Marathon

Apache Mesos is an open-source distributed-systems kernel and cluster resource manager developed at Berkeley around 2009 by Benjamin Hindman and maintained by the Apache Software Foundation. It abstracts CPU, memory, storage, ports, GPUs, and other resources from individual machines, presenting a datacenter or cloud environment as a shared resource pool. Its two-level, bring-your-own-scheduler model offers resources to application frameworks while allowing them to provide their own scheduling policies through Mesos APIs. Frameworks such as Hadoop, Spark, Kafka, Elasticsearch, Marathon, Chronos, and Aurora could be assembled on top of it; the system was used at large scale by organizations including Twitter and Apple. Mesos runs agents across clusters with replicated masters and ZooKeeper for fault tolerance, supports Docker and AppC containers, pluggable resource isolation, HTTP APIs, a web interface, and Linux, macOS, and Windows. The Apache Mesos project has retired, and its website directs users to the Apache Attic for details.

Chronos is a scheduling component built on Apache Mesos for running scheduled work.

Apache Aurora is a retired Apache Mesos framework for running long-lived services and cron jobs across a shared pool of machines. It keeps jobs running by rescheduling them onto healthy machines after failures, and provides rolling updates with automatic rollback, resource quotas, multi-user deployment, and a templating-capable configuration DSL. Aurora registers services with Apache ZooKeeper for client discovery. It was distributed under the Apache License 2.0.

Monad is a high-performance, decentralized layer-1 blockchain. The platform uses parallel execution and is described in the videos as supporting 300-millisecond block times, 600-millisecond finality, throughput of up to 10,000 transactions per second, and front-running protection. Its ecosystem includes applications such as the Moose on-chain aggregator and X-Stocks tokenized equities, while Monad is also presented as supporting private settlement zones and payment use cases for banks.

Microsandbox is a local-first microVM runtime and library for running untrusted workloads, including AI-agent tasks, user code, plugins, CI jobs, development environments, scrapers, and automation. It uses hardware-level isolation and runs standard OCI container images from registries such as Docker Hub and GHCR, with Docker-like image, command, shell, and volume workflows. The project provides SDKs for TypeScript, Rust, Python, Go, and Ruby, along with a CLI for booting and controlling sandboxes. Applications can create microVMs as child processes without a setup server or long-running daemon; sandboxes can also run detached for long-lived sessions. Agent Skills and an MCP server support agent-created sandboxes, and the project documents secret handling intended to keep keys out of the VM. It runs on Linux, macOS, and Windows, with platform-specific virtualization requirements, and is described as beta software.