charm AT lists.siebelschool.illinois.edu

Subject: Charm++ parallel programming system

List archive

[charm] FW: FW: Projections

From: "Ortega, Bob" <bobo AT mail.smu.edu>
To: Ronak Buch <rabuch2 AT illinois.edu>
Cc: "charm AT lists.cs.illinois.edu" <charm AT lists.cs.illinois.edu>
Subject: [charm] FW: FW: Projections
Date: Tue, 15 Dec 2020 18:01:18 +0000
Accept-language: en-US
Arc-authentication-results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=mail.smu.edu; dmarc=pass action=none header.from=mail.smu.edu; dkim=pass header.d=mail.smu.edu; arc=none
Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=Ar2A6IC83FHA9lD/Cme1AM1BH9xjP781xDSaVlybKtY=; b=A/PSYWdb7F+3vGA2KTqmuZ2t7KcitEmLSrvQsNk3PPJe0CZIczz5oZfkpw/ZtnReNOlYUs9HzRPboZVzIGG84soc+TC+WKz2vWImTYT69UJyu31PWiYAWRpBNuOy6XU6t3Taa+PViHBrKkcIS/+7Z5SY5bUJ54Me17UZiaYDFfWGxwt3NSJnVAMiTL3CbZpcU6lsgJeFx+um6jwHAYqNvkpTl/zUSL8O4vdvEd65hNUCLuagQq8h6buroRQEUrOgwZn1ep9OldPPvtKvZgctkh+VhL+iZok/llN5XX0vVWIj0XL247MtkWV06tlGuOxKnza43L7PKf43G6dyA8dkFw==
Arc-seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=ZT5JEH8Q3kyLlXmfa1H9NYom71n+waWXa+AvuLz2/czpxHPmG+howZWFli5CMXVqoP2YWpmp8v0s17UQopqvBEFED73qqkcivaRYOJlP/2SOXFYQQhDb8Xpn+VAbVbJ1uLsQtZqfMJP8xNXLYJhaThsg5GImo+L/yO+ybNqZURvxok6DIuXPAH6V9r5n8RCxtTfJs/jPSkvOCsTWQWOSK8raH0Zx1Xc6b/saycXe+FCt0xZogDifnp3nvPdB0ELk4poT4i+oWTh2MdiNo2HEnrg8EvM4gYQOQByk0iH07xz2w5UgjbAzLYjFX1F9totU6bGN9HRKTmMYppQaOpESDQ==
Authentication-results: illinois.edu; spf=pass smtp.mailfrom=bobo AT mail.smu.edu; dkim=pass header.d=smu365.onmicrosoft.com header.s=selector2-smu365-onmicrosoft-com; dmarc=none

Ronak,

Here are some interesting screenshots from Projections I’m trying to understand:

Namd projections run against 1m atom dataset and 2500 steps:

Namd projections run against 1m atom dataset and 1500 steps:

Not sure but is this signaling some kind of network issue in case 1 and then in case 2 showing that processing is done by ~357s? Sorry if these questions are simple but we were attempting to extend the processing time by increasing the number of steps so we could observe processing on a longer scale.

Again, thank you for your time and support.

Bob

From: Ronak Buch <rabuch2 AT illinois.edu>
Date: Monday, December 14, 2020 at 6:21 PM
To: "Ortega, Bob" <bobo AT mail.smu.edu>
Subject: Re: [charm] FW: Projections

Ah, I see. You can try the verbs build of Charm++ if you'd like, it's a bit outdated relative to the UCX layer, but should still provide better performance than MPI if it works

(granted, the performance differences between the various layers many not be that large, so it may not be worth worrying about).

-Ronak

[charm] FW: Projections, Ortega, Bob, 12/10/2020
- <Possible follow-up(s)>
- Re: [charm] FW: Projections, Ronak Buch, 12/10/2020
  - Re: [charm] FW: Projections, Ortega, Bob, 12/10/2020
  - Message not available
    - Re: [charm] FW: Projections, Ronak Buch, 12/11/2020
      - Message not available
        
        Message not available
        
        Message not available
        
        Message not available
        Message not available
        Re: [charm] FW: Projections, Ortega, Bob, 12/14/2020
      - Message not available
        
        Message not available
        
        Message not available
        
        Message not available
        Message not available
        Message not available
        Re: [charm] FW: Projections, Ronak Buch, 12/14/2020
        Re: [charm] FW: Projections, Ortega, Bob, 12/14/2020
        
        Message not available
        Message not available
        Message not available
        [charm] FW: FW: Projections, Ortega, Bob, 12/15/2020