TY - GEN
T1 - Exploiting hardware transactional memory in main-memory databases
AU - Leis, Viktor
AU - Kemper, Alfons
AU - Neumann, Thomas
PY - 2014
Y1 - 2014
N2 - So far, transactional memory - although a promising technique - suffered from the absence of an efficient hardware implementation. The upcoming Haswell microarchitecture from Intel introduces hardware transactional memory (HTM) in mainstream CPUs. HTM allows for efficient concurrent, atomic operations, which is also highly desirable in the context of databases. On the other hand HTM has several limitations that, in general, prevent a one-to-one mapping of database transactions to HTM transactions. In this work we devise several building blocks that can be used to exploit HTM in main-memory databases. We show that HTM allows to achieve nearly lock-free processing of database transactions by carefully controlling the data layout and the access patterns. The HTM component is used for detecting the (infrequent) conflicts, which allows for an optimistic, and thus very low-overhead execution of concurrent transactions.
AB - So far, transactional memory - although a promising technique - suffered from the absence of an efficient hardware implementation. The upcoming Haswell microarchitecture from Intel introduces hardware transactional memory (HTM) in mainstream CPUs. HTM allows for efficient concurrent, atomic operations, which is also highly desirable in the context of databases. On the other hand HTM has several limitations that, in general, prevent a one-to-one mapping of database transactions to HTM transactions. In this work we devise several building blocks that can be used to exploit HTM in main-memory databases. We show that HTM allows to achieve nearly lock-free processing of database transactions by carefully controlling the data layout and the access patterns. The HTM component is used for detecting the (infrequent) conflicts, which allows for an optimistic, and thus very low-overhead execution of concurrent transactions.
UR - http://www.scopus.com/inward/record.url?scp=84901792922&partnerID=8YFLogxK
U2 - 10.1109/ICDE.2014.6816683
DO - 10.1109/ICDE.2014.6816683
M3 - Conference contribution
AN - SCOPUS:84901792922
SN - 9781479925544
T3 - Proceedings - International Conference on Data Engineering
SP - 580
EP - 591
BT - 2014 IEEE 30th International Conference on Data Engineering, ICDE 2014
PB - IEEE Computer Society
T2 - 30th IEEE International Conference on Data Engineering, ICDE 2014
Y2 - 31 March 2014 through 4 April 2014
ER -