Big data storage platform
The platform is used for storing media data, system performance logs, computer incident logs, logs of bank transactions and messages downloaded from social networks. The density of downloaded data is 200 TB per 24 hours, with the maximum recording size up to 50 TB and the number of downloaded records up to 100 million per 24 hours.
The solution consists of two main components: big data storage and information search and a unified analytics engine for big data processing. АК. Big Data is driven by open source products: distributed data storage — RIAK KV + RIAK S2, APACHE Hadoop, APACHE SOLR indexing; engine for large-scale data processing — APACHE SPARK.
The big data storage component is a distributed key-value storage that allows you to unify hundreds of servers in a cluster and provide the specified reliability of data storage and performance of request processing.
117036, Moscow, Kedrova St., 15
Other developer projects: