<ahref="http://www.amazon.co.jp/Systems-Performance-Enterprise-Brendan-Gregg-ebook/dp/B00FLYU9T2/ref=sr_1_6?s=english-books&ie=UTF8&qid=1414762573&sr=1-6&keywords=system+performance"onclick="__gaTracker('send', 'event', 'outbound-article', 'http://www.amazon.co.jp/Systems-Performance-Enterprise-Brendan-Gregg-ebook/dp/B00FLYU9T2/ref=sr_1_6?s=english-books&ie=UTF8&qid=1414762573&sr=1-6&keywords=system+performance', 'Systems Performance: Enterprise and the Cloud');">Systems Performance: Enterprise and the Cloud</a>を読んでいます。システム管理の分野で顕著な成果として2013年度のLISA Awardというのを受賞した書籍らしいのですが、これがまた面白い。たぶん自分が何を説明しているのかを正確に理解している人が、わかりやすく伝える努力を惜しまずに書くとこんな書籍になるのだと思う。これは最後まで読まねば。
著者の人のスライドを発見したので貼り付け:
<divstyle="margin-bottom: 5px;">
<strong><atitle="Linux Performance Tools 2014"href="//www.slideshare.net/brendangregg/linux-performance-tools-2014"target="_blank">Linux Performance Tools 2014</a></strong> from <strong><ahref="//www.slideshare.net/brendangregg"target="_blank">Brendan Gregg</a></strong>
<ahref="https://www.amazon.co.jp/exec/obidos/ASIN/B00FLYU9T2/simsnes-22/ref=nosim/"onclick="__gaTracker('send', 'event', 'outbound-article', 'https://www.amazon.co.jp/exec/obidos/ASIN/B00FLYU9T2/simsnes-22/ref=nosim/', 'Systems Performance: Enterprise and the Cloud');"target="_blank"name="amazletlink">Systems Performance: Enterprise and the Cloud</a>
posted with <ahref="http://www.amazlet.com/"onclick="__gaTracker('send', 'event', 'outbound-article', 'http://www.amazlet.com/', 'amazlet');"title="amazlet"target="_blank">amazlet</a> at 14.10.31
> There are known knowns; there are thing we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns — there are things we do not know we do not know.
>
> — U.S. Secretary of Defense Donald Rumsfeld, Feb 12, 2002
Input/Output operations per second is a measure of the rate of data transfer operation.
#### Throughput
The rate of work performed. Especially in communications, the term is used to refer to the data rate (byte per second or bit per second).
#### Responsetime
The time for an operation to complete. This includes any time spent waiting and time spent being serviced (service time), including the time to transfer the result.
#### Latency
A measure of time an operation spends waiting to be serviced.
#### Utilization
For resources that service requests, utilization is a measure of how busy a resource is, based on how much time in a given interval it was actively performing work.
#### Saturation
The degree to which a resource has queued work it cannot service.
#### Bottleneck
In system performance, a bottleneck is a resource that limits the performance of the system. Identifying and removing systematic bottlenecks is a key activity of system performance.
#### workload
The input to the system or the load applied is the workload.
### CPU視点でみた場合の各パーツの処理に要する時間とわかりやすい卑近な例に例えたら
こんな感じの表が掲載されていました:
<ahref="https://www.flickr.com/photos/42332031@N02/15680611275"onclick="__gaTracker('send', 'event', 'outbound-article', 'https://www.flickr.com/photos/42332031@N02/15680611275', '');"title="CPU_cycletime_and_its_scaled_time by Kazuhiro MUSASHI, on Flickr"><imgclass="aligncenter"src="https://farm4.staticflickr.com/3952/15680611275_e9971829fd.jpg"alt="CPU_cycletime_and_its_scaled_time"width="500"height="316"/></a>