From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on inbox.vuxu.org X-Spam-Level: X-Spam-Status: No, score=0.3 required=5.0 tests=DKIM_ADSP_CUSTOM_MED, DKIM_SIGNED,DKIM_VALID,FORGED_GMAIL_RCVD,FREEMAIL_FORGED_FROMDOMAIN, FREEMAIL_FROM,HEADER_FROM_DIFFERENT_DOMAINS,HTML_MESSAGE, MAILING_LIST_MULTI,RCVD_IN_DNSWL_NONE autolearn=no autolearn_force=no version=3.4.4 Received: from tb-ob20.topicbox.com (tb-ob20.topicbox.com [173.228.157.66]) by inbox.vuxu.org (Postfix) with ESMTP id 4A5B428A65 for ; Tue, 28 May 2024 21:33:35 +0200 (CEST) Received: from tb-mx0.topicbox.com (tb-mx0.nyi.icgroup.com [10.90.30.73]) by tb-ob20.topicbox.com (Postfix) with ESMTP id 2552B33A60 for ; Tue, 28 May 2024 15:33:34 -0400 (EDT) (envelope-from bounce.mM510201f156a99eb26c5c7268.r522be890-2105-11eb-b15e-8d699134e1fa@9fans.bounce.topicbox.com) Received: by tb-mx0.topicbox.com (Postfix, from userid 1132) id D860B1A3AC10; Tue, 28 May 2024 15:33:33 -0400 (EDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed; d=9fans.net; h=from:to :subject:message-id:in-reply-to:references:date:mime-version :content-type:content-transfer-encoding:list-help:list-id :list-post:list-subscribe:reply-to:list-unsubscribe; s=dkim-1; t=1716924813; x=1717011213; bh=LyXkq3j4/Wepf2xJ93ukxu9iTsNBUlVu d0EMmbgyjgQ=; b=kB5BWWeViUp/saj5bCehSodvlUwHEYTARXKLU5rPSxlt7wrj b1rEy+AYwL6WUh36oz2STqoV3ZhLNsEItkcNJOhZhOEh/ItbnVFamXsni2RL05sL Pxgb/idWlmvJaul0GHYophMKZu/v1rjbBmnBY1Xoc2yWgK7dJQ5abkEpzjc= From: wb.kloke@gmail.com To: 9fans <9fans@9fans.net> Subject: Re: [9fans] Throwing in the Towel Message-Id: <17169248040.5f383EFCe.47999@composer.9fans.topicbox.com> In-Reply-To: <16277.1716771974@lunacy.ugrad.cs.cmu.edu> References: <16277.1716771974@lunacy.ugrad.cs.cmu.edu> Date: Tue, 28 May 2024 15:33:24 -0400 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary=17169248041.0a93dE8E.47999 Content-Transfer-Encoding: 7bit Topicbox-Policy-Reasoning: allow: sender is a member Topicbox-Message-UUID: 26990252-1d29-11ef-ac8c-b81740decc0b Archived-At: =?UTF-8?B?PGh0dHBzOi8vOWZhbnMudG9waWNib3guY29tL2dyb3Vwcy85?= =?UTF-8?B?ZmFucy9UMmNhNjc0ODZjN2ExM2E3Ny1NNTEwMjAxZjE1NmE5OWViMjZjNWM3?= =?UTF-8?B?MjY4Pg==?= List-Help: List-Id: "9fans" <9fans.9fans.net> List-Post: List-Software: Topicbox v0 List-Subscribe: Precedence: list Reply-To: 9fans <9fans@9fans.net> List-Unsubscribe: , Topicbox-Delivery-ID: 2:9fans:437d30aa-c441-11e9-8a57-d036212d11b0:522be890-2105-11eb-b15e-8d699134e1fa:M510201f156a99eb26c5c7268:1:9UdkWUTanF8WoaRD3zMx5RjDYt_sUJzaoBr52Knzx34 --17169248041.0a93dE8E.47999 Date: Tue, 28 May 2024 15:33:24 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable For the napkin calculation: On disk, the IEntry is 38Bytes. Alas, writes oc= cur always in (the ssd internal) blocksize. So, essentially (assuming 4096 = byte blocksize, which is quite optimistic), we have a write efficiency of l= ess than 1 percent. A good firmware in the ssd could avoid needing a new block for the write, i= f all bits are changed in teh same direction by the new data. This is defea= ted by the binary search in the index buckets. The entries must be sorted, = and the number of used entries is tob updated. So, for each new entry the w= hole (internal) block has to be remapped. Therefore an erasure block must b= e erased after about=C2=A0 1m/4k index writes. It seems, venti in its current form is a ssd killer, if they are used for t= he isects. ------------------------------------------ 9fans: 9fans Permalink: https://9fans.topicbox.com/groups/9fans/T2ca67486c7a13a77-M51020= 1f156a99eb26c5c7268 Delivery options: https://9fans.topicbox.com/groups/9fans/subscription --17169248041.0a93dE8E.47999 Date: Tue, 28 May 2024 15:33:24 -0400 MIME-Version: 1.0 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
For the napkin calculation: On disk, the IEntr= y is 38Bytes. Alas, writes occur always in (the ssd internal) blocksize. So= , essentially (assuming 4096 byte blocksize, which is quite optimistic), we= have a write efficiency of less than 1 percent.

A good firmware in the ssd could avoid needing a new block for the w= rite, if all bits are changed in teh same direction by the new data. This i= s defeated by the binary search in the index buckets. The entries must be s= orted, and the number of used entries is tob updated. So, for each new entr= y the whole (internal) block has to be remapped. Therefore an erasure block= must be erased after about  1m/4k index writes.

It seems, venti in its current form is a ssd killer, if they ar= e used for the isects.
= --17169248041.0a93dE8E.47999--