EOS Block Producer performance analytics: weekly stats. Sept 14 , 2018

johnchamberlain (48)in #eos • 6 years ago (edited)

Who are the best and the worst block producers ?

Each week I'll make a post listing the top and bottom block producers by Average CPU Time/Tx, μs

Determining performance of EOS block producers isn't easy, there's lots to take into account besides what is perceived from individual EOS block producer compliance and marketing efforts, or the various reports and summaries compiled.

The first chart identifies the Top 10 Block Producers by Avg CPU Time/Tx, μs. all 10 are under 3'000 μs microseconds (0.003 second).

Vasiliy's BP Performance Analytics - https://eosbpstats.com

The better Block Producers perform on Average CPU tx processing the greater the ability to process more transactions per staked tokens.

Vasiliy's BP Performance Analytics - https://eosbpstats.com

To really know block producers requires time and research from which informed thoughtful vote choices can be made let's face it, very few of the total EOS holders will spend the time to evaluate 30 Block Producers and caste an informed vote. So what are the incentives to vote at all ?

The main focus in this post is which BPs give more "BANG" per staked EOS for transaction processing, but this isn't the only factor to consider in vote choice. There can be a number of reasons a block producer would appear to be performing above or below average. Security and I'll emphases here that BP security policies can influence how data is reported publicly. So we can assume Block producers unlikely to sacrifice security simply to appear better on public performance analytics and even that below average can mean better than average security policies.

The Bottom 4 Block Producers by Avg CPU Time/Tx, μs. all 4 are over 3'000 μs microseconds (0.003 second). Remember this is a metric measured from the public EOS blockchain and that we could assume better security opposed to say hardware issues for these block producers.

Vasiliy's BP Performance Analytics - https://eosbpstats.com

The worse Block Producers perform on Average CPU tx processing means less transactions per staked tokens. Note could also indicate better security policies.

Vasiliy's BP Performance Analytics - https://eosbpstats.com

To make metrics easier to compare BP Performance Analytics excludes "spam" transactions from the "blocktwitter" account known to be greylisted by some Block Producers.

BP performance analytics is independently created (not by block producers) and reporting data is compiled from the EOS public blockchain.

What do you think of these performance metrics?

#eosio #blockchain #analytics

6 years ago in #eos by johnchamberlain (48)

Sort:

michaeldharris (38) 6 years ago

Great post. I'm liking the EOS communities vigilance about BPs even in the very earliest stages. Congrats to starteosiobp for no longer being the slowest BP. I know a lot of people have been pointing out their terrible (relatively) avg. CPU time/tx so I'm assuming they listened and upgraded their hardware.

Also, thanks for pointing out in the post that tx speed isn't the only metric that people should look for in voting fir BPs. While it should definitely be an important metric since tx times help the network improve, there are also qualitative points to consider. For example, cypherglass was in the bottom four, but they've done so much for incubation, outreach, education, partnerships, etc. They should definitely be recognized and compensated for their efforts. Hopefully they will use the BP rewards to upgrade the equipment to lower the tx speeds. They would then be an unstoppable force.

Thanks again for the post 👍

$0.00

1 vote

[-]