Rockaway Blockchain Profile

Hi,

Tomas here from Rockaway Blockchain Fund. We would love to help Vega in this stage of launch.

Team / Company

Rockaway Blockchain Fund (RBF) is VC investor into digital assets since 2017. The fund portfolio consists of 30% of Funds (Polychain, Pantera, Multicoin, etc…) and 70% direct investments (Solana, Terra, 1inch, Wintermute, etc…). We are also investors in Vega since the seed round in July 2019. As part of our value-add we are running node infrastructure for projects we have invested in. Our technical team is based in Prague, Czech Republic and we host also our infrastructure in this region. We are long term investor aiming to hold our Vega investment and support the network for at least 5-7 years.

Experience

We are operating Solana, Ethereum and Terra nodes in production environment. During our work we identified issues running Tendermint based networks, so we are currently focusing on hardening the node infrastructure to increase the availability time. We will also share these findings with the community. In addition to production operations, we are also participating in testnets (currently Solana, Agoric, Terra).

Security

In terms of physical security, the hardware is located in Tier III datacenter with access available only based on ID cards, there are currently 2 persons authorized to access the hardware in DC.

In terms of software security all servers are Linux based with access available only via authorized SSH keys, through internal VPN. Keys to the node accounts (i.e., consensus keys) are stored on separate bare metal machine only accessible within internal network, which is behind strict firewall. The keys themselves are loaded on Yubico Hardware Security Modules. Seeds to the nodes are sharded and are safely held by multiple parties.

Reliability

To prevent slashing for Tendermint based blockchains we are operating 2 nodes. The primary node is located in the SafeDX Tier III DC in Prague, the backup node is SwissLayer. By having our nodes outside of the common colocation centers (i.e. most of the Solana nodes are run in Hetzner DCs – Falkenstein and Helsinky accounts for cca 30% of nodes), we are substantially improving decentralization of the networks.

We are using our own HW from multiple trusted vendors with mission critical level of support. Our infrastructure is monitored with Prometheus with visualization layer in Grafana and alerted by OpsGenie.

Performance

We are using latest AMD Epyc CPUs (Milan) and Intel Xeon CPUs, with minimum 128GB of RAM. Currently our primary datacenter is served with dual 1G line with possible upgrade to 10G (when needed). Our already established relationship with Hardware vendors allows us to deploy new machines within a week.

Communication

We are always accessible via Twitter (DMs), our ID is @rbf_capital. We are also happy to chat via email (staking@rbf.capital) or via Telegram (@EmiT_RBF)

Governance

We are actively monitoring the existing proposals through our portfolio management process and participate in voting. We are also currently looking for DeFi Expert, which will be involved with the team and the community in developing new proposals on full time basis.

4 Likes

Hey guys, love the fact that you’re not just using AWS / Google / some other cloud and that you’re in bit unusual geographical locations :rocket: .

2 Likes

Hi Tomas, great to hear your long term intentions with the network. Just wondering how it has been running your own hardware? Have there been any surprise trips to the DC in the night?
And are you using kubernetes for orchestration?

Thanks, we think that using AWS/Google/other cloud is against the basic principle of decentralization, we try to be as independent as possible.

1 Like

Hi Zach, there has not been any issue, which involved DC trip :slight_smile: We are using proven HW vendors and everything that has moving parts is doubled :slight_smile:

We are monitoring HW thoroughly, for example we are monitoring total writes to SSD disks, when they are reaching its designed endurance, disk is replaced with a new one.

Currently there is no need for Kubernetes because we are trying to keep production infrastructure as simple as possible.
Some of the new projects we would like to deploy in the future are using Kubernetes, so we are looking into it.

1 Like

Thanks Tomas, that’s great to hear it’s going smoothly :surfing_man:

Totally agree that public cloud providers are best avoided for decentralisation!

Definitely a plus to be using own hardware and we want to avoid concentration of cloud providers, data centres, countries, etc. as much as possible.