Simply VC and Akash — the unstoppable cloud
The cloud as we currently know it is run by a select few tech giants. This comes with a host of problems, including a very centralised internet, a complete lack of privacy and a reduction of consumer power. This centralisation is also made apparent when one provider has outages and so much of the internet goes offline.
Akash is solving this by building a decentralised cloud computing marketplace. Leveraging the capabilities of blockchain technology, they aim to replace tech giants with smaller data centres who can make use of their currently underutilised capacity. They have even produced the Supermini, a specially made device that is perfect for individuals wanting to become providers themselves.
The Akash team is currently testing the various functionality of the Akash network for scale and correctness. The current testing phase focuses on the testing of a Decentralised Exchange (DEX) that matches user submitted jobs (batches of computational work they would like to be carried out) to computational power providers.
Akash’s vision fits right into Simply VC’s vision, of distributing and decentralising the currently very centralised internet for better redundancy, security and privacy. In fact, we have recently partnered with Akash to pursue our common vision.
Our infrastructure is set up such that it is in line with our vision and beliefs. At Simply Staking we run and manage our infrastructure from our own Tier III data centre in Malta, providing us with power, internet connection and hardware redundancies. In addition, we utilise geographically distributed private hardware to provide further service redundancy.
Our Akash setup
To participate in the Akash network, we are using Dell PowerEdge servers, provisioned into VMs and Docker Containers via VMware Hypervisors, with each Akash VM having 8 cores, 32 GB of RAM and 200 GB of HDD space. The servers are set up such that failure of one machine will cause all VMs to fail over to a standby server. All data is stored on SANs making this failover functionality possible, with no data loss. All the above is protected by multiple layers of next-generation firewalls.
Our validator is protected behind a sentry architecture, such that the validator only communicates to the outside network through a set of geographically distributed sentries. This architecture protects our validator from DDoS attacks and mitigates Man-in-the-Middle attacks.
Monitoring
Monitoring is a crucial component of running and maintaining our validator infrastructure. For Akash validating, we use our own PANIC monitoring and alerting tool. This is open source and freely available here: https://github.com/SimplyVC/panic_cosmos
PANIC constantly monitors our nodes, alerting us in case of validator or sentry downtime, validator peer drops, bonded balance change and so on. Alerts are received in the form of phone calls, Telegram messages, and emails, depending on their severity.
When it comes to monitoring the state of our computation providers, we use:
- Grafana Dashboards to monitor the state (CPU/GPU, RAM and HDD usage) of our machines
- Graylog to store and alert on the system logs of each machine
- OpsGenie for alerts with escalation policies
