Gossip-based visibility control for high-performance geo-distributed transactions

• Published in: The VLDB journal Vol. 30; no. 1; pp. 93 - 114
• Main Authors: Fan, Hua, Golab, Wojciech
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 2021; Springer Nature B.V
• Subjects: Best Papers of VLDB 2019; Computer centers; Computer Science; Concurrency; Concurrency control; Data centers; Data replication; Database Management; Gossip; Protocol (computers); Special Issue Paper; Transaction processing; Visibility; Wide area networks; Geo-distributed data; Replication; Concurrency control; Distributed transactions
• ISSN: 1066-8888; 0949-877X
• DOI: 10.1007/s00778-020-00626-5

Providing ACID transactions under conflicts across globally distributed data is the Everest of transaction processing protocols. Transaction processing in this scenario is particularly costly due to the high latency of cross-continent network links, which inflates concurrency control and data replication overheads. To mitigate the problem, we introduce Ocean Vista—a novel distributed protocol that guarantees
strict serializability
. We observe that concurrency control and replication address different aspects of resolving the visibility of transactions, and we address both concerns using a multi-version protocol that tracks visibility using version watermarks and arrives at correct visibility decisions using efficient gossip. Gossiping the watermarks enables asynchronous transaction processing and acknowledging transaction visibility in batches in the
concurrency control
and
replication
protocols, which improves efficiency under high cross-data center network delays. In particular, Ocean Vista can access
conflicting
transactions in parallel and supports efficient
write-quorum
/
read-one
access using one round trip in the common case. We demonstrate experimentally in a multi-data center cloud environment that our design outperforms a leading distributed transaction processing engine (TAPIR) more than tenfold in terms of peak throughput, albeit at the cost of additional latency for gossip and a more restricted transaction model. The latency penalty is generally bounded by one wide area network (WAN) round trip time (RTT), and in the best case (i.e., under light load) our system nearly breaks even with TAPIR by committing transactions in around one WAN RTT.