The Wiert Corner – irregular stream of stuff

Jeroen W. Pluimers on .NET, C#, Delphi, databases, and personal interests

  • My badges

  • Twitter Updates

    • RT @samgerrits: Caroline en asielzoekers, een tweeluik. Links: dwepen met een speldje gekregen van een Iraanse asielzoeker, rechts: nou ja… 4 hours ago
    • RT @delphijunkie: Yeah, nah. I'm good thanks Twitter. https://t.co/eTMPUoeSEa 4 hours ago
    • RT @d_feldman: Microsoft: We have world class AI research Google: We have world class AI research Meta: We’re one or two steps behind in AI… 4 hours ago
    • RT @SchipholWatch: Op dit moment is kerosine zo’n tien keer goedkoper dan alternatieve synthetische brandstof. De overheid moet dit prijsve… 4 hours ago
    • RT @jasongorman: One aspect of LLMs many folks overlook is the energy cost of training one. GPT-3 used an ~936 MWh and training it took 102… 4 hours ago
  • My Flickr Stream

  • Pages

  • All categories

  • Enter your email address to subscribe to this blog and receive notifications of new posts by email.

    Join 4,178 other subscribers

Some links and graphs on ESXi capping/throtteling disk speeds

Posted by jpluimers on 2022/01/06

As promised in “Solution” on ESXi 6.7 smartinfo throwing error Cannot open device, here are a few links in capping throttling disk speeds by ESXi followed by a few graphs of my own:

My own observations on ESXi 6.7 update 3:

  1. One rsync operation:
    1 rsync from 860 EVO SSD to 960 PRO NVMe

    1 rsync from 860 EVO SSD to 960 PRO NVMe

  2. Two rsync operations:
    2 rsync from 860 EVO SSD to 960 PRO NVMe

    2 rsync from 860 EVO SSD to 960 PRO NVM

  3. Resume actions were about 10 times faster than the single rsync read speeds:
    Resume action

    Resume action

  4. Suspend actions were between 4 and 6 times faster than rsync write speeds:
    Start of suspend action

    Start of suspend action

     

    Finish of suspend action

    Finish of suspend action

For each rsync operation, I had a separate SSH session going, and the speed doubled.

The resume action of all Virtual Machines was almost a flat speed curve.

The suspend action of all Virtual Machines started fast (when all machines were suspending) and finished slower (when only the largest virtual machines were still suspending)

–jeroen

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

 
%d bloggers like this: