top top - 13:05:18 up 1:49, 1 user, load average: 0.17, 0.13, 0.06 Tasks: 107 total, 3 running, 103 sleeping, 0 stopped, 1 zombie Cpu0 : 6.3%us, 0.9%sy, 0.0%ni, 92.3%id, 0.4%wa, 0.1%hi, 0.1%si, 0.0%st Mem: 514444k total, 507384k used, 7060k free, 10976k buffers Swap: 321260k total, 33856k used, 287404k free, 168936k cached
On 10/8/07, Eduardo Tongson <[EMAIL PROTECTED]> wrote: > > Press '1' when running top(1). > > E > On 10/8/07, war cries <[EMAIL PROTECTED]> wrote: > > Guys, > > > > Just checked the bootup process of our server (DELL PowerEdge 1750) and > it > > uses one processor. The Xeon processor is using HT technology so that's > why > > my /proc/cpuinfo output detected it as processor 0 and 1. But when I > used > > "top" command, the output tells me that HT is disabled, please correct > me if > > I'm wrong with my analysis. > > ___ > > top > > > > top - 09:50:11 up 1 day, 16:13, 1 user, load average: 0.66, 0.63, 0.58 > > Tasks: 162 total, 1 running, 161 sleeping, 0 stopped, 0 zombie > > Cpu(s): 0.7%us, 0.2%sy, 0.0%ni, 99.1%id, 0.0%wa, 0.0%hi, 0.0%si, > > 0.0%st > > Mem: 1035348k total, 956120k used, 79228k free, 167356k buffers > > Swap: 1966072k total, 0k used, 1966072k free, 540536k cached > > ___ > > > > Guys this server is a Proxy/Caching Server (squid) and Content Filtering > > (dansguardian). We want to maximize the processing capability of this > server > > because currently it is serving a thousand (1000 min - 5000 max) > researchers > > in our institution. > > > > Need your advise. Thank you. > > > > > > > > On 10/5/07, Orlando Andico <[EMAIL PROTECTED]> wrote: > > > Well, what the OS thinks is really irrelevant w.r.t. Hyper-Threaded > > processors. > > > > > > Hyper-Threading only gives 5% performance improvement over a single > > > thread, on CPU-bound activities. A thread is much, much less than a > > > core. > > > > > > That actually explains why you only see one or two bars with lots of > > activity. > > > > > > To quote one of my colleagues from Redwood Shores: > > > > > > "Yes. Hyperthreading only works if there are lots of L2 cache misses. > > > Then the CPU can work on the pretend CPU's work while waiting. But if > > > there were no cache misses then there would be zero benefit to HT. So > > > the tighter your code is the less HT works. > > > > > > HT is basically a fraud. You are seeing exactly what you will with > > > almost anu software, not just with TimesTen. If you get even 5% > > > improvement then you did good. No one should expect much more." > > > > > > > > > > > > > > > On 10/5/07, Michael Tinsay <[EMAIL PROTECTED]> wrote: > > > .. > > > > I have a server that has a dual-core, HT-enabled processor. Ubuntu > says > > it > > > > has 4 processors. htop shows 4 bars for processors, but most of the > > time > > > > only one or two have activity. > > > _________________________________________________ > > > Philippine Linux Users' Group (PLUG) Mailing List > > > [email protected] (#PLUG @ irc.free.net.ph) > > > Read the Guidelines: http://linux.org.ph/lists > > > Searchable Archives: http://archives.free.net.ph > > > > > > > > > _________________________________________________ > > Philippine Linux Users' Group (PLUG) Mailing List > > [email protected] (#PLUG @ irc.free.net.ph) > > Read the Guidelines: http://linux.org.ph/lists > > Searchable Archives: http://archives.free.net.ph > > > _________________________________________________ > Philippine Linux Users' Group (PLUG) Mailing List > [email protected] (#PLUG @ irc.free.net.ph) > Read the Guidelines: http://linux.org.ph/lists > Searchable Archives: http://archives.free.net.ph >
_________________________________________________ Philippine Linux Users' Group (PLUG) Mailing List [email protected] (#PLUG @ irc.free.net.ph) Read the Guidelines: http://linux.org.ph/lists Searchable Archives: http://archives.free.net.ph
