f



Re: Dictionary problem with Parallel processing

> From: vijayakumar
>
> I am facing strange problem, my environment is Sun Solaris, SAS 9.1.3.
>
> Situation:
>
> I want to create summary report for the libraries about no of obs,
> when was the data set created and library name etc.
>
> One main job started to run which in turn creates around 30 to 50
> parallel jobs in which I have to create summary report specifically
> for two libraries for all those parallel jobs.
>
>  I have used dictionary. tables and dictionary. members to get the
> report created but every now and then I am getting an error saying the
> "Lock was not available for the dataset and held by some process".
>
> Solution in Mind:
>
> Is this can be avoid if I use sashelp.vtable and sashelp.vslib?

[sound of alarm bells going off here]

> Since
> as far as I know the dictionaries will open the libraries to get the
> above said key things but these sashelp views can be other way around
> to avoid these issue?
>
> I need your expert's advice on this to resolve this issue.

summary: totally avoid reading sashelp views in production code

reason: they are derivatives of SQL dictionaries
and are an order of magnitude slower
i.e.: takes 10 times as long
than Contents which reads information for one specific libref.

see
http://www.sascommunity.org/wiki/List_Processing_Development#Speed_Consi
derations

instead use
Proc Contents data = Libref._all_
              out  = Work.ListTables;

that is a list of variables
you have to reduce that to a list of tables.

Ron Fehd  the module/routine/subroutine maven  CDC Atlanta GA USA RJF2
at cdc dot gov
0
rjf2 (3342)
12/22/2008 10:27:13 PM
comp.soft-sys.sas 142828 articles. 1 followers. Post Follow

2 Replies
666 Views

Similar Articles

[PageSpeed] 11

Thanks Ian and Ronald will try to use exist with proc contents to
produce the report.

On Dec 22, 11:27=A0pm, r...@CDC.GOV ("Fehd, Ronald J. (CDC/CCHIS/
NCPHI)") wrote:
> > From: vijayakumar
>
> > I am facing strange problem, my environment is Sun Solaris, SAS 9.1.3.
>
> > Situation:
>
> > I want to create summary report for the libraries about no of obs,
> > when was the data set created and library name etc.
>
> > One main job started to run which in turn creates around 30 to 50
> > parallel jobs in which I have to create summary report specifically
> > for two libraries for all those parallel jobs.
>
> > =A0I have used dictionary. tables and dictionary. members to get the
> > report created but every now and then I am getting an error saying the
> > "Lock was not available for the dataset and held by some process".
>
> > Solution in Mind:
>
> > Is this can be avoid if I use sashelp.vtable and sashelp.vslib?
>
> [sound of alarm bells going off here]
>
> > Since
> > as far as I know the dictionaries will open the libraries to get the
> > above said key things but these sashelp views can be other way around
> > to avoid these issue?
>
> > I need your expert's advice on this to resolve this issue.
>
> summary: totally avoid reading sashelp views in production code
>
> reason: they are derivatives of SQL dictionaries
> and are an order of magnitude slower
> i.e.: takes 10 times as long
> than Contents which reads information for one specific libref.
>
> seehttp://www.sascommunity.org/wiki/List_Processing_Development#Speed_Con=
si
> derations
>
> instead use
> Proc Contents data =3D Libref._all_
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 out =A0=3D Work.ListTables;
>
> that is a list of variables
> you have to reduce that to a list of tables.
>
> Ron Fehd =A0the module/routine/subroutine maven =A0CDC Atlanta GA USA RJF=
2
> at cdc dot gov

