-
Notifications
You must be signed in to change notification settings - Fork 228
Home
FIRST-TIME CONTRIBUTORS: Please read the "Resources for Contributors" section on the right sidebar and follow the guidelines closely! |
---|
Apache Helix is a generic cluster management framework used for the automatic management of partitioned, replicated and distributed resources hosted on a cluster of nodes. Helix automates reassignment of resources in the face of node failure and recovery, cluster expansion, and reconfiguration.
To understand Helix, you first need to understand cluster management. A distributed system typically runs on multiple nodes for the following reasons:
- scalability
- fault tolerance
- load balancing
Each node performs one or more of the primary functions of the cluster, such as storing and serving data, producing and consuming data streams, and so on. Once configured for your system, Helix acts as the global brain for the system. It is designed to make decisions that cannot be made in isolation. Examples of such decisions that require global knowledge and coordination:
scheduling of maintainence tasks, such as backups, garbage collection, file consolidation, index rebuilds repartitioning of data or resources across the cluster informing dependent systems of changes so they can react appropriately to cluster changes throttling system tasks and changes While it is possible to integrate these functions into the distributed system, it complicates the code. Helix has abstracted common cluster management tasks, enabling the system builder to model the desired behavior with a declarative state model, and let Helix manage the coordination. The result is less new code to write, and a robust, highly operable system.
- Automatic assignment of resources and partitions to nodes
- Node failure detection and recovery
- Dynamic addition of resources
- Dynamic addition of nodes to the cluster
- Pluggable distributed state machine to manage the state of a resource via state transitions
- Automatic load balancing and throttling of transitions
- Optional pluggable rebalancing for user-defined assignment of resources and partitions
Modeling a distributed system as a state machine with constraints on states and transitions has the following benefits:
- Separates cluster management from the core functionality of the system.
- Allows a quick transformation from a single node system to an operable, distributed system.
- Increases simplicity: system components do not have to manage a global cluster. This division of labor makes it easier to build, debug, and maintain your system.
Pull Request Description Template
ZooKeeper API module for Apache Helix
DataAccessor for Assignment Metadata
Concurrency and Parallelism for BucketDataAccessor
WAGED Rebalance Pipeline Redesign
WAGED rebalancer Hard Constraint Scope Expansion
IdealState Dependency Removal Progression Remove requested state in Task Framework