Paper Title: Improving Cassandra's Performance with Docker and NVMe

Paper Abstract: The Cassandra distributed database achieves high throughput and large capacity with unparalleled resiliency by scaling linearly to many nodes and keeping multiple copies of the data. The cost of this high resiliency is increased latency and inefficient use of CPU cycles and storage. A new way to reduce the costs is by decoupling the storage from the nodes running the database by deploying a Cassandra Cluster over dockers connected to NVMe over Fabric drives. Using NVMe and reducing the required replication factors lowers latency significantly. Decoupling the volumes from the database nodes enables much more efficient use of CPU cycles by allowing multiple instances to share CPU and memory resources on the same servers, while improving the system’s overall flexibility and resiliency.

Paper Author: Kais Belgaied, Storage & Servers Division CTO, Sanmina Corp
David Paulsen, , Sanmina Corp

Author Bio: Kais Belgaied is currently CTO for Sanmina's Storage and Servers division, driving the company's adoption and growth strategy around NVMe. A prolific innovator with over 50 patents, he was previously the architect of the control plane scale-out and the availability of all-flash array clusters at the highly regarded startup Nimble Storage (acquired by HPE). He has also been an architect and team leader at VMware and Sun Microsystems. He is a self-starter with a proven record of successful project delivery in areas such as kernel development, software-defined storage, systems and networking virtualization, network protocols, and OS security and access controls. He has published articles and given or contributed to conference presentations on topics such as cache management, scalability, and hardware offload. He earned an MS and a postgraduate degree in computer science and applied mathematics from the Polytechnique Institute of Grenoble (France), and an MBA from Columbia.

Author 2 Bio: David Paulsen is a Member of Technical Staff at Sanmina’s Storage and Server’s Division (Newisys), where he works on all-flash storage systems, software RAID, and management middleware. He has over 25 years experience in software development, including positions at Coraid, Ricoh Innovations, and Hewlett-Packard. He has focused recently on software projects based on Python and C++, along with Docker, Cassandra, and MongoDB. He received his training in mathematics and computer science from University of California at Santa Barbara.