0
12/22/2008 11:40:23 PM
On Dec 22, 11:27=A0pm, r...@CDC.GOV ("Fehd, Ronald J. (CDC/CCHIS/
NCPHI)") wrote:
> > From: vijayakumar
>
> > I am facing strange problem, my environment is Sun Solaris, SAS 9.1.3.
>
> > Situation:
>
> > I want to create summary report for the libraries about no of obs,
> > when was the data set created and library name etc.
>
> > One main job started to run which in turn creates around 30 to 50
> > parallel jobs in which I have to create summary report specifically
> > for two libraries for all those parallel jobs.
>
> > =A0I have used dictionary. tables and dictionary. members to get the
> > report created but every now and then I am getting an error saying the
> > "Lock was not available for the dataset and held by some process".
>
> > Solution in Mind:
>
> > Is this can be avoid if I use sashelp.vtable and sashelp.vslib?
>
> [sound of alarm bells going off here]
>
> > Since
> > as far as I know the dictionaries will open the libraries to get the
> > above said key things but these sashelp views can be other way around
> > to avoid these issue?
>
> > I need your expert's advice on this to resolve this issue.
>
> summary: totally avoid reading sashelp views in production code
>
> reason: they are derivatives of SQL dictionaries
> and are an order of magnitude slower
> i.e.: takes 10 times as long
> than Contents which reads information for one specific libref.
>
> seehttp://www.sascommunity.org/wiki/List_Processing_Development#Speed_Con=
si
> derations
>
> instead use
> Proc Contents data =3D Libref._all_
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 out =A0=3D Work.ListTables;
>
> that is a list of variables
> you have to reduce that to a list of tables.
>
> Ron Fehd =A0the module/routine/subroutine maven =A0CDC Atlanta GA USA RJF=
2
> at cdc dot gov

Thanks Ian and Ronald will try with exist and proc contents tomorrow
and let u guys know :-)
0
12/22/2008 11:42:34 PM
Reply:

Similar Artilces:

Re: Dictionary problem with Parallel processing #2
Vijay, I doubt if a view of a DICTIONARY file can do any better than the DICTIONARY file since the view executes the underlying DICTIONARY file. There are probably more sophisticated methods involving the underlying operating system to know when each job is finished. However, you might use the EXIST function to loop through the required data sets until all data sets exist. Ian Whitlock ================ vijayakumar <vijayk.kannan@GMAIL.COM> wrote: Dear Folks, I am facing strange problem, my environment is Sun Solaris, SAS 9.1.3. Situation: I want to create summary report for the libraries about no of obs, when was the data set created and library name etc. One main job started to run which in turn creates around 30 to 50 parallel jobs in which I have to create summary report specifically for two libraries for all those parallel jobs. I have used dictionary. tables and dictionary. members to get the report created but every now and then I am getting an error saying the ?Lock was not available for the dataset and held by some process?. Solution in Mind: Is this can be avoid if I use sashelp.vtable and sashelp.vslib? Since as far as I know the dictionaries will open the libraries to get the above said key things but these sashelp views can be other way around to avoid these issue? I need your expert?s advice on this to resolve this issue. Regards, Vijay ...

Dictionary problem with Parallel processing
Dear Folks, I am facing strange problem, my environment is Sun Solaris, SAS 9.1.3. Situation: I want to create summary report for the libraries about no of obs, when was the data set created and library name etc. One main job started to run which in turn creates around 30 to 50 parallel jobs in which I have to create summary report specifically for two libraries for all those parallel jobs. I have used dictionary. tables and dictionary. members to get the report created but every now and then I am getting an error saying the =93Lock was not available for the dataset and held by some process=94. Solution in Mind: Is this can be avoid if I use sashelp.vtable and sashelp.vslib? Since as far as I know the dictionaries will open the libraries to get the above said key things but these sashelp views can be other way around to avoid these issue? I need your expert=92s advice on this to resolve this issue. Regards, Vijay ...

Re: SAS Parallel processing
Ram, if (for obvious reasons since you are rsubmitting code) you can not give us ready-to-run code then at least show us the error messages you get. Kind regards Robert > -----Urspruengliche Nachricht----- > Von Ram > Gesendet: Dienstag, 31. Juli 2007 02:51 > > Hi, > > I am relatively new to SAS parallel processing scene and trying to > work on the following piece of code using SAS 9 on a PC. The remote > server has 4 processors and this is a UNIX based server. On my PC > windows based SAS, I execute the following code after connecting to > the remote serve...

