Amounted to billions of dollars. In Newegg, tens of millions of users every day browsing merchandise and produce subsequent operations under a single transaction and so on. We build data systems must cope with the increasing amount of data, with the robustness and reliability. Currently, we use Cassandra Newegg to build the next generation of on-line system. Cassandra is no single point of failure in the distributed storage system, having good writing performance and concurrency performance random read, the data model supports the main Key-Value method. In the practice of Newegg, many business scenarios require secondary indexes, Cassandra comes with a secondary index poor performance can not meet the requirements; Cassandra cluster to integrate Solr cluster model, the development team need to maintain data synchronization Cassandra and Solr, read and write performance It will therefore reduce development costs high. To solve this problem, Newegg Big Data Team developed based on Solr, Cassandra built-in secondary index.
Original Address: http://www.infoq.com/cn/presentations/solr-and-cassandra-two-level-cache-practice
Reproduced in: https: //www.cnblogs.com/zhoujg/p/5181333.html