Our EOS technology Stack

Will be replicated across two Cloud locations

EOS42 - TECH STACK

Cloud Locations

Google & Azure: We will be hosting our Block producers and Nodes in the Google and Azure cloud solutions.
Block Producer: Our Primary block producer will be with Google and our backup with Azure.
Nodes: Each location will have 2 EOS nodes behind a network load balancer.

Block Producing & Node Hardware

CPU: 8 core Processor
Memory: 128GB
Storage: 512GB internal SSD storage.
Bandwidth: 16Gbps

Networking:

16Gbps: Our networking subnets will allow for speeds upto 16Gbps
Network Load balancing: Our nodes will be situated behind a Network load balancer that will balance all networking requests between both nodes. The Network load balancer is a pass-through load balancer and makes no changes to any of the TCP packets.
DDOS protection: The Google load balancer will also act as our DDOS protection. We have thoroughly tested the abilities of the Google load balancer to handle a DDOS attack.

Security

Automated updates: Our servers will install security updates automatically.
Bi-annual audits: We will employ an external security consultant to carry out bi-annual security audits on our infrastructure.
No incoming ports on our block producers: Our block producers will not allow any incoming connections. This ensures our Block Producers can stay hidden from the internet whilst still synching up to the blockhain and producing blocks.

Monitoring

EOS process failure: The EOS processes on our servers will be monitored 24/7 and will include event handlers to automate the restarting of the process.
Verifying EOS nodes are synching: Further to checking that our EOS processes are running, we will also verify that our nodes & block producers are staying in sync with the blockchain.
Fork check: We will also monitor the head_block numbers on upto 21 nodes and compare that number against our nodes and block producers to ensure we have not forked onto our own chain.

Support

VictorOps Any outage notifications that occur on our infrastructure will be captured by our monitoring solution and send over to VictorOps to alert the current on call engineer.
Rotating team of engineers 24/7 We have will a rotating team of 3 engineers monitoring our infrastructure 24/7
Continuous improvement By leveraging our incident timeline we are able to record monitoring data, system actions, and human remediation. Ous team can leverage these insights to drive continuous improvement.

Future - Hypervisor Baremetal Cloud Solution

Data Centre: Post launch we will be working on moving all our infrastructure to our own dedicated data centre space.
Hypervisor Cloud: We will be installing our own bare metal hardware running hypervisors to build up our custom cloud solution.
Multi Data centre: We will scale to a multi data centre solution to become truly independantly decentralized, by not having to rely on Google, Azure or Microsoft.