twitter / distributedlog
- среда, 11 мая 2016 г. в 03:12:54
Java
A high performance replicated log service.
DistributedLog (DL) is a high-performance, replicated log service, offering durability, replication and strong consistency as essentials for building reliable distributed systems.
DL is able to provide milliseconds latency on durable writes with large number of concurrent logs. Also handle high volume reads and writes per second from thousands of clients.
Messages are persisted on disk and replicated to store multiple copies to prevent data loss. They are guaranteed to be consistent among writers and readers in terms of strict ordering.
DL provides an efficient serving layer that optimized for running in a multi-tenant datacenter environment such as Mesos or Yarn. The serving layer is able to support large scale writes fan-in and reads fan-out.
DL supports various workloads from latency sensitive online transaction processing (OLTP) applications (e.g. WAL for distributed database and in-memory replicated state machines), real-time stream ingestion and computing to analytical processing.
DL is designed for I/O isolation in real-world workloads, to support large scale of number of logs for multi tenants.
DL has a modern layered-architecture design, that separates stateless serving tier from stateful storage tier. It allows scaling storage independent of scaling cpu and memory, to support large scale writes fan-in and reads fan-out.