This topic is intended to capture Subnet Management activities over time for the shefu subnet, providing a place to ask questions and make observations about the management of this subnet.
At the time of creating this topic the current subnet configuration is as follows:
DFINITY will submit an NNS proposal today to reduce the notarization delay on the subnet, shefu, similar to what has happened on other subnets in recent weeks (you can find all details in this forum thread).
Voted to adopt proposal 133059. The subnet shefu being an application subnet, where new subnets of this type are already being created by default with the proposed 300ms for the initial_notary_delay_millis as seen in this proposal.
Thanks for the work @dsharifi. Also could you help me find if there is a way to see what has already been done for the Synchronous message submission endpoint and the Latency-aware ingress routing parts of the TOKAMAK roadmap?
Voted to adopt proposals 133059, 133060, 133064-133070 & 133072-133077.
This set of proposals is intended to reduce network latency by reducing the notarization delay to 300ms for 15 of the 18 subnets that currently have a notarization delay of 600ms, all of which are application subnets. 15 other subnets (14 application + system (Bitcoin)) already have a notarization delay of 300ms. I understand that performance testing has been done on this change, as noted here.
2 removed nodes (Switzerland - down, US - up) replaced with nodes in Canada and Japan. The second swap was to improve decentralisation. Looks good, I’ve voted to adopt.
Decentralisation Stats
Subnet node distance stats (distance between any 2 nodes in the subnet) →
Smallest Distance
Average Distance
Largest Distance
EXISTING
0 km
6096.726 km
15663.139 km
PROPOSED
0 km
6729.961 km (+10.4%)
15089.7 km (-3.7%)
This proposal slightly increases decentralisation, considered purely in terms of average geographic distance (and therefore there’s a slight theoretical increase in localised disaster resilience).
Subnet characteristic counts →
Continents
Countries
Data Centers
Owners
Node Providers
EXISTING
3
10
13
13
13
PROPOSED
3
10
13
13
13
Largest number of nodes with the same characteristic (e.g. continent, country, data center, etc.) →
The above subnet information is illustrated below, followed by a node reference table:
Map Description
Red marker represents a removed node (transparent center for overlap visibility)
Green marker represents an added node
Blue marker represents an unchanged node
Highlighted patches represent the country the above nodes sit within (red if the country is removed, green if added, otherwise grey)
Light grey markers with yellow borders are examples of unassigned nodes that would be viable candidates for joining the subnet according to formal decentralisation coefficients (so this proposal can be viewed in the context of alternative solutions that are not being used)
Known Neurons to follow if you're too busy to keep on top of things like this
If you found this analysis helpful and would like to follow the vote of the LORIMER known neuron in the future, consider configuring LORIMER as a followee for the Subnet Management topic.
Other good neurons to follow:
Synapse (follows the LORIMER and CodeGov known neurons for Subnet Management, and is a generally well informed known neuron to follow on numerous other topics)
CodeGov (actively reviews and votes on Subnet Management proposals, and is well informed on numerous other technical topics)
WaterNeuron (the WaterNeuron DAO frequently discuss proposals like this in order to vote responsibly based on DAO consensus)
Proposal 133449 replaces node pri5a which appears as “Status: Offline” in the dashboard but it seems to be running well according to the Node Provider Rewards tool. @sat@SvenF Am I missing something?
Hi @timk11 there have been several networking issues with all nodes from Sygum bank last week that our networking team worked on together with Sygnum, the dre-heal tool automatically selects nodes that need to be replaced.
If I look carefully at the graph I see this nodes also has some failed blocks as block master on the 18th, at the time we run the dre-heal tool to replace nodes. Note that each node in a subnet takes turn in being a block master, so that’s why you do not see a large number of failures unless the node is complete dead.
Thans @SvenF for clarifying that. If I’m reading the graph correctly it shows 3 failed blocks and 16659 successfully proposed blocks. Correct me if I have that wrong. This also ties in with my questions in another post about the rate of block failure that would warrant replacement.
