shows, ect, on my Imac. I was disappointed by how slow the downloads would take. So I set up my macbook right in front of the Imac to compare the speeds. I went to the Movie Trailers section on apple.com and I clicked on The Last Airbender 1080p (149MB) at the same time for both computers (on WiFi of the same network). My macbook finished the download 5 minutes faster than the Imac. I then tried the 42MB version with a minute difference with the Macbook. I then went to youtube and it was almost the same speed. I tried Itunes and it was about the same speed as well. I then did an internet speed test with both and got these results:
    Macbook: Download 18.45Mb/s, Upload 2.28Mb/s, Ping 93
    IMac: Download 20.88Mb/s, Upload 2.3Mb/s, Ping 89
    I don't know if any of this information helps or if anyone will even read this far but the apple store is extremely far from my house and would try almost anything to fix this issue. My questions are:
    1) Is this a known issue that has a fix, if so how do i fix it?
    2) Is this possibly a faulty computer that i would have to return?
    3) Is this the way it's supposed to operate and if I don't like it take it back
    Thanks in advance for anyone who even reads this post.

    You may also want to check your router. Some routers use a priority bandwidth feature that will dedicate more bandwidth to one machine. If for some reason the macbook's download was started earlier than the imac's, then this might be part of the discrepancy. There are a lot of factors to think about when it comes to wifi bandwidth.
    I do agree with the one comment about testing one computer at a time instead of simultaneously.
    Also, when you said you have the movies from the Macbook on your iMac, can you elaborate? If you are using a shared library, then your iMac is going to be using part of your download speed for updates to your shared itunes library, where your Macbook is only going to be uploading the list. If I am incorrect in my understanding of the sharing of iTunes library, someone please let me know.

  • Why is my 2008 Macbook faster than my 2012 MacBook Pro?

    I got a basic, 13Inch Macbook back in 2008. It has been used, nonstop, for the past almost 5 yrs since i bought it. I never turn it off, it always gets used on a bed/couch, tonnes of windows/programs constantly open. Its full of crappy DLed programs, movies, music and whatnot, has only recently been updated to 10.6 OS X. Its dying, the charger barely works, and it over heats and blacksout sometimes and yet, it runs so much faster than my end of 2012 13in MacBook Pro, 2.9GHz model. Slower to the point that i still use my old one and let Hubby take the new one to the Construction site with him. Ive used it maybe once or twice since i bought it in November. Booting up is slower, general use is slow, opening programs is slower. And its got almost nothing on it!
    Just wondering why its slower when its newer and supposedly better? I thought i was upgrading?

    Wipe the drive on the new system and Reinstall OS X. Factory installs aren't all they are supposed to be. Sometimes corrupted from the Get Go.
    Only by wiping the drive and doing a Fresh Clean Install will you know if your slowness was caused by some type of OS corruption or posibly a hardware problem.
    Also you don't mention what RPM the drive is in your older Mac but Apple puts really Slow 5400RMP drives in the newer models by default. These Slow 5400RPM drives will slow down the whole system. Slow to boot. Slow to load programs and files into those programs and slow to save back to the drive. Also OS X is constantly writing to and reading from the hard drive so a slower drive will again slow down the whole system.
    If it is less then 14 days old you can return it No Questions Asked for a full refund and then maybe try another, different, machine.

  • Why is iMovie 5 x faster than AP3?

