How to efficiently snapshot transactional data: hardware or software controlled

The quest for real-time business intelligence requires exe-cuting mixed transaction and query processing workloads on the same current database state. However, as Hari-zopoulos et al. [6] showed for transactional processing, co-execution using classical concurrency control techniques will not yield the necessary performance – even in re-emerging main memory database systems. Therefore, we designed an in-memory database system that separates transaction processing from OLAP query processing via periodically re-freshed snapshots. Thus, OLAP queries can be executed without any synchronization and OLTP transaction process-ing follows the lock-free, mostly serial processing paradigm of H-Store [8]. In this paper, we analyze different snapshot mechanisms: Hardware-supported Page Shadowing, which lazily copies memory pages when changed by transactions, software controlled Tuple Shadowing, which generates a new version when a tuple is modified, software controlled Twin Tuple, which constantly maintains two versions of each tu-ple and HotCold Shadowing, which effectively combines Tu-ple Shadowing and hardware-supported Page Shadowing by clustering update-intensive objects. We evaluate their per-formance based on the mixed workload CH-BenCHmark which combines the TPC-C and the TPC-H benchmarks on the same database schema and state. 1