This proposal replaces node pri5a which appeared as “Status: Offline” in the dashboard prior to the proposal’s execution. It appeared to be working well in the Node Provider Reward tool but I take on board the explanation of the issues affecting nodes from this node provider. Additionally, decentralisation is improved with respect to country. The number of nodes in the US is still above the number allowed in the target topology but has been lowered from 4 nodes to 3 nodes.
Voted to adopt proposal 133449. The proposal replaces the dead node pri5a and node ugevo on subnet shefu with the nodes wxpaf and voskz. The replacement of the healthy node comes as a decentralization improvement on the Country metric.
Voted to adopt proposal 134183, as the reasoning is sound and the description matches the payload. This proposal replaces 2 healthy nodes, both of which appear as “Active” on the IC dashboard. The proposed change improves decentralisation with respect to country and brings the target topology parameters to within the requirements.
TLDR: I’m planning to adopt. Improves decentralisation in terms of country diversity. This is slightly at the expense of average geographic distance between nodes, and also the max number of nodes per continent (will be 7 instead of 6 after this proposal executes), however neither of these are formal IC Target Topology metrics.
Motivation:
replacing node psz2h-gwldw-4xyex-ca33n-lojke-aiert-jk7b5-mcnzp-4ohor-jrrw3-aae to optimize network topology
replacing node axo3p-axths-mmqxt-2gawc-zakoe-mgbww-zv7hv-ppnp6-twqz7-umjz4-kqe to optimize network topology
2 removed American nodes replaced with a node in China and Lithuania.
Decentralisation Stats
Subnet node distance stats (distance between any 2 nodes in the subnet) →
Smallest Distance
Average Distance
Largest Distance
EXISTING
0 km
6729.961 km
15089.7 km
PROPOSED
0 km
6348.994 km (-5.7%)
14604.536 km (-3.2%)
This proposal slightly reduces decentralisation, considered purely in terms of geographic distance (and therefore there’s a slight theoretical reduction in localised disaster resilience).
Subnet characteristic counts →
Continents
Countries
Data Centers
Owners
Node Providers
Node Operator
EXISTING
3
10
13
13
13
13
PROPOSED
3
12 (+16.7%)
13
13
13
13
This proposal slightly improves decentralisation in terms of jurisdiction diversity.
Largest number of nodes with the same characteristic (e.g. continent, country, data center, etc.) →
The above subnet information is illustrated below, followed by a node reference table:
Map Description
Red marker represents a removed node (transparent center for overlap visibility)
Green marker represents an added node
Blue marker represents an unchanged node
Highlighted patches represent the country the above nodes sit within (red if the country is removed, green if added, otherwise grey)
Light grey markers with yellow borders are examples of unassigned nodes that would be viable candidates for joining the subnet according to formal decentralisation coefficients (so this proposal can be viewed in the context of alternative solutions that are not being used)
Known Neurons to follow if you're too busy to keep on top of things like this
If you found this analysis helpful and would like to follow the vote of the LORIMER known neuron in the future, consider configuring LORIMER as a followee for the Subnet Management topic.
Another good neuron to follow is Synapse (follows the LORIMER and CodeGov known neurons for Subnet Management, and is a generally well informed known neuron to follow on numerous other topics)
The proposal replaces 2 healthy Active status nodes form Chicago 3 and Fremont, US with Awaiting nodes from HongKong 1, HK and Vilnius 2, LT in order to optimize network topology.
Voted to adopt proposal 134183. The proposal replaces two nodes from subnet shefu:
Removed Nodes: psz2h, axo3p.
Added Nodes: 2a7fq and 3lgs3.
The proposal was verified using the DRE tool to verify the metrics stated. All nodes replaced are healthy but this replacements improve the network topology on the country metric by reducing the number of nodes in the US from 3 to 1.