    I noticed that slideshow exports in AP3 seemed to take a long time and did a test with iMoive 09. I exported the exact same slideshow of 60 pictures and 1 song from both iMoive and AP3. The iMovie export was 5 times faster than the AP3 export.
    Why is this and can anything be done to speedup AP3 slideshow exports?
    I have a fairly quick Mac, 2 x 2.26 Ghz Quad-Core Intel Xeon, 16 meg of ram, ATI Radeon HD 4870.
    Any thoughts?
    Ken

    You may also want to check your router. Some routers use a priority bandwidth feature that will dedicate more bandwidth to one machine. If for some reason the macbook's download was started earlier than the imac's, then this might be part of the discrepancy. There are a lot of factors to think about when it comes to wifi bandwidth.
    I do agree with the one comment about testing one computer at a time instead of simultaneously.
    Also, when you said you have the movies from the Macbook on your iMac, can you elaborate? If you are using a shared library, then your iMac is going to be using part of your download speed for updates to your shared itunes library, where your Macbook is only going to be uploading the list. If I am incorrect in my understanding of the sharing of iTunes library, someone please let me know.

  • 2011 MBP - why is the optical port faster than the hard drive port?

    In looking at an earlier discussion (April I think) I found a screen grab that showed a 15" MBP with a 6.0Gbp/s hard drive connection and a 3.0Gbp/s optical bay connection. That is image number one below.
    Images two and three are from my brand new MBP 17". Can someone please explain the Link Speed and Negotiated Link Speed differnces? What is the 1.5 Gbp/s about?
    I would appreciate any input guys.
    Thanks!
    Hugh

    Why did Apple change this?
    9-pin FireWire ports use FireWire 800, which is faster than the old 6-pin FireWire 400 ports.
    Is there an adapter I can buy so that I can backup my Macbook?
    Yes.
    (48784)

  • Why Would Crystal Reports read more than it prints

    Hi everyone,
    In my case My record selection is based on the following:
    {FACILITY}=1
    And {payment_code} In ["110","111","112","115","116","117"]
    And {service_code} = "COPAY"
    And {date_of_receipt} in [{?From Date} to {?To Date}]
    And yet for any parameter selection set crystal reports would read - say - 1000 records but prints only 25 records, the question is if the records found according to the selected values from parameter and the record selection formula, is only 25 records, then why would it read 1000 records? what is the reason? and how to avoid it?
    Thanks in advance!

    It may also be what the numbers are telling you is the filter (WHERE) cannot be pushed down to the Server so CR has to filter it client side on the second pass.
    Likely due to the data range, quite often the date format is not compliant to the Server so CR has to "interpret" the date format so it ends up filtering client side.
    Try commenting out that part of the selection formula and see if you get the same number of records read. If you do then you could try using a SQL Express to convert the date fields or possibly using a Command and Command Parameter to set the date range may work for you. This way the DB is definitely getting the date formatted for it's server side use.
    Don

  • SQL Server UPDATE much faster than Oracle -- why??

    The following MS-SQL Server (2000) UPDATE takes 10 seconds, while the same Oracle 10.2.0.1 UPDATE takes over 15 minutes. The data is exactly the same (data set of 97,000 rows) and the cpu/disk/memory configs are similar for both databases. The Oracle tables have indexes and statistics gathered. The Oracle Explain Plan shows that indexes are being used for the SELECT table joins, but not for the UPDATE table.
    Any ideas?
    MS-SQL UPDATE:
    update lib_cust_summary
    set rev_2003 =
    SELECT
    sum(ohd.PRICE_EXTENSION)
    FROM
    oehist_summary ohs
    ,customers c
    ,oehist_detail ohd
    ,lib_cust_summary lcs
    WHERE c.cust_nbr     = lcs.cust_nbr
    and c.cust_nbr = ohs.cust_nbr
    and ohs.INVOICE_NBR = ohd.INVOICE_NBR
    and c.cust_nbr = lcs2.cust_nbr
    and ohs.CUST_NBR NOT LIKE 'XFER%'
    and     ohd.prod_grp not in ('AV','FE')
    AND ohs.INVOICE_DATE between {ts '2003-01-01 00:00:00'} and {ts '2003-12-31 00:00:00'}
    group by c.cust_nbr
    from lib_cust_summary lcs2;
    Oracle UPDATE
    update lib_cust_summary lcs /*+ index(lcs lib_cust_summary_pk) */
    set rev_2003 =
    SELECT /*+ index(lcs lib_cust_summary_pk) */
    sum(ohd.PRICE_EXTENSION)
    FROM
    oehist_summary ohs
    ,customers c
    ,oehist_detail ohd
    WHERE c.cust_nbr = lcs.cust_nbr
    and c.cust_nbr = ohs.cust_nbr
    and ohs.INVOICE_NBR = ohd.INVOICE_NBR
    and ohs.CUST_NBR NOT LIKE 'XFER%'
    and     ohd.prod_grp not in ('AV','FE')
    AND ohs.INVOICE_DATE between '01-JAN-2003' and '31-DEC-2003'
    group by c.cust_nbr
    )

