[ale] Tool search

[Jim Kinney]

These are looking promising. 

As I want to measure resource consumption, I can designate a memory page as a resource and count page writes as a measure of data flowing through a job run. Should be able to step through cgroups to isolate a pid and it's children. 

Yeah. This is getting hard.

On October 16, 2020 8:49:32 AM EDT, Jerald Sheets <questy at gmail.com> wrote:
>
>
>> On Oct 16, 2020, at 8:25 AM, Ed Cashin via Ale <ale at ale.org> wrote:
>> 
>> I was watching for responses.  This doesn't sound simple at all to
>me.
>> 
>> For a while they were adding dtrace to Linux, and I think Oracle
>Linux had it.  I don't know whether it made it into the mainline, but
>if so, you could probably put together what you need using dtrace. 
>You'd have to consider all the ways (direct IO, minor page faults)
>bytes can get into RAM, though, and filter out stuff you're not
>interested in (local disk, etc.).
>> 
>> If there's no special tool and no dtrace, ftrace might be of interest
>to you.
>
>There’s “bpftrace”
>http://www.brendangregg.com/blog/2018-10-08/dtrace-for-linux-2018.html
><http://www.brendangregg.com/blog/2018-10-08/dtrace-for-linux-2018.html>
>
>The OEL version:
>https://www.oracle.com/linux/downloads/linux-dtrace.html
><https://www.oracle.com/linux/downloads/linux-dtrace.html>
>
>There’s a project:
>https://github.com/dtrace4linux/linux
><https://github.com/dtrace4linux/linux>
>
>(I like this one.  It’s the actual Oracle DTracefor linux  code, but
>doesn’t modify the kernel…. It uses a KLM to provide the endpoints
>instead.
>
>
>
>—j

-- 
Computers amplify human error
Super computers are really cool
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.ale.org/pipermail/ale/attachments/20201016/f3f81bd0/attachment.html>