Return-Path: X-Original-To: apmail-flink-user-archive@minotaur.apache.org Delivered-To: apmail-flink-user-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 18FF418A33 for ; Mon, 6 Jul 2015 08:29:35 +0000 (UTC) Received: (qmail 85238 invoked by uid 500); 6 Jul 2015 08:29:35 -0000 Delivered-To: apmail-flink-user-archive@flink.apache.org Received: (qmail 85156 invoked by uid 500); 6 Jul 2015 08:29:34 -0000 Mailing-List: contact user-help@flink.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: user@flink.apache.org Delivered-To: mailing list user@flink.apache.org Received: (qmail 85146 invoked by uid 99); 6 Jul 2015 08:29:34 -0000 Received: from Unknown (HELO spamd1-us-west.apache.org) (209.188.14.142) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 06 Jul 2015 08:29:34 +0000 Received: from localhost (localhost [127.0.0.1]) by spamd1-us-west.apache.org (ASF Mail Server at spamd1-us-west.apache.org) with ESMTP id 791E8D26BB for ; Mon, 6 Jul 2015 08:29:34 +0000 (UTC) X-Virus-Scanned: Debian amavisd-new at spamd1-us-west.apache.org X-Spam-Flag: NO X-Spam-Score: 2.9 X-Spam-Level: ** X-Spam-Status: No, score=2.9 tagged_above=-999 required=6.31 tests=[DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=3, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=disabled Authentication-Results: spamd1-us-west.apache.org (amavisd-new); domainkeys=fail (1024-bit key) reason="fail (message has been altered)" header.from=jsparks@cray.com header.d=cray.com; dkim=pass (1024-bit key) header.d=cray.com Received: from mx1-us-west.apache.org ([10.40.0.8]) by localhost (spamd1-us-west.apache.org [10.40.0.7]) (amavisd-new, port 10024) with ESMTP id fNEqcbBjJcKX for ; Mon, 6 Jul 2015 08:29:27 +0000 (UTC) Received: from esa1.cray.iphmx.com (esa1.cray.iphmx.com [68.232.142.33]) by mx1-us-west.apache.org (ASF Mail Server at mx1-us-west.apache.org) with ESMTPS id CE7A5205B4 for ; Mon, 6 Jul 2015 08:29:26 +0000 (UTC) DomainKey-Signature: s=cray1024; d=cray.com; c=nofws; q=dns; h=X-IronPort-AV:X-Cray-OBMMKR:Received:Received:Received: From:To:Subject:Thread-Topic:Thread-Index:Date:Message-ID: References:In-Reply-To:Accept-Language:Content-Language: X-MS-Has-Attach:X-MS-TNEF-Correlator:x-originating-ip: Content-Type:MIME-Version:Return-Path; b=Ew03PfNhZoYFlJ2vZZXRVb8zyyocciznzwbdYYO0uOO1HG7CZGlZ0XkX 3AsbjDKeS70D6J9Qsw4YiL0M5dcngfDNefDxOIPxg7iQOj4E/rzn1e9z9 N9gJgS3nz4ioGPlNPHpGIAit5t9Z1X7WyGq4NifJWk2j4F5NjhmGoeLLA k=; DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=cray.com; i=@cray.com; q=dns/txt; s=cray1024; t=1436171367; x=1467707367; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=KP5kd5kqc1ZYJieVDdvAVVqvyPZfnXJt/qx/EDlshXs=; b=TtiE9deYwPsBAQ5AvIJ6ZYzZSNcgxohgGFfgDmJEN+7ci+ZzYxhZGUGK z9KnmNtBNWsClxdVUmDinDN76R3jzamazZpdGyK1i2lDI3zcEIV9xH3xD y1B7ZgmuC4AJsjMDXwvqUSMchVYFzhFZYtQiPjXWGA2tkzXL+6jM6tsa4 M=; X-IronPort-AV: E=Sophos;i="5.15,413,1432594800"; d="scan'208,217";a="2715385" X-Cray-OBMMKR: 1433258124 Received: from cray-smtp-2.cray.com (HELO CFWEX01.americas.cray.com) ([136.162.34.11]) by esa1.cray.iphmx.com with ESMTP/TLS/AES128-SHA; 06 Jul 2015 09:28:19 +0100 Received: from CFWEX02.americas.cray.com (172.30.74.25) by CFWEX01.americas.cray.com (172.30.88.25) with Microsoft SMTP Server (TLS) id 14.3.224.2; Mon, 6 Jul 2015 03:28:18 -0500 Received: from CFWEX01.americas.cray.com ([169.254.1.128]) by cfwex02.americas.cray.com ([169.254.2.204]) with mapi id 14.03.0224.002; Mon, 6 Jul 2015 03:28:17 -0500 From: Bill Sparks To: "user@flink.apache.org" Subject: RE: data conversion between flink and "other" paradigms Thread-Topic: data conversion between flink and "other" paradigms Thread-Index: AdC3vWnOOl/hHKitTgqazg6q+/BtlQAMYryA//+tP00= Date: Mon, 6 Jul 2015 08:28:16 +0000 Message-ID: <121F01D0B574FC4DA43817B7D9CF419E243FB07E@CFWEX01.americas.cray.com> References: <121F01D0B574FC4DA43817B7D9CF419E243FB011@CFWEX01.americas.cray.com>, In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [192.168.234.40] Content-Type: multipart/alternative; boundary="_000_121F01D0B574FC4DA43817B7D9CF419E243FB07ECFWEX01americas_" MIME-Version: 1.0 --_000_121F01D0B574FC4DA43817B7D9CF419E243FB07ECFWEX01americas_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Fabian. Thanks for the info and pointer to python. I'll check it out. -Bill ________________________________ From: Fabian Hueske [fhueske@gmail.com] Sent: Monday, July 06, 2015 3:23 AM To: user@flink.apache.org Subject: Re: data conversion between flink and "other" paradigms Hi Bill, a DataSet is just a logical concept in Flink. DataSets are often not persis= ted and just streamed along operators. At the moment, there is no way to ac= cess an intermediate DataSet of a Flink program directly (this might change= in the future). You can process data in another function by implementing a Java user functi= on (for example a MapPartition function) and sending the data through JNI t= o a C function (if you need the full data set, you must set the parallelism= to 1). Flink's Python API follows a similar approach to ship data from Fli= nk to an external Python process. Best, Fabian 2015-07-06 9:30 GMT+02:00 Bill Sparks >: Just a question if there was some prior-art here. Just say someone wanted t= o use flink for processing, but at some point they wanted to call another f= unction via say JNI/C which doesn't understand DataSet's. How would one go = about this ... I'm assuming the code would have to convert the data to a co= mmon format prior to calling the function. Regards, Bill. --_000_121F01D0B574FC4DA43817B7D9CF419E243FB07ECFWEX01americas_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Fabian.

 

Thanks for the info and pointer to python. I'll check it out.

 

-Bill

From: Fabian Hueske [fhueske@gmail.com] Sent: Monday, July 06, 2015 3:23 AM
To: user@flink.apache.org
Subject: Re: data conversion between flink and "other" par= adigms

Hi Bill,

a DataSet is just a logical concept in Flink. DataSets are often not persis= ted and just streamed along operators. At the moment, there is no way to ac= cess an intermediate DataSet of a Flink program directly (this might change= in the future).

You can process data in another function by implementing a Java user functi= on (for example a MapPartition function) and sending the data through JNI t= o a C function (if you need the full data set, you must set the parallelism= to 1). Flink's Python API follows a similar approach to ship data from Flink to an external Python process.<= br>
Best, Fabian



2015-07-06 9:30 GMT+02:00 Bill Sparks <jspar= ks@cray.com>:

Just a question if there was some prior-art here. Just say someone wante= d to use flink for processing, but at some point they wanted to call anothe= r function via say JNI/C which doesn't understand DataSet's. How would one = go about this ... I'm assuming the code would have to convert the data to a common format prior to calling th= e function.

 

 

Regards,

   Bill.


--_000_121F01D0B574FC4DA43817B7D9CF419E243FB07ECFWEX01americas_--