Index > Introduction

Introduction

ROMA is one of the data storing system for distributed key-value stores. It is a completely decentralized distributed system that consists of multiple processes, called nodes, on several machines. It is based on pure P2P architecture like a distributed hash table, thus it provides high availability and scalability.

ROMA is written in Ruby. However, following choices are available to access to ROMA.

• Client libraries of Ruby and Java are available.
• ROMA protocol is compatible with memcached text-based one so that, any memcached client libraries allows users to interact with ROMA.

Key Features

• Fault-Toralance
• There is no single point of failure in ROMA.
• The data stored in ROMA is replicated automatically. The number of replicas is configured by the users.
• ROMA automatically detects disable nodes (process) and then separates the node from the ROMA ring.

• High Scalability
• New nodes can be joined to ROMA dynamically.
• The procedure to join to ROMA is just lunching new ROMA nodes (processes).
• Joining new nodes to ROMA results in linearly-increasing its storage size and though put

• Memcached Protocol
• ROMA employs memcached text-based communication protocol.
• This means any memcached client libraries and telnet client are able to interact with ROMA

• Plug-in Architecture
• ROMA has plug-in architecture to be extended its function.
• It is possible to extend ROMA without changing its source code by preparing plug-ins