Date: Wed, 28 Oct 2020 08:11:10 +0000 From: Arthur Chance <freebsd@qeng-ho.org> To: Ralf Mardorf <ralf-mardorf@riseup.net>, freebsd-questions@freebsd.org Subject: Re: What is the "better / best " method to multi-boot different OSes natively WITHOUT VirtualBox(es) ? Message-ID: <d6155ea1-850f-79b7-9a62-1c67c12a6488@qeng-ho.org> In-Reply-To: <20201028073644.52fed6c6@archlinux> References: <CALMiprbGBaSJQUAA=1HDZAjvsVNK7dqB_5mBb5DKzV16F3hxHg@mail.gmail.com> <20201024123148.4929fb9e.freebsd@edvax.de> <alpine.BSF.2.21.9999.2010272340090.72530@fledge.watson.org> <20201028073644.52fed6c6@archlinux>
next in thread | previous in thread | raw e-mail | index | archive | help
On 28/10/2020 06:36, Ralf Mardorf wrote: > On Tue, 27 Oct 2020 23:57:05 +0000 (UTC), doug@safeport.com wrote: >> I would be interested in opinions on the best hard disks. > ^^^^^^^^ > a good chosen word, since not many > undisputed facts do exist > > In my experiences a batch of series x of vendor y could be very good or > bad and for the next batch it could be vice versa or a good series is > discontinued. Interesting are statistics from Google. I don't have a > link at hand, but I remember that they successfully used consumer > instead of enterprise disks and IIRC what Google experienced with > countless disks, was similar to my experiences with just a few disks. I > read an article years ago. I suspect you mean Backblaze rather than Google. Here's a link to their page on hard drives https://www.backblaze.com/b2/hard-drive-test-data.html You can get to the quarterly stats and useful hints from there. > IOW related to the lifespan it seems to be a lotto draw. > > Related to performance simple technical facts and clear measurements > do exist. It was already hinted by this thread. > > Shingled magnetic recording makes drives inexpensive, because it does > increase storage capacity. Due to the procedure it is required to > rewrite tracks, which results in performance lags. Different SMR drives > suffer from different performance lags. However, more expensive drives, > not using SMR technology tend to be faster when writing data. > > In the end you need to read a lot of papers and decide on your own what > is "the best" for you or in your opinion. And know that will change a few months later anyway. :-) -- The number of people predicting the demise of Moore's Law doubles every 18 months.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d6155ea1-850f-79b7-9a62-1c67c12a6488>