kdb+ and q
Q+A on all things kdb+ database and the q language
Showing results for 
Search instead for 
Did you mean: 

Forum Posts

slow performance of win32 version of KDB

Hi, I like to test some simple thing on windows version of 32bit KDB, I inputted a table with 99722 row, with date, sym and other 15 columns of numeric data. However, if I just take a simple average of that 15 column, it take more than 3 minutes to r...

carfield1 by New Contributor
  • 7 replies
  • 0 kudos

Sparse Arrays

Can anyone point to discussion or work on sparse arrays in k/q? This must have come up somewhere, but I don't see anything in the open resources.Cheers

i3text by New Contributor
  • 1 replies
  • 0 kudos

[TorQ] .gw.syncexec[] select servers by procname or attribute

How can I make .gw.syncexec[] work like .servers.getservers[]? I have separate hdbs for each stock market. How can I query for prc_stocks but not hk_stocks? .servers.getservers[] supports selecting servers by procname or attributes. thxproctype,procn...

LamHinYan by New Contributor
  • 4 replies
  • 0 kudos

functional select update

I have a table tab, and I have a list of columns allCols I want to calculate their correlation against anther column rr. Below doesn't seem to work: ?[tab; enlist 1b; 0b;allCols!{ x, cor, y }'[4#`rr; allCols] ];but if I just do a simple average, ?[ta...

KdbNoob by New Contributor
  • 2 replies
  • 0 kudos

When is dropping down to K necessary?

As I understand it, q is to some extent a subset of K4. Since resources for q are rather meager, I'm also drawing on stuff I can find related to K3 and J. When translating code from these languages to q, I'd like to know for what kinds of tasks I wou...

i3text by New Contributor
  • 5 replies
  • 0 kudos