    update lib_cust_summary lcs /*+ index(lcs
    s lib_cust_summary_pk) */
    set rev_2003 =Why are you forcing the optimizer to read a full table by using an index?
    SELECT /*+ index(lcs lib_cust_summary_pk) */What happens if you remove this hint also?
    sum(ohd.PRICE_EXTENSION)
    FROM
    oehist_summary ohs
    ,customers c
    ,oehist_detail ohd
    WHERE c.cust_nbr = lcs.cust_nbr
    and c.cust_nbr = ohs.cust_nbr
    and ohs.INVOICE_NBR = ohd.INVOICE_NBR
    and ohs.CUST_NBR NOT LIKE 'XFER%'
    and     ohd.prod_grp not in ('AV','FE')
    AND ohs.INVOICE_DATE between '01-JAN-2003' and
    d '31-DEC-2003'You should not compare dates to strings
    AND ohs.INVOICE_DATE between to_date('01-JAN-2003','DD-MON-YYYY') and
    to_date('31-DEC-2003','DD-MON-YYY')What are the plans now and after the changes?

  • Would this query be faster in Oracle 10?

    I'm running 2 queries on Oracle 8i:
    Query 1:
    select count(Member_ID) from
    select Member_ID, c2, c3, ... c100 from T1
    union
    select Member_ID, c2, c3, ... c100 from T2
    union
    union
    select Member_ID, c2, c3, ... c100 from T10
    Query2:
    select count(Member_ID) from
    select Member_ID from T1
    union
    select Member_ID from T2
    union
    union
    select Member_ID from T10
    The columns c2 to c100 in Query 1 involve a lot of cast statements. Query 1 takes about 81 seconds and Query 2 takes only about 1/4 second – which is a big difference. These 2 queries are equivalent in terms of what the would return. The reason I am doing Query 1 is because I want the inner query to be a view.
    My question is does anyone know if Oracle 10 would be smart enough to 'forget' about columns c2 to c100 in Query 1 as they are not needed in Query 1? Could the execution times of the 2 queries be similar in Oracle 10?
    Please don't ask me to see the whole queries as they too big and beside the point.

    Denis,
    The answer to your question is that Oracle is smart
    enough to know it simply cannot 'forget' columns c2
    until c100. The fact that in your case both the
    queries return the same result, does not mean the two
    are semantically equivalent.
    In query 1 you use a union operator, meaning it has
    to do a SORT UNIQUE step. In other words: only when
    the rows (member_ID,c2,c3,...,c100) are completely
    identical they will be merged. In query 2 the rows
    only get merged when member_ID's are identical.
    Regards,
    Rob.Thanks for your help! I didn't think about that! In this case the Member_ID is unique (but I didn't say that from the outset). I'm going to try this thing over using Union All instead of Union. If that doesn't help me - I may write back with an update on the post...

  • Why full index scan is faster than full table scan?

    Hi friends,
    In the where clause of a query,if we give a column that contains index on it,then oracle uses index to search data rather than a TABLE ACCESS FULL Operation.
    Why index searching is faster?

