TY - GEN
T1 - Hardware transactions in nonvolatile memory
AU - Avni, Hillel
AU - Levy, Eliezer
AU - Mendelson, Avi
N1 - Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 2015.
PY - 2015/1/1
Y1 - 2015/1/1
N2 - Hardware transactional memory (HTM) implementations already provide a transactional abstraction at HW speed in multicore systems. The imminent availability of mature byte-addressable, nonvolatile memory (NVM) will provide persistence at the speed of accessing main memory. This paper presents the notion of persistent HTM (PHTM), which combines HTM and NVM and features hardwareassisted, lock-free, full ACID transactions. For atomicity and isolation, PHTM is based on the current implementations of HTM. For durability, PHTM adds the algorithmic and minimal HW enhancements needed due to the incorporation of NVM. The paper compares the performance of an implementation of PHTM (that emulates NVM aspects) with other schemes that are based on HTM and STM. The results clearly indicate the advantage of PHTM in reads, as they are served directly from the cache without locking or versioning. In particular, PHTM is an order of magnitude faster than the best persistent STM on read-dominant workloads.
AB - Hardware transactional memory (HTM) implementations already provide a transactional abstraction at HW speed in multicore systems. The imminent availability of mature byte-addressable, nonvolatile memory (NVM) will provide persistence at the speed of accessing main memory. This paper presents the notion of persistent HTM (PHTM), which combines HTM and NVM and features hardwareassisted, lock-free, full ACID transactions. For atomicity and isolation, PHTM is based on the current implementations of HTM. For durability, PHTM adds the algorithmic and minimal HW enhancements needed due to the incorporation of NVM. The paper compares the performance of an implementation of PHTM (that emulates NVM aspects) with other schemes that are based on HTM and STM. The results clearly indicate the advantage of PHTM in reads, as they are served directly from the cache without locking or versioning. In particular, PHTM is an order of magnitude faster than the best persistent STM on read-dominant workloads.
UR - http://www.scopus.com/inward/record.url?scp=84946010635&partnerID=8YFLogxK
U2 - 10.1007/978-3-662-48653-5_41
DO - 10.1007/978-3-662-48653-5_41
M3 - Conference contribution
AN - SCOPUS:84946010635
SN - 9783662486528
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 617
EP - 630
BT - Distributed Computing - 29th International Symposium, DISC 2015, Proceedings
A2 - Moses, Yoram
PB - Springer Verlag
T2 - 29th International Symposium on Distributed Computing, DISC 2015
Y2 - 7 October 2015 through 9 October 2015
ER -