Project

General

Profile

Slug #191

Slow lex gbasis

Added by John Abbott almost 12 years ago. Updated almost 11 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Category:
Portability
Target version:
Start date:
19 Jun 2012
Due date:
% Done:

100%

Estimated time:
Spent time:

Description

The following example comes from an exercise I gave during Algebra Computazionale

Use QQ[a,b,x,y,z], Lex;
I := Ideal(x-a^3-b^3-3, y-a^5-b^5-5, z-a^7-b^7-7);
t0 := CpuTime();
RGB := ReducedGBasis(I);
PrintLn "Tempo impiegato per calcolare RGB: ", DecimalStr(CpuTime()-t0);

On my computer, with the latest development version of CoCoA-5 and CoCoALib, the time taken is about 850s. On a student's computer (MS Windows, I believe), using the version of CoCoA-5 which was available 1-2 months ago on the website, the same computation takes about 60s.

I am mentioning this because I think we have not recently changed anything in the G-basis computation code...

I've put this in CoCoALib because I believe that's where the problem lies (even though it came to light when using CoCoA-5).


Related issues

Related to CoCoA-5 - Feature #19: How to get the version number of CoCoA-5?Closed2011-11-04

History

#1 Updated by Anna Maria Bigatti almost 12 years ago

  • Status changed from New to In Progress

John Abbott wrote:

The following example comes from an exercise I gave during Algebra Computazionale
[...]
On my computer, with the latest development version of CoCoA-5 and CoCoALib, the time taken is about 850s. On a student's computer (MS Windows, I believe), using the version of CoCoA-5 which was available 1-2 months ago on the website, the same computation takes about 60s.

Also my current version takes about 60s.
I guess you have some experimental part in your working version....
I think I really should add some extra info to VersionInfo ;-)

#2 Updated by John Abbott almost 12 years ago

  • Status changed from In Progress to Closed
  • % Done changed from 0 to 100

I did make veryclean; sync with CVS, and make.
Now mine takes about a minute too... very odd!

No idea why it was so slow before :-(

#3 Updated by Anna Maria Bigatti almost 12 years ago

John Abbott wrote:

No idea why it was so slow before :-(

I had a similar problem to what you described yesteray: old cocoa running while compiling the new one. Maybe the old one is not deleted until it is linking time.

#4 Updated by Anna Maria Bigatti over 11 years ago

  • Project changed from CoCoA to CoCoALib

#5 Updated by Anna Maria Bigatti over 11 years ago

  • Category set to Various

#6 Updated by Anna Maria Bigatti over 11 years ago

  • Project changed from CoCoALib to CoCoA-5
  • Category deleted (Various)

#7 Updated by Anna Maria Bigatti over 11 years ago

  • Target version set to CoCoA-5.0.3

#8 Updated by Anna Maria Bigatti almost 11 years ago

  • Category set to Portability
  • Assignee set to John Abbott

Also available in: Atom PDF