Re: transactional scanner vs regular non transactional scanner?

We saw in our SSCC testing that simple selects will not necessarily go through the transactional scanner. You won't get unexpected results in MVCC because we only offer read committed data without snapshots, but this is probably not ideal long term. For IDU operations it's a bug. Regards, Sean

Re: Does trafodion caches the results for jdbc queries?

But in theory it could work on Ubuntu with a manual install? On Thu, Jul 30, 2015, 02:49 Amanda Moran wrote: > Hi thre Radu- > > We are currently only able to use the installer to install > Trafodion on RedHat 6 and on Cloudera or Hortonworks. > > Thanks. > > Amanda > > On Wed, Jul 29, 2015 at

Re: Does trafodion caches the results for jdbc queries?

Hi thre Radu- We are currently only able to use the installer to install Trafodion on RedHat 6 and on Cloudera or Hortonworks. Thanks. Amanda On Wed, Jul 29, 2015 at 4:47 PM, Radu Marias wrote: > Yes, I agree it's not a good benchmark environment. Initially I got > problems installing, start

Re: Does trafodion caches the results for jdbc queries?

Yes, I agree it's not a good benchmark environment. Initially I got problems installing, starting it and running any query, now I managed to run some test queries in sandbox with no crash and decent performance at least. The next step is trying to install it on a cluster environment with several no

Re: Does trafodion caches the results for jdbc queries?

If you are using a sandbox to do this testing, then that is not an environment conducive to performance testing. It will probably not give you reliable and consistent results. A sandbox is limited in memory and is subject to limitations of the environment used to build it. The most important

trafodion.org domain/servers

Agree with what Hans said about the sandbox. Changing topics now, related to the downloads page Hans mentioned. As a Apache (incubating) project, we won't be distributing binaries, and so the downloads page needs to go away, or at least migrate to downstream vendors. Likewise, the entire trafodion

Re: Draft for August Trafodion podling status report

Hi Dave, sorry, and thanks for the clarification. I think the simple answer is no. Hans On Wed, Jul 29, 2015 at 3:45 PM, Dave Birdsall wrote: > Hi, > > Thanks, Hans. I think I agree with you. > > My question though was whether it is in fact open source at the moment. If > it is not, we should n

Re: Does trafodion caches the results for jdbc queries?

Hi Radu, In addition, we can help verify the query plan for you to assure the queries can perform reasonably well in your environment, if it works for you. To produce a query plan for query , do the following. log plan.log clear; prepare zzz from : explain zzz; explain options 'f' zzz; --rep

Re: Linuxize trafci and sqlci

+1 On Wed, Jul 29, 2015 at 2:23 PM, Rohit Jain wrote: > Support this wholeheartedly! Find it pretty irritating and I am an old > timer :-) I am sure folks new to Trafodion have a hard time with it from a > usability perspective. > > On Thursday, July 30, 2015, Hans Zeller wrote: > > > Yes, I'v

Re: Draft for August Trafodion podling status report

Hi, Thanks, Hans. I think I agree with you. My question though was whether it is in fact open source at the moment. If it is not, we should not mention it in the monthly report. (Just a small point.) Dave On Wed, Jul 29, 2015 at 3:31 PM, Hans Zeller wrote: > Hi, > > IMHO, the recipe for creat

Re: Draft for August Trafodion podling status report

Hi, IMHO, the recipe for creating the sandbox should be on the wiki and therefore open source. Since the sandbox itself is a binary, the term open-source probably does not apply to it. There could be multiple people and organization offering a sandbox binary, either as a downloadable OVA file or t

RE: Re: Index usage

Further discussing this issue with the experts, it is determined that the index layout for the salted index also contains the salt of the base table only. So my analysis doesn't hold good anymore. We support SALT LIKE TABLE only for the indexes. Selva -Original Message- From: Selva Govin

Re: Draft for August Trafodion podling status report

Hi, Is the Sandbox open source? Dave On Wed, Jul 29, 2015 at 12:24 PM, Qifan Chen wrote: > On community development (the 1st most important issue), I wonder if we can > mention the following. > > 1. Actively help Trafodion users to develop skills and knowhow to use > Trafdoion (the cases of us

RE: transactional scanner vs regular non transactional scanner?

Hi Eric, We do have auto-commit variants for both Order Entry and Debit Credit benchmarks. Regards, Atanu Atanu Mishra Esgyn Corporation Milpitas, CA m: +1.408.219.9765 e: atanu.mis...@esgyn.com -Original Message- From: Eric Owhadi [mailto:eric.owh...@esgyn.com] Sent: Wednesday, J

RE: transactional scanner vs regular non transactional scanner?

