pcp
[Top] [All Lists]

Re: [pcp] pcp grafana and graphite - How to convert pcp metric values in

To: "Frank Ch. Eigler" <fche@xxxxxxxxxx>
Subject: Re: [pcp] pcp grafana and graphite - How to convert pcp metric values into percent
From: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Thu, 03 Jul 2014 09:06:35 +1000
Cc: Amer Ather <aather@xxxxxxxxxxx>, Martin Spier <mspier@xxxxxxxxxxx>, pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <20140702203904.GA21540@xxxxxxxxxx>
References: <y0mha353ky5.fsf@xxxxxxxx> <y0m8uoh3kf4.fsf@xxxxxxxx> <002d01cf9284$dc8fe570$95afb050$@internode.on.net> <002f01cf9285$f0c63160$d2529420$@internode.on.net> <CAM1aq-HgP5+Tsq_sCLWH6GrLOn8UXQWaEVkVaC1p=TT_kMn6Tg@xxxxxxxxxxxxxx> <20140629220735.GA13993@xxxxxxxxxx> <CAM1aq-Hr2ss+b82p1EMCRTjK_Vsbnde_tKVv53MQfP25p4e-aw@xxxxxxxxxxxxxx> <y0m1tu41lld.fsf@xxxxxxxx> <CAM1aq-EGP8=DGUva+JSk3WMeLRisi0Weg3spN1+SLghgrOKVtQ@xxxxxxxxxxxxxx> <53B46CB4.40409@xxxxxxxxxxxxxxxx> <20140702203904.GA21540@xxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.6.0
G'day Frank.

On 03/07/14 06:39, Frank Ch. Eigler wrote:
...
Indeed so.  A new issue under contention is whether the client side of
PMAPI - libpcp - could/should have rate-conversion capabilities, so
each client programs wouldn't have to reinvent that wheel.

And this direction is something I would strongly support, ... but

One more piece of history (at the risk of boring the readers) ... the universal "goodness" of this approach was recognized, er I'd guess 15 years ago, after we'd repeated similar client-side logic in pmval, then pmie, then pmchart, then pmview, then pmdumptext, and then a re-written pmview, ...

At that time it was known as the "performance metrics class" and became something of the holy grail.

The sad story is that we tried (at least 3 times I believe) to build the performance metrics class and failed each time. The only surviving code from these efforts is the libpcp_qmc code in the source tree that is used by pmtime, pmchart and pmdumptext (and pmview although that's not really in the tree yet).

I wish I had better recall of the reasons for the failures to help inform the current deliberations ... but the C vs C++ divide was at least part of the problem back then.

<Prev in Thread] Current Thread [Next in Thread>