Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.databases.ms-sqlserver > #1505

Re: doubt

Show all headers | View raw

 (rja.carnegie@gmail.com) writes:
> Users, and I speak as one, can be trained to accept 
> appalling standards of service.  We've got an enterprise 
> reporting system that /usually/ returns an error instead 
> of a report.

Reminds of the ERP system they have my client. Being a consultant, I don't
have to use it. But I recall that when I was an employee I had to attest
an invoice. It was the sleaziest interface I've ever seen. But it came from
a vendor which in not in scope for this newsgroup.
 
> I suppose the technical answer to the question is in 
> terms of the activity level on the server - where a lot 
> less than 100% still counts as under stress - or on
> the duration between starting an operation and 
> completing it.  But what can you do ab?ut it? I suppose
> stop running your personal Bitcoin software on 
> the server...

But it may be running a reindexing in the middle of the night or a brutal
ETL operation, and you really want the machine to max out. And as long as
the job is done in the morning, no one cares.

Subbu's question was short and meaningless. To be useful, you first need to
define "slow". That is, you need a baseline. Once you have a baseline and 
you know that a key procedure for adding orders takes, say, 100 ms, you 
could set up an alert when it suddenly takes one second or more. Of if
there is a queue you process, you can set up an alert when the queue exceeds 
a certain size. Or an alert if your ETL job is not finished by 05:00. Etc.

But for the broad and sweeping question that Subbu asked, there is really 
only one answer: someone has to complain about the performance. 



-- 
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx

Back to comp.databases.ms-sqlserver | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

doubt SUBBU N <nsubbu1986@gmail.com> - 2013-05-28 23:06 -0700
  Re: doubt Erland Sommarskog <esquel@sommarskog.se> - 2013-05-29 07:32 +0000
  Re: doubt SUBBU N <nsubbu1986@gmail.com> - 2013-05-30 02:55 -0700
    Re: doubt Erland Sommarskog <esquel@sommarskog.se> - 2013-05-30 23:08 +0200
      Re: doubt rja.carnegie@gmail.com - 2013-05-30 15:10 -0700
        Re: doubt Erland Sommarskog <esquel@sommarskog.se> - 2013-05-31 07:45 +0000
  Re: doubt Ross Presser <rpresser@gmail.com> - 2013-06-04 10:18 -0700
    Re: doubt Erland Sommarskog <esquel@sommarskog.se> - 2013-06-04 22:24 +0200
      Re: doubt Ross Presser <rpresser@gmail.com> - 2013-06-04 13:34 -0700

csiph-web