Putting back the thread on the dev list, as I accidentally messed up :-) -Original Message- From: Oliver Bucaojit [mailto:oliver.bucao...@esgyn.com] Sent: Wednesday, July 29, 2015 11:57 AM To: Eric Owhadi Cc: Shang-Sheng Tung; Dennis Markt Subject: RE: transactional scanner vs regular non

Re: transactional scanner vs regular non transactional scanner?

Hi Rohit, We should be going through the transactional methods for all IDU and select operations, but we don't currently. This is a mostly invisible issue when using the TRX endpoint coprocessor, but it is a major problem for SSCC at the moment specifically because we do provide snapshots and the

Re: transactional scanner vs regular non transactional scanner?

Oliver, Don't we now force transactional use for all statements? Otherwise, we cannot guarantee consistency if some workloads use transactions and some don't? Maybe that It's not required for selects as yet, since the behavior for selects today would be the same with or without transactions, but w

Re: Linuxize trafci and sqlci

Support this wholeheartedly! Find it pretty irritating and I am an old timer :-) I am sure folks new to Trafodion have a hard time with it from a usability perspective. On Thursday, July 30, 2015, Hans Zeller wrote: > Yes, I've been wanting this (and to a lesser degree wanting to fix this) > fo

Re: Does trafodion caches the results for jdbc queries?

I am not sure that a sandbox is really good to assess performance and whether it will give you reliable consistent performance. The sandbox is limited to the environment it was built on and there are limitations on the amount of memory available to it. The sandbox was built only to kick tires with

Re: Does trafodion caches the results for jdbc queries?

Hi Radu, My suggestion would be to benchmark the system with warmed-up caches, since this will give more meaningful and repeatable numbers. It probably depends on the type of benchmark, but in most cases the benchmark itself will contain multiple statements, with caching going on between them, mak

Re: Linuxize trafci and sqlci

Yes, I've been wanting this (and to a lesser degree wanting to fix this) for years. Time to get rid of FC, but it might not be trivial to do, since there are several pieces of software involved. I think the foundation layer does some manipulation of stdin and/or stdout, which has caused us some tro

RE: transactional scanner vs regular non transactional scanner?

Hi Eric, The transactional scanner is used within a transaction, it will go and access our transactional state on the server-side to gather the changes made since the transaction was begun. The changes are merged with the scan of the underlying HBase table to get the result. When a scan is calle

Linuxize trafci and sqlci

Hi Folks, I've noticed that both trafci and sqlci uses FC (fix command) for manipulation of historical commands. This is kind of weird on a Linux system where utilities tend to use up arrow and cursor movement to modify a historic string for new execution. Would it be a good idea to create

transactional scanner vs regular non transactional scanner?

About https://issues.apache.org/jira/browse/TRAFODION-1420 About implementing the small scanner optimization: Turns out that the small scanner will only work for non-transactional scans with small work. Implementing a transactional small scanner is probably possible but is more work. Before I st

Re: Does trafodion caches the results for jdbc queries?

yes, there exists a control switch to turn off query cache in compiler. Such switch is called query_cache and can be used in a control query default statement (CQD in short) to control how much memory that the query cache can use. When we set it to a value of 0, the cache is disabled. To set the

Re: Draft for August Trafodion podling status report

On community development (the 1st most important issue), I wonder if we can mention the following. 1. Actively help Trafodion users to develop skills and knowhow to use Trafdoion (the cases of user Layla and Radu) 2. Sandbox develop method (on Windows) ready The report looks good. --Qifan On We

Re: Does trafodion caches the results for jdbc queries?

These 2 caching are indeed very good for performance in real environments. But I'm trying to run some internal benchmarks on trafodion and collect the queries duration, in this case the cache generates inconsistency for my tests. I can drop the connection and re-run the query on a new connection fo

Re: Does trafodion caches the results for jdbc queries?

You mentioned also that sometimes you get a slower response time when reconnecting, Radu. As far as I understand (only from what others told me), there could be two reasons. One is that your new connection might be to a different mxosrvr process, which does not have the cache established earlier. T

Re: Draft for August Trafodion podling status report

Thanks for putting this together, it looks good. I believe Rohit will be talking at ApacheCon in Budapest (Ken confirmed), should we include this? On Wed, Jul 29, 2015 at 9:54 AM, Dave Birdsall wrote: > Hi Suresh, > > Looks good. A couple more notes: > > On wiki migration, it's expected that rel

Re: Does trafodion caches the results for jdbc queries?

Another point is that if you can take the advantage of query cache by preparing the query once and executing it multiple times, further reducing the compilation time per SQL invocation. ​

Trafodion Daily Test Result - 2015-07-29_08-30-00

Daily Automated Testing Jenkins Job: https://jenkins02.trafodion.org/job/Check-Daily/42/ Archived Logs: http://logs.trafodion.org/Daily/2015-07-29_08-30-00 Changes since previous daily build: No changes Test Job Results: FAILURE core-regress-catman1-cm5.3 (9 min 31 sec) FAILURE core-regress-

Re: Does trafodion caches the results for jdbc queries?

Hi, I suspect this is normal behavior. The Trafodion SQL compiler caches query plans and metadata. So if you execute the same query multiple times in a session, most of the compilation phases are bypassed after the first time. Also, HBase itself caches data in its RegionServers. So, the first ti

Does trafodion caches the results for jdbc queries?

I'm using daily build *trafodion-20150726_0830.tar.gz* in *sandbox 1.1* with several sql clients like SQuirreL, DbVisualizer and 0xDBE from Intellij. Noticed that in some cases when I execute a query for the first time it takes longer than the next subsequent executions. It's not the first query

Re: Draft for August Trafodion podling status report

Hi Suresh, Looks good. A couple more notes: On wiki migration, it's expected that relevant content from the pre-Apache Trafodion wiki will be contributed to the project shortly and that content integrated into the new Apache wiki and web site over the next month or so. On licensing issues, JIRA

RE: Re: Index usage

Here is my prognosis of the problem, but I may be wrong. For the index with no salting keys, the index layout is such that given an index key value, it is possible to get to the base table row easily because base table salt column is part of the index. For the index with salting, the index layou

Draft for August Trafodion podling status report

Hi, Please give your feedback on this status report. I copied the July version and made some edits. Thank you Suresh Trafodion Trafodion is a webscale SQL-on-Hadoop solution enabling transactional or operational workloads on Hadoop. Trafodion has been incubating since 2015-05-24. Four most

RE: Re: Index usage

I think there is some issue. I have attached the invoke of the table. When the query involves index t12i1 it chooses the index plan even for the queries with the predicates. When the query involves index t12i2, the index is not chosen unless it is full index scan. The index t12i2 is salted while t

Re: Re: Index usage

hi Ming, yes, in that case the index only scan plan should be chosen. Regards, --Qifan On Wed, Jul 29, 2015 at 9:29 AM, wrote: > hi, QiFan, > If Layla only select the index column , not * , it should choose the index > scan, is that right? I can do a test myself ... > thanks,Ming > - Origi

Re: Re: Index usage

hi, QiFan, If Layla only select the index column , not * , it should choose the index scan, is that right? I can do a test myself ... thanks,Ming - Original Message - From: Qifan Chen To: "Martin, Layla (HP DualStudy)" Cc: "dev@trafodion.incubator.apache.org" Subject: Re: Index usage Da

Re: Index usage

Yes, that will be a very good exercise. did you quit the sqlci session? The shape is only useful within a session, or when it is being turned off. You also need to update stats after each iteration of table load, for the column used in the predicate. On Wed, Jul 29, 2015 at 8:50 AM, Martin, Lay

RE: Index usage

You meen Frequency in terms of how often a value occurs? That’s exactly what I want to do. I have a function which generates random numbers (in a specific range). But even with random numbers between 0 and 49,999 and a total of 50,000 rows, the index is used. I now assume that something is set

Re: Index usage

You probably can change the frequency of a value in the column (f=1,2,5,10, ...) to find out the break-even point. The CQD provided will force a full scan (without the shape). On Wed, Jul 29, 2015 at 8:32 AM, Martin, Layla (HP DualStudy) < layla.mar...@hp.com> wrote: > Hi Qifan, > > > > But thi

RE: Index usage

Hi Qifan, But this doesn’t just undo the control query shape from before, does it? I would like to see at which point the optimizer decides to use indexes rather than a full table scan. Best regards Layla Layla Martin HP DualStudy - Telefon +49 7031 4504682 - layla.mar...@hp.com

Re: Index usage

Yes, apply the following CQD to completely turn off index. CQD HIDE_INDEXES 'ALL'; On Wed, Jul 29, 2015 at 6:05 AM, Martin, Layla (HP DualStudy) < layla.mar...@hp.com> wrote: > Hi, > > Thanks for your help. > It now uses the index! Can I switch the index usage of again? I think it > still uses

RE: Index usage

Hi, Thanks for your help. It now uses the index! Can I switch the index usage of again? I think it still uses the index, even if it doesn’t make sense … (if I’m using 50,000 different values for 50,000 rows …). I’ve tried “CONTROL QUERY SHARE OFF, but it didn’t do the thing … It still uses the