Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: =?UTF-8?Q?Arne_Vajh=C3=B8j?= <arne@vajhoej.dk>
Newsgroups: comp.os.vms
Subject: Re: RX2800 sporadic disk I/O slowdowns
Date: Sun, 20 Oct 2024 20:32:41 -0400
Organization: A noiseless patient Spider
Lines: 38
Message-ID: <vf47f9$k0cc$3@dont-email.me>
References: <veu99d$3derp$1@dont-email.me> <veumb8$3gbt8$1@dont-email.me>
 <vf1ku6$3vhv$1@dont-email.me> <vf1lvv$3s2a$1@dont-email.me>
 <vf1m6n$444r$2@dont-email.me> <vf1mi6$3s2a$2@dont-email.me>
 <vf1nhv$444r$8@dont-email.me> <vf1nof$3s2a$3@dont-email.me>
 <vf1uo4$8qpm$2@dont-email.me> <vf2vsv$dvmh$1@dont-email.me>
 <vf3rvn$itru$1@dont-email.me> <vf3sbs$iugp$1@dont-email.me>
 <vf3tdg$itru$10@dont-email.me> <vf42hq$k0cc$1@dont-email.me>
 <vf45v3$kfm0$1@dont-email.me> <vf46bu$k0cc$2@dont-email.me>
 <vf477e$kfm0$4@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Mon, 21 Oct 2024 02:32:42 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="e557ab2ded888c175ae698c2a2678a45";
	logging-data="655756"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19j6N4BGXxDFDmJQ+Qz2PyJ7ZUJeSyab10="
User-Agent: Mozilla Thunderbird
Cancel-Lock: sha1:ZLvhJufTNddXgZznnhlFWwon7YY=
In-Reply-To: <vf477e$kfm0$4@dont-email.me>
Content-Language: en-US
Bytes: 2738

On 10/20/2024 8:28 PM, Lawrence D'Oliveiro wrote:
> On Sun, 20 Oct 2024 20:13:50 -0400, Arne Vajhøj wrote:
>> On 10/20/2024 8:06 PM, Lawrence D'Oliveiro wrote:
>>> On Sun, 20 Oct 2024 19:08:41 -0400, Arne Vajhøj wrote:
>>>> Let me try again.
>>>>
>>>> DB write to plates & system crash => OK but slow
>>>
>>> The DB knows how to make this fast. Remember its cache is faster than
>>> any disk controller.
>>
>> This is where the DB is writing to plates.
>>
>> You can add a fourth scenario:
>>
>> DB write to DB cache & system crash => guaranteed problem with
>> transaction
> 
> Transaction resilience is a standard thing with databases (and journalling
> filesystems) going back decades.

Yes.

But they can't do miracles.

To be sure to come up ok after a system crash it is either write to
plates or write to a cache that will survive the system crash (raid
controller cache with battery backup).

> Some DBMSes don’t even want to work through filesystems, they would rather
> manage the raw storage themselves. This is why POSIX async I/O exists
> <https://manpages.debian.org/7/aio.7.en.html>.

Yes. That is to avoid any dangerous OS/filesystem cache (and possible
for better performance).

Arne