× The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.



Yeah, Pete - was wondering what changed it.

I can't find a direct statement of the benefit for Java. Do you already have multiple processors? Then Java will use them. But it's going to multi-thread, no matter what.

I'm interested in the document or site that says it's good to turn this on for Java - cuz I've not heard that.

I do have a link that discusses this on POWER7, which has 4 hardware threads. It says you do not get 4x gain from it - at most, maybe 2x. Older processors since the Northstar of around 2002 have only 2 hardware threads.

The V6R1 article on this system value is at http://tinyurl.com/ykftc7p and has another link to some advice on when to set it to what. The link has a comment that turning it on lets the entire machine do more work, leaving it off lets single tasks run faster.

http://tinyurl.com/yzrrs7y is about simultaneous Multi-threading on POWER7 Processors

HTH
Vern*
*
Pete Hall wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Vern Hamberg wrote:
Since the default is to have it turned on, I'd say it's best to leave it
that way. That is the gist of an InfoCenter article on when to use this
feature. It appears that things should not go more slowly when this is
on, but you won't gain double the speed from the 2 paths in a CPU in
earlier processor since NorthStar. POWER7 seems now to support 4
hardware threads on a core.

Well, see, the thing is, it's currently set to '0', which appears to
mean that threads will be executed sequentially instead of concurrently.
I don't know how it got set that way. Presumably that's what came over
from the previous hardware. I'm looking to turn the support on so Java
threads will get executed more efficiently. As I said, the setting is
inherited from the controlling partition on the hardware we are running,
so I can't test the change on the dev LPAR, and an IPL of the complete
box is necessary to affect the change, meaning a service outage. Hence
the questions: Is this a good thing to do? Will any bad things happen?

- --
Pete Hall
pete@xxxxxxxxxxxxxx
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAkuMAq8ACgkQXczQcKdXKg7+qQCdH79KlpTmeCBpQDX0KA6+Dpiu
auAAoIgtZ2XTH+Tk6bnJwrMqEMi3nPPo
=Z9Ro
-----END PGP SIGNATURE-----

As an Amazon Associate we earn from qualifying purchases.

This thread ...

Replies:

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].

Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.