Re: Unusual SAS processing Problem
What I am trying to understand is if the same model with same specification same class variables and same dataset ran well last week then sudenly what happened that is is taking 1 hr 51 mins to process ? Regards Sudip On Apr 2, 4:31 pm, sudip.memp...@GMAIL.COM (sudip chatterjee) wrote: > What I am trying to understand is if the same model with same > specification same class variables and same dataset ran well last week > then sudenly what happened that is is taking 1 hr 51 mins to process > ? > > Regards > Sudip Well, it sure would have been nice if you had provide...

Re: SAS time problem and formula problem
Hello Vijay I can only answer your first question since the formula was not readable for me. (SAS-L removes attachments, incl the picture showing your formula) Consider this data step ... data in; format hr hhmm5.; hr = '45:00't; /* <-- 45 hours and 0 minutes -- */ hrn = hr/(60*60); /* <-- number of decimal hours from 0 minutes. Using the hour() function will return 21 which is incorrect -- */ loghrn = log(hrn); /* <-- gives you the 3.807 -- */ run; I think the issue is the value of hr. You see 45:00 which is the representation of the SAS hhmm5. format. The SAS stored value is a numeric 162000, the number of seconds since the time 0 or 00:00. When you use the formula log(hr) you are doing log(162000) and not log(45) as you may want. HTH Magnus On Mon, 3 Nov 2003 11:51:14 +0000, Vijay Reddi <Vijay.Reddi@SSL- INTERNATIONAL.COM> wrote: >Hi All, > >I have two problems, and I really hope you can help me. > >The first problem is that I want to take the Log of a time variable (format >is hhmm6.), but the answer returned is wrong. I have used the following >code: > >/* p is a time variable with format hh:mm5.*/ > >data 'C:\Vijay Data\Research project\Project\Data\craw4'; > set craw3_1; > ln_aa=log(p); > format ln_aa hhmm5.; >run; > >For example if p=45:00 (hh:mm) then ln_aa equals 00:00 (hh:...

Re: SAS time problem and formula problem #2
Please re-post your second question and present the formula in-line, as part of the message. SAS-L does not allow attachments. Also make it a separate posting, with its own subject. When a person has two or more unrelated questions, there is no reason to bundle them in one posting. It's not as if one has to pay for postage. Making a separate posting for each question, with appropriate subjects, makes it much easier to carry on and follow the discussions. On Mon, 3 Nov 2003 11:51:14 +0000, Vijay Reddi <Vijay.Reddi@SSL- INTERNATIONAL.COM> wrote: >Hi All, > >I have two probl...

Re: problem when installing SAS. windows x64 does not support SAS?
2 things you can do: 1) switch to 32-bit vista (business / ultimate) 2) or wait for a new version of SAS supporting 64-bit windows On Wed, May 14, 2008 at 10:54 AM, <elodie.gillain@gmail.com> wrote: > Hi everyone > > I am trying to install SAS on a Vista machine. > > The setup wizard says that Windows x64 does not support the SAS > version I am trying to install. > > What can I do? > > I greatly appreciate your help. > -- =============================== WenSui Liu ChoicePoint Precision Marketing Phone: 678-893-9457 Email : wensui.liu@choicepoint.com ...

Re: problem when installing SAS. windows x64 does not support SAS? #2
Elodie, Take a look at: http://support.sas.com/kb/16/568.html HTH, Art --------- On Wed, 14 May 2008 07:54:57 -0700, elodie.gillain@GMAIL.COM wrote: >Hi everyone > >I am trying to install SAS on a Vista machine. > >The setup wizard says that Windows x64 does not support the SAS >version I am trying to install. > >What can I do? > >I greatly appreciate your help. ...

