For those of you not yet familiar with Clustrix, we have developed ClustrixDB, the industry’s first scale-out SQL database engineered for the cloud. ClustrixDB is uniquely and ideally suited to handle massive volumes of ACID-compliant transactional workloads while concurrently running real-time analytics on the same operational data. (See: https://cloudtools.rackspace.com/apps/835#!overview ).
Delivering a database like ClustrixDB with linearly-scaling performance, automated fault-tolerance and self-healing in a cloud would be a significant challenge. Success would depend heavily upon a well-architected infrastructure on which to deliver this level of enterprise value at the SLAs our customers had come to expect from us. And we were actively seeking a cloud with such capabilities.
Where Did ClustrixDB Grow Up?
Prior to our OCT ’13 GA release of ClustrixDB Software Version, we had only delivered ClustrixDB for production use on our own dedicated, turn-key Appliance clusters comprising 3-to-N nodes. Each node had eight 64-bit physical cores, 48GB RAM, 1GB NVRAM, 7 x 160GB SSDs, dual 40Gbps Infiniband fabrics for node interconnect, and multiple Gb Ethernet NICs for front-end client SQL access.
Years of development and commercial success had proven our Appliance to be the optimal delivery vehicle for on-premise or co-located datacenter Enterprise deployments, but extremely difficult, if not impossible, to deploy in a cloud. This was due to: