1. Introduction

Bagri is a native XML database built on top of distributed cache. It is pure Java app, a kind of IMDG clustered solution. XML is not the only data format supported by the system. It also accepts JSON documents and can be extended for others. So, it is capable to process and manage semi-structured documents where data model is specified in the document itself. CSV is another format which can be supported by the system. Bagri does not require any kind of pre-registered meta-data (like XML schema) to process incoming (XML) documents properly. Data comes to the system as documents, where they’re parsed and efficiently stored. Then clients can perform queries against stored documents by means of XQuery. Thus, Bagri is a kind of document database also. Bagri is a client-server system. It can be deployed as one or many data servers which holds data and run distributed queries on it. Bagri supports any number of clients connected to the system in parallel. For small-sized deployments the system supports embedded deployment mode too.