Re: SAS 9.2 Update on problem re Help files
On Mon, 25 Aug 2008 09:25:35 -0400, "Peter Flom" <peterflomconsulting@MINDSPRING.COM> said: > Hi again > > A week ago, I asked this list about a problem with 9.2: When you click > "help" it runs an index, but it doesn't save the index when you quite > SAS, so, each time you quit SAS, you lose the index. > > Replies received here did not work, so I wrote to SAS. > > They said, essentially, that it's a known bug, and they are working on > it. So you're responsible for the item in today's Tech Support Notes. Congratulations...

Re: Transfering datasets to Unix /SAS from PC/SAS
Md Alam: Most often failures of ftp to transfer a valid SAS dataset from Unix to a PC platform result from transferring the file using a MODE other than BINARY. Try a binary mode transfer. Sig -----Original Message----- From: owner-sas-l@listserv.uga.edu [mailto:owner-sas-l@listserv.uga.edu] On Behalf Of Md Alam Sent: Friday, November 04, 2005 11:34 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Transfering datasets to Unix /SAS from PC/SAS - Still problem Thanks for all your help. I tried using FTP to tranfer datasets from unix to PC. But when I am trying to open the datasets in PC after being tra...

Re: RE : Re: copulas and SAS
Adel, Think Joke! But, more seriously, it did prompt me to look up the origins of the copula as a functional. Joe On 9/27/06, adel F. <adel_tangi@yahoo.fr> wrote: > > Hi, > I do not know how this is related to my orginal question. > > My understanding of copula is, it is a joint distribution with uniform > margins, the problem how to obtain a copula which fits better a data? > > If there is any suggestion from members of SAS list, about examples for > copulae with SAS, I will appreciate that > > Thanks a lot > > Adel > Joe Whitehurst <joew...

Parallel Processing in SAS
I've heard that there are opportunties to do some parallel processing if your server has multiple CPUs. We have 4 CPUs running on windows 2003 SP1. Anyone have any insight or articles about Parallel processing? I heard only a few procs can actually utilize and take advantage of parallel processing, but wasn't sure. Any insight is appreciated! Jer wrote: > I've heard that there are opportunties to do some parallel processing if > your server has multiple CPUs. I think there are also opportunities for parallel processing with single CPU machines that support threading; on...

SAS Parallel processing
Hi, I am relatively new to SAS parallel processing scene and trying to work on the following piece of code using SAS 9 on a PC. The remote server has 4 processors and this is a UNIX based server. On my PC windows based SAS, I execute the following code after connecting to the remote server. options autosignon=yes ; rsubmit; /* Begin simulation */ options symbolgen ls=96; options sascmd='!sascmd -nosyntaxcheck' autosignon=yes cpucount=ACTUAL; %macro StatsModel(startLoop=, endLoop=, inputfn=,nsims=); filename params "/data/&inputfn"; data parameters; infile par...

SAS/Warehouse Administrator Problem?? or SAS/ACCESS problem??
hello, I've built a process in SAS/Warehouse Administrator, where i'm trying to extract distinct cities from the 3 tables(source schema) in Oracle into a new table (target schema) called CITY and again into UniqueCity (target schema) bcos of some redundancy. The process looks like this. Unique City | | mapping | | CITY ...

Re: Re: Problems with my
Hi David, > > > Even Fortran has a LOGICAL data type! Yes it does. And modern versions (> F90) allow users to define their own types in terms of fundamental data types, improving readability & code even more. However, the usual way that this sort of thing is coded in Fortran (usually F77), is as a series of cryptic arguments to subroutines (eg LinPack); logical variables are rarely used, but I'm happy to be contradicted. My point was that while this can be done in Mathematica, a better way might be to use options & logical variables such as ...

Re: Problem with dictionary.tables
Antje, Try following code see if it works: proc sql noprint; select "perm."||memname into :dsn separated by " " from dictionary.tables where libname = "PERM"; quit; data result; set perm.&dsn; run; YU On 6/26/06, antje.schuele@googlemail.com <antje.schuele@googlemail.com> wrote: > > Dear Newsgroup, > > I'd like to set a few tables together. That's why I read out my needed > tables with the help of dictionary.tables: > > Here's my code: > > proc sql noprint; > select memname into...

