Building a Columnar Database on RAMCloud

1. Front Matter

   Pages i-xix

   PDF

2. Introduction

   • Christian Tinnefeld

   Pages 1-6

3. Related Work and Background

   • Christian Tinnefeld

   Pages 7-42

4. A Database System Architecture for a Shared Main Memory-Based Storage

   1. Front Matter

      Pages 43-43

      PDF

   2. System Architecture

      • Christian Tinnefeld

      Pages 45-51

   3. Data Storage

      • Christian Tinnefeld

      Pages 53-62

   4. Data Processing

      • Christian Tinnefeld

      Pages 63-66

5. Database Operator Execution on a Shared Main Memory-Based Storage

   1. Front Matter

      Pages 67-67

      PDF

   2. Operator Execution on One Relation

      • Christian Tinnefeld

      Pages 69-77

   3. Operator Execution on Two Relations

      • Christian Tinnefeld

      Pages 79-91

6. Evaluation

   1. Front Matter

      Pages 93-93

      PDF

   2. Performance Evaluation

      • Christian Tinnefeld

      Pages 95-99

   3. High-Availability Evaluation

      • Christian Tinnefeld

      Pages 101-102

   4. Elasticity Evaluation

      • Christian Tinnefeld

      Pages 103-108

7. Conclusions

   1. Front Matter

      Pages 109-109

      PDF

   2. Conclusions

      • Christian Tinnefeld

      Pages 111-112

8. Back Matter

   Pages 113-130

   PDF

This book examines the field of parallel database management systems and illustrates the great variety of solutions based on a shared-storage or a shared-nothing architecture. Constantly dropping memory prices and the desire to operate with low-latency responses on large sets of data paved the way for main memory-based parallel database management systems. However, this area is currently dominated by the shared-nothing approach in order to preserve the in-memory performance advantage by processing data locally on each server. The main argument this book makes is that such an unilateral development will cease due to the combination of the following three trends: a) Today’s network technology features remote direct memory access (RDMA) and narrows the performance gap between accessing main memory on a server and of a remote server to and even below a single order of magnitude. b) Modern storage systems scale gracefully, are elastic and provide high-availability. c) A modern storage system such as Stanford’s RAM Cloud even keeps all data resident in the main memory. Exploiting these characteristics in the context of a main memory-based parallel database management system is desirable. The book demonstrates that the advent of RDMA-enabled network technology makes the creation of a parallel main memory DBMS based on a shared-storage approach feasible.

• Cloud computing
• Database operator execution
• Distributed systems
• In-Memory database technology
• Remote direct memory access
• Shared main memory-based storage

• University of Potsdam, Hasso Plattner Institute, Potsdam, Germany

  Christian Tinnefeld

Dr. Christian Tinnefeld received his B.Sc. and M.Sc. degrees from the Hasso Plattner Institute at the University of Potsdam, Germany, where he also pursued his doctoral studies. His main research interests are In-Memory Databases and Cloud Computing. In the former area, he has been collaborating with SAP for six years and contributed to initial concepts of the SAP HANA database. In the latter area, Christian has been collaborating with the RAM Cloud Project at Stanford University, California, USA for three years.

Policies and ethics