Sponsored Content
Operating Systems Solaris Assign a process to a core on start Post 302910728 by achenle on Sunday 27th of July 2014 06:17:28 PM
Old 07-27-2014
I do wonder what problem is being solved here. Binding to processors is almost never worthwhile. It might get you somewhat better performance for a memory-latency-bound application on a NUMA machine where you can guarantee your application uses memory with better locality to the processor(s) you bind to, but that's about it.

If you're grasping that hard for every last processor cycle, how much time have you spent profiling your application and improving its performance directly based on profiled performance data?
 

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

how to start a process and make it sleep for 5 mins and then kill that process

how to start a process and make it sleep for 5 mins and then kill that process (6 Replies)
Discussion started by: shrao
6 Replies

2. UNIX for Dummies Questions & Answers

Getting core id, physical id of running process

Hi All, I need a help. I need to know: 1. the individual core id, physical id of a running process on multicore inside a program. The system file /proc/cpuinfo shows all the ids of all the processes. But I need to know runtime, what exact core id and physical id , the running process has.... (0 Replies)
Discussion started by: debrajde
0 Replies

3. Solaris

coreadm diasble a process from core dumping

Hello All Is it possible to disable a specific process from core dumping ? In my environment I have 2 bespoke application processes which needs to be stopped from core dumping but any other process should be allowed to core dump. If I do : coreadm -d process it will stop all per processes... (3 Replies)
Discussion started by: baner_n
3 Replies

4. UNIX for Dummies Questions & Answers

How to force core dump of a process

We have an application that terminates with segmentation violation errors in the logs. No source code is available since this is a third party software that is way past its maintenance life cycle. Under these circumstances is there a way to force a core dump of the process for further analysis?? ... (3 Replies)
Discussion started by: Un1xNewb1e
3 Replies

5. Shell Programming and Scripting

Script - How to automatically start another process when the previous process ends?

Hi all, I'm doing automation task for my team and I just started to learn unix scripting so please shed some light on how to do this: 1) I have 2 sets of datafiles - datafile A and B. These datafiles must be loaded subsequently and cannot be loaded concurrently. 2) So I loaded datafile A... (10 Replies)
Discussion started by: luna_soleil
10 Replies

6. Programming

how to monitor the child process on which cpu core

Hi all. Sorry to express my questions wrongly in my early post,I repost my question again here. My pc has dual core, I wirte an application with two process, parents process and child process. My quetion is how to realize :if the child process is on core 0,it will tell me I'm on core 0,if it... (1 Reply)
Discussion started by: yanglei_fage
1 Replies

7. UNIX for Dummies Questions & Answers

Script to start background process and then kill process

What I need to learn is how to use a script that launches background processes, and then kills those processes as needed. The script successfully launches the script. But how do I check to see if the job exists before I kill it? I know my problem is mostly failure to understand parameter... (4 Replies)
Discussion started by: holocene
4 Replies

8. Red Hat

Process does not dump any core files when crashed even if coredumpsize is unlimited

Hello Im using redhat and try to debug my application , its crashes and in strace I also see it has problems , but I can't see any core dump I configured all the limit ( im using .cshrc ) and it looks like this : cputime unlimited filesize unlimited datasize unlimited... (8 Replies)
Discussion started by: umen
8 Replies

9. Shell Programming and Scripting

Get the process/thread running on which core

I use top -H -p 1256 to show process 1256, and then press "f" then press "j" to display it is running on which core, is there a better mothed, I want to be automated to get this (1 Reply)
Discussion started by: yanglei_fage
1 Replies

10. Red Hat

Core Dump of a process in Red Hat Linux 5.9

Hello All, I am new joiner of this forum.I am new to Linux shell scripting. At present I have identified 1 application which stalls very frequently (PID is say xyz) and I am not having much information in its application log to identify the root cause of stalling. I need to take the core dump... (19 Replies)
Discussion started by: Anjan Ganguly
19 Replies
loratune(1M)															      loratune(1M)

NAME
loratune - initiate LORA tuneup to improve alignment of processing resources SYNOPSIS
DESCRIPTION
The command improves the performance of an HP-UX system in LORA (Locality-Optimizated Resource Alignment) mode, by triggering the kernel to do optimization tuning of the system and application memory. The command may return before the tuning is complete. This tuning can continue in the background for several minutes after the command returns. If the command is interrupted, or the command process is otherwise terminated, no subsequent tuning operations will be initiated, but those in progress will continue until completion. When no options are specified, all locality domains are tuned. Options This option limits the tuning to the locality domain specified by lid. Application Usage In Locality-Optimized Resource Alignment (LORA) mode, performance of applications may improve when their processors and the memory that they use are aligned in the same locality. The HP-UX kernel attempts to maintain such alignment at all times, but it is possible for mis- alignment to occur when the system workload transitions significantly. If that happens, the command can be used to reestablish alignment and improve application performance. Here are some examples of workload transition events that may cause misalignment of processing resources: o Starting or terminating an application that consumes a large amount of processor or memory resources. o The workload demand surges far above the normal level and then recedes. o Dynamic platform operations that add or subtract processor or memory resources from the operating system's control. If one of those events has occurred, and application performance is not as high as it can be, then it is appropriate to invoke the command. RETURN VALUES
The command returns a 0 when successful, and a non-zero value when unsuccessful. ERRORS
The command will not initiate any tuning and will return a non-zero value if the user does not have appropriate privilege. WARNINGS
The command can consume a significant amount of system processing resources to restore optimal alignment. Ideally, the command should be invoked prior to the time when performance is critical, not during the time when performance is critical. The command should be invoked after the workload transition is complete, not during workload transition. Although tuning can continue in the background for several min- utes, it is not necessary to wait any more than five minutes between invoking the command and launching new applications. EXAMPLES
If eight instances of SAP were running in localities with locality domain identifiers 4 and 5, and four of those instances were shut down, then it would be appropriate to tuneup the remaining four instances. An example command would be: If an HP-UX system in LORA mode had one of its cells deactivated, then it would be appropriate to tuneup the entire system. An example command would be: AUTHOR
was developed by HP. SEE ALSO
mpsched(1), numa_policy(5), lora(7). Itanium(R)-based Systems Only loratune(1M)
All times are GMT -4. The time now is 02:34 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy