From issues-return-7820-archive-asf-public=cust-asf.ponee.io@systemml.apache.org Tue Feb 13 21:51:05 2018 Return-Path: X-Original-To: archive-asf-public@cust-asf.ponee.io Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by mx-eu-01.ponee.io (Postfix) with SMTP id A06BB18067B for ; Tue, 13 Feb 2018 21:51:04 +0100 (CET) Received: (qmail 90955 invoked by uid 500); 13 Feb 2018 20:51:03 -0000 Mailing-List: contact issues-help@systemml.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@systemml.apache.org Delivered-To: mailing list issues@systemml.apache.org Received: (qmail 90946 invoked by uid 99); 13 Feb 2018 20:51:03 -0000 Received: from pnap-us-west-generic-nat.apache.org (HELO spamd4-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 13 Feb 2018 20:51:03 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd4-us-west.apache.org (ASF Mail Server at spamd4-us-west.apache.org) with ESMTP id 2FFDBC0110 for ; Tue, 13 Feb 2018 20:51:03 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd4-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: -109.511 X-Spam-Level: X-Spam-Status: No, score=-109.511 tagged_above=-999 required=6.31 tests=[ENV_AND_HDR_SPF_MATCH=-0.5, KAM_ASCII_DIVIDERS=0.8, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_SPF_WL=-7.5, USER_IN_WHITELIST=-100] autolearn=disabled Received: from mx1-lw-eu.apache.org ([10.40.0.8]) by localhost (spamd4-us-west.apache.org [10.40.0.11]) (amavisd-new, port 10024) with ESMTP id cZCCIaz2OpHY for ; Tue, 13 Feb 2018 20:51:02 +0000 (UTC) Received: from mailrelay1-us-west.apache.org (mailrelay1-us-west.apache.org [209.188.14.139]) by mx1-lw-eu.apache.org (ASF Mail Server at mx1-lw-eu.apache.org) with ESMTP id 58A205F180 for ; Tue, 13 Feb 2018 20:51:01 +0000 (UTC) Received: from jira-lw-us.apache.org (unknown [207.244.88.139]) by mailrelay1-us-west.apache.org (ASF Mail Server at mailrelay1-us-west.apache.org) with ESMTP id 8F933E00EA for ; Tue, 13 Feb 2018 20:51:00 +0000 (UTC) Received: from jira-lw-us.apache.org (localhost [127.0.0.1]) by jira-lw-us.apache.org (ASF Mail Server at jira-lw-us.apache.org) with ESMTP id 4C09D240F6 for ; Tue, 13 Feb 2018 20:51:00 +0000 (UTC) Date: Tue, 13 Feb 2018 20:51:00 +0000 (UTC) From: "Govinda Malavipathirana (JIRA)" To: issues@systemml.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (SYSTEMML-2083) Language and runtime for parameter servers MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/SYSTEMML-2083?page=3Dcom.atlass= ian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D1= 6363016#comment-16363016 ]=20 Govinda Malavipathirana commented on SYSTEMML-2083: --------------------------------------------------- Hi I went through the documentation, built up the SystemML locally and ran som= e samples as well, observing on paper yet to done. What I have seen it not = only=C2=A0specific on machine learning use cases but works on iterative alg= orithms (linear-algebraic) as well? and What could be the next step for me? > Language and runtime for parameter servers > ------------------------------------------ > > Key: SYSTEMML-2083 > URL: https://issues.apache.org/jira/browse/SYSTEMML-2083 > Project: SystemML > Issue Type: Epic > Reporter: Matthias Boehm > Priority: Major > Labels: gsoc2018 > > SystemML already provides a rich set of execution strategies ranging from= local operations to large-scale computation on MapReduce or Spark. In this= context, we support both data-parallel (multi-threaded or distributed oper= ations) as well as task-parallel computation (multi-threaded or distributed= parfor loops). This epic aims to complement the existing execution strateg= ies by language and runtime primitives for parameter servers, i.e., model-p= arallel execution. We use the terminology of model-parallel execution with = distributed data and distributed model to differentiate them from the exist= ing data-parallel operations. Target applications are distributed deep lear= ning and mini-batch algorithms in general. These new abstractions will help= making SystemML a unified framework for small- and large-scale machine lea= rning that supports all three major execution strategies in a single framew= ork. > =C2=A0 > A major challenge is the integration of stateful parameter servers and th= eir common push/pull primitives into an otherwise functional (and thus, sta= teless) language. We will approach this challenge via a new builtin functio= n \{{paramserv}} which internally maintains state but at the same time fits= into the runtime framework of stateless operations. > Furthermore, we are interested in providing (1) different runtime backend= s (local and distributed), (2) different parameter server modes (synchronou= s, asynchronous, hogwild!, stale-synchronous), (3) different update frequen= cies (batch, multi-batch, epoch), as well as (4) different architectures fo= r distributed data (1 parameter server, k workers) and distributed model (k= 1 parameter servers, k2 workers).=C2=A0 -- This message was sent by Atlassian JIRA (v7.6.3#76005)