    Sometimes it is faster to use index and sometimes it is faster to use full table scan. If your statistics are up to date Oracle is far more likely to get it right. If the query can be satisfied entirely from the index, then an index scan will almost always be faster as there are fewer blocks to read in the index than there would be if the table itself were scanned. However if the query must extract data from the table when that data is not in te index, then the index scan will be faster only if a small percentage of the rows are to be returned. Consiter the case of an index where 40% of the rows are returned. Assume the index values are distributed evenly among the data blocks. Assume 10 rows will fit in each data block thus 4 of the 10 rows will match the condition. Then the average datablock will be fetched 4 times since most of the time adjacent index entries will not be in the same block. The number of single datablock fetches will be about 4 times the number of datablocks. Compare this to a full table scan that does multiblock reads. Far fewer reads are required to read the entire table. Though it depends on the number of rows per block, a general rule is any query returning more than about 10% of a table is faster NOT using an index.

  • Would sorting in UNIX faster than using order by?

    Hi all, recently I was trying to optimize a query that returns a million rows. After using explain plan, I found that the 'order by' (3 columns needed) clause contributes to a significant portion of the cost.
    I wonder if the process can be speed up by doing the query straight without ordering, then use the UNIX sort command (with the help of other commands, such as sed, awk... etc.) to perform the 'order by' instead. Of course you'll need to supply a whole bunch of strange parameters to sort command to make it work as expected.
    But in terms of performance when the result set is huge, which one is faster? 'Order by' or UNIX sort?
    Many thanks.

    What are you sorting by ?
    If it is a date and the output formats it with a 'Mon' component, then I'd bet on Oracle rather than any user written logic converting JAN to 01 etc.
    As for the rest of it, it may depend on the memory allocated to the processes (so PGA setting may improve query performance). A million rows would probably require a disk sort, so the location/speed of that file would impact it too.

  • Why is Chrome so much faster than Safari?

    I'm an Apple and Safari lover, but love goes only so far. I have tried everything with Safari in Yosemite, but it's really in a sad state when it won't even open my Google account page, or my own website. Sadly, I have found so many pages that either load so slowly that I give up, or don't open at all in Safari, yet Chrome always works, and fast. In fact, Firefox and Opera, both of which I have never used regularly are much better than Safari right now.
    Does anyone have a clue as to why Safari has turned into such a poor browser in Yosemite? But please don't suggest I go to safe mode. I only want a decent browser, which I would prefer was Safari and not Chrome, but I do not want to deconstruct or rebuild my Macbook Air OS installation.

    Funny. I'm finding exactly the opposite to be true. I've used Chrome since it came out, and it used to be far superior to Safari. Lately I find it to be very buggy--lots of jittery, time consuming page loads and crashes. Frustrated, I imported all of my stuff to Safari where loading times, scrolling, and navigation are like butter. No crashes and no loading issues on my MB Pro. So far so good.
    I do wonder if it has something to do with the "baggage" I developed on Chrome over the years. I kept things pretty tidy--purging history, cookies, etc-- but it still just kept getting worse.
    I guess Firefox is an option, but one thing is certain: I will not be returning to IE ****.

  • Why is Chromium so much faster than Safari?

    Safari is starting to drag its feet when launching just about any site. Does Chromium have something Safari doesn't?

    HI,
    Empty your Safari cache more often. Command + Option + E
    Why use Google Chrome?
    Carolyn

  • Why is SeaMonkey 2.19 faster than FireFox 22?

    FF22 is soo slow on onepagelove.com and other sites than SM219.
    FF22 even stalls on site with more than 4 tabs open.
    The site is all about web & graphic design ;)
    I am basically using the same add-ons for both browsers.
    Both browsers had a total clean install including registry & hidden folders.

    There shouldn't be any difference in performance between SeaMonkey and Firefox when it comes to web pages because they both use the same rendering engine. It is when you start customizing a browser via extensions or otherwise that things are changing and a lot of extensions can have quite an impact on the performance. Extensions are mostly written in JavaScript code and that is always slower than C++ code that makes the most part of the internal code.

Maybe you are looking for