Publication
Distributed Transactional Systems Cannot Be Fast
Related publications (47)
When is it safe to run a transactional workload under Read Committed?
The popular isolation level multiversion Read Committed (RC) exchanges some of the strong guarantees of serializability for increased transaction throughput. Nevertheless, transaction workloads can sometimes be executed under RC while still guaranteeing se ...
ASSOC COMPUTING MACHINERY2023Efficient Protocols for Enforcing Causal Consistency in Geo-Replicated Key-Value Data Stores
Modern large-scale data platforms manage colossal amount of data, generated by the ever-increasing number of concurrent users. Geo-replicated and sharded key-value data stores play a central role when building such platforms. As the strongest consistency m ...
EPFL2020Fast General Distributed Transactions with Opacity
Aleksandar Dragojevic, Stanko Novakovic, Georgios Chatzopoulos
Transactions can simplify distributed applications by hiding data distribution, concurrency, and