Re: Problem with SAS macro
Mandan, I think this is closer to what you want. %macro empdata_mod1(show); %local loop msg ; %window employee group=base #5 @4 "Date of Access: &sysday, &sysdate.." #8 @10 "Employee Data Base" #17 @29 "Press" @35 "Enter" attr=underline @41 "to continue." group=month #9 @10 "Data for" @19 mon 3 protect=yes #11 @4 "Or select another month (blank to quit):" @45 mon 3...

Re: problem on SAS/graph
On Mon, 16 May 2005 14:51:08 -0400, Ben <benpub7@YAHOO.COM> wrote: >Hi friends, >I want to produce pdf file using ods pdf='somefile.pdf' for its PROCLABEL >option , > >but it seemed I could not get rid of the default sasgraph.pdf, even I set >up goptions gsfname="somefile.pdf". it's cost me several hours without >progress. > >Thanks and strange log; ERROR: The sasPrtC device does not support GACCESS or GSFNAME. ERROR: Driver SASGDDMX will not load. Ben Sounds like you need a filename statement. Your error message may go away if you c...

Re: Problem Notes for SAS
Chang, I have occasionally thought about subscribing to this particular list, but have never gotten around to it. One reason is potential volume -- about how many messages does this subscription generate in an average day or week? Mike Rhoads Westat RhoadsM1@Westat.com -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Chang Chung Sent: Monday, June 09, 2008 10:24 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Problem Notes for SAS hi, just to let you know who doesn't already, I find the problem notes for sas feed is a good one to subscribe: http...

Re: SAS Date problem
Make that the YYMMDD informat. In other words: date = input(put(date,8.),yymmdd8.); On Sat, 1 May 2004 19:52:36 -0400, Howard Schreier <Howard_Schreier@ITA.DOC.GOV> wrote: >Try > > date = input(put(date,8.),mmddyy8.); > >On Sat, 1 May 2004 20:04:39 GMT, Giulio Belrango ><giuliobelrango@ROGERS.COM> wrote: > >>I work in an IBM Mainframe OS390 environment. >>I'm reading a 4 byte field that contains a date in binary format ie PIB4. >or >>PIC 9(8) comp, the date is 020041201 ie CCYYMMDD. I'm trying to convert >this >>to...

Re: Sas macro problem
On Sun, 30 Nov 2008 13:24:09 -0800, hcurington@GMAIL.COM wrote: >Hello, > >I am new to sas programming, and I am having trouble executing this >sas macro: > >%macro getbinbf(indat,bin_number,target,varin); >/* Produce Logit Plot for Variable without Transformation */ >proc rank data=&&indat groups=&bin_number out=out; >var &varin; >ranks bin; >run; >proc means data=out noprint nway; >class bin; >var &target &varin; >output out=bins sum(&target)=&target mean(&varin)=&varin; >run; >datbins; >set bins; >elogit=log((&target+(sqrt(_freq_)/2))/(_freq_-&target+(sqrt(_freq_)/ >2))); >run; >proc datasets lib=work; >delete out; >run; >quit; >%mend; > >We are supposed to understand how the above code can be used to >produce a logit plot for the binary response variable. I understand >that positional parameters are being used here, but how do I run this >macro with my dataset Assign5.hmeq? Does bin_number refer to using >the actual number 1 or 0? The target variable in the sas dataset is >"BAD." I am not sure what the 'varin' variable is referring to, maybe >an independent variable? I tried using this call statement right >after the macro: > >%getbinbf (Assign5.hmeq, BAD, BAD, BAD) >but I get errors saying 'option or parameter is not recognized' and >will be ignored and 'expecting an integer...

