Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1syuRg-0030MS-Re for pgsql-bugs@arkaria.postgresql.org; Thu, 10 Oct 2024 14:47:17 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.94.2) (envelope-from ) id 1syuRf-001kT0-Ty for pgsql-bugs@arkaria.postgresql.org; Thu, 10 Oct 2024 14:47:16 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1syuRf-001kSs-LN for pgsql-bugs@lists.postgresql.org; Thu, 10 Oct 2024 14:47:16 +0000 Received: from mail-pf1-x444.google.com ([2607:f8b0:4864:20::444]) by magus.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 (Exim 4.94.2) (envelope-from ) id 1syuRc-000M8S-Bx for pgsql-bugs@lists.postgresql.org; Thu, 10 Oct 2024 14:47:15 +0000 Received: by mail-pf1-x444.google.com with SMTP id d2e1a72fcca58-71e03be0d92so853703b3a.3 for ; Thu, 10 Oct 2024 07:47:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1728571631; x=1729176431; darn=lists.postgresql.org; h=mime-version:subject:references:in-reply-to:message-id:to:from:date :from:to:cc:subject:date:message-id:reply-to; bh=X7q6TqRWzxBjHla9MM/uOloaHOkUNsu92ZnihyCgxy8=; b=bK65LIkjZ99vDB7DzR2SNmDZ9LlNXJtlvz7Wvvn2mj4+JDEWHF5k2Hb3djkDmyQpaA 9seKpsp4LY73R8/1pz2i5JBROsLetC3vNqlgdnfd2dfzvI5dYQG4Ef31G4zamM/29GGa +gB1Bhs2bqIuLjSBYbgrOYRBNt7Nwt/GmbL8ix5OTEkesfJmrBhTWaVg4g4gUU1cBkOz YI1mlFdrTZBrH+IWD333wOStAlkak+Y+Youa2qbwOgn5yhK3mBofsABFRTRHZGOKkyx8 ZnIkHNAsoIG7ZLWGhb8A3ktI9RxnphHv16OHOQ8c1iIjHg5+OPdq/7H6U5L/uOpc5iGl XSQQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1728571631; x=1729176431; h=mime-version:subject:references:in-reply-to:message-id:to:from:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=X7q6TqRWzxBjHla9MM/uOloaHOkUNsu92ZnihyCgxy8=; b=xVhxLMRJ2Zlnl11EWbaDMqWMqlTj64JRl61qyohS3SYoKEOYQcjyAdrj3Nl4Fr6w50 DnDUUJZoiqY4yJK+kdO7xdWhpg3hOfsto61+5kqYqnS8G5MeR1RDXygM2HHKNJBwj1/u ESdXibqFIBYZlzpCwMzokfPKhjXxZg4i6ccrYfXGRw06MM0hMnP7CqsvD+yab4nVNUOH 5U+Do5wSPs8Do49JdbXTqzF8Kih0VcWh/Pj5sevdnn0L7a8P04k9zVXceFXwogET/ML+ jjeTSMVBDsnFQgtj66pdb9Cd17LHHwfAvoLrRnXIYnMHPev1NH/olnhgVCqZs13eyrCx PmiQ== X-Forwarded-Encrypted: i=1; AJvYcCWQEvGC3hsYp3VElHC7wp+H65C1IAv0bqK5MPkdUn5y+k7rUsAxKJUhqfaL14RjbqXWOBL1lBwHRcQT@lists.postgresql.org X-Gm-Message-State: AOJu0Yyh9CKSzKpncnw6BlWrbJt9pVgdkemChv7TaN5TNPBRe2gH+kQs yBDL4qB0Xfowu2Qs/S0P085FK9nmSYAXIQK/Xd1FbzBDI110Ojym X-Google-Smtp-Source: AGHT+IHPIP2vUXK5zGP6mCV65YxoVypVQZVz3a/CutxD32UAflXSygFUygsdMpCy60eaFsg8pi7TlA== X-Received: by 2002:a05:6a21:30cc:b0:1d8:b15c:f604 with SMTP id adf61e73a8af0-1d8b15cf80fmr3483157637.32.1728571630968; Thu, 10 Oct 2024 07:47:10 -0700 (PDT) Received: from [127.0.0.1] ([2400:8901::f03c:94ff:fe2f:b760]) by smtp.gmail.com with ESMTPSA id d2e1a72fcca58-71e2a9f6982sm1105263b3a.77.2024.10.10.07.47.09 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Thu, 10 Oct 2024 07:47:10 -0700 (PDT) Date: Thu, 10 Oct 2024 22:47:00 +0800 From: Zhang Mingli To: Ba Jinsheng , "=?utf-8?Q?pgsql-bugs=40lists.postgresql.org?=" , Tomas Vondra Message-ID: <3c7df680-97e6-4f20-ab27-65fb8282f95e@Spark> In-Reply-To: <20f9671d-6f40-429b-a749-a2f53086e0dd@vondra.me> References: <20f9671d-6f40-429b-a749-a2f53086e0dd@vondra.me> Subject: Re: Question of Parallel Hash Join on TPC-H Benchmark X-Readdle-Message-ID: 3c7df680-97e6-4f20-ab27-65fb8282f95e@Spark MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="6707e8eb_59aae3cc_f40f" List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Archived-At: Precedence: bulk --6707e8eb_59aae3cc_f40f Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hi, Zhang Mingli www.hashdata.xyz On Oct 10, 2024 at 22:41 +0800, Tomas Vondra , wrote: > > You can try setting parallel=5Fsetup=5Fcost and parallel=5Ftuple=5Fcost= to 0, > and reducing min=5Fparallel=5F=7Btable,index=7D=5Fscan=5Fsize. That sho= uld force a > parallel plan, and you'll see if parallelism is helpful. In addition to this, it seems you=E2=80=99r eager a parallel-aware HashJo= in. Check if=C2=A0=C2=A0enable=5Fparallel=5Fhash =3D on, an explain(verbose) = will show related GUCs. --6707e8eb_59aae3cc_f40f Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline
Hi,


Zhang Mingli
www.hashdata.xyz
On Oct 10, 2024 at 22:41 +0800, Tomas Vondra <to= mas=40vondra.me>, wrote:

You can try setting parallel=5Fsetup=5Fcost and parallel=5Ftuple=5Fcost t= o 0,
and reducing min=5Fparallel=5F=7Btable,index=7D=5Fscan=5Fsize. That shoul= d force a
parallel plan, and you'll see if parallelism is helpful.
In addition to this, it seems you=E2=80=99r eager a= parallel-aware HashJoin.
Check if&=23160;&=23160;enable=5Fparallel=5Fhash =3D on, an explain(verbo= se) will show related GUCs.
--6707e8eb_59aae3cc_f40f--