Re: Data processing problem
On Mon, 8 Sep 2008 09:56:14 -0700, jn mao <jn_mao@YAHOO.COM> wrote: >Hi SASLs,I have a large dataset. I need to calculate the duration by each two connected dates bydrug for each patient, then divided by NT. Only select the 2nd data of NT to divide duration.?My data:ID date drugname NTaa 1/2/08 xx 10aa 2/1/08 xx 14aa 2/3/08 yy 11aa 3/1/08 xx 15aa 3/2/08 yy 16The data I want:ID duration drugname NT NT_duraa 31 xx 14 14/31aa 28 xx 15 15/28aa 29 yy 16 16/29?The duration is the difference of two connented dates for the sam...

Re: SAS Problem #3
On Fri, 30 Apr 2004 05:13:05 +1200, Dean Edwards <dean@DREAMNET.CO.NZ> wrote: >"Thank you to all the people with your helpful replies to my earlier SAS problem. However "SAS Problem" is an extremely vague and uninformative Subject header. > > > >My first problem was to get the target points = (number of households for each area). > >and calculate the number of households for 26 areas that earn say $12,824 or less for example. This has been solved. > >. > >. I have resolved the statistical problem and it is just a coding issue problem now...

Re: Dictionary problem #2
Hi, Antonmuhin, Alex and Batista, as I can use only Python 2.1 dict() is not allowed I guess. The .copy() trick works on the simple example I gave. Thanks a lot for your hints. But my real problem is more complex (nested dicts) like this: p_settings = {} for item in items : p = myList # a list for i in range(10) : for setting in p[i] : p_settings['anotherString']['aString'] = myString object_dict = {'id':'anotherString', 'settings':p_settings} object_perms_list.append(object_dict) here the copy()-trick does not work. How ...

Web resources about - Re: Dictionary problem with Parallel processing - comp.soft-sys.sas

Dictionary - Wikipedia, the free encyclopedia
According to Nielsen (2008) a dictionary may be regarded as a lexicographical product that is characterised by three significant features: (1) ...

Dictionary.com Dictionary & Thesaurus goes 7.0 with interface improvements
In September, Dictionary.com Dictionary & Thesaurus was updated to version 6.0 with a significant redesign that introduced a fresh look and feel ...

Price Drop: Textkraft with German Dictionary
Textkraft with German Dictionary 3.1.3 Device: iOS iPad Only Category: Productivity Price: Free, Version: 3.1.3 ( iTunes ) Description: Das ...

'Rabid Feminist' and Other Sexist Sentences Should Be Excluded From the Oxford Dictionary ASAP
It's a source that's meant to inform and educate but the Oxford English Dictionary recently came under fire for being sexist. Michael Oman-Reagan, ...

'Mx.' instead of 'Mr.' or 'Mrs.?' It's in the dictionary now
Among the words officially added to dictionary.com this week is &quot;Mx.,&quot; pronounced &quot;mix&quot; and defined as &quot;a title of respect ...

'Identity' is Dictionary.com's word of the year - CNN.com
Race, sexuality and gender are the live-wire topics that inspired Dictionary.com to pick "identity" as its word of the year for 2015.

Jimmy Fallon, Nathan Lane play 'Kid Dictionary' on 'The Tonight Show' - EW.com
Jimmy Fallon, Nathan Lane play 'Kid Dictionary' on 'The Tonight Show' - EW.

The Original Hacker's Dictionary For Small Business Owners
A cybersecurity book and glossary for newbies and small business owners.

But That Isn't A Word: Oxford Dictionary Chooses The Face With Tears Of Joy Emoji As Word Of The Year ...
Because even dictionary companies want to feel edgy from time to time, Oxford Dictionaries has announced the 'face with tears of joy' emoji is ...

An Online Dictionary Plans to Catalog a Million Emerging English Words
With viral memes and hashtags sweeping the internet on the daily, language is evolving faster than conventional dictionaries can keep up. You ...

Resources last updated: 3/16/2016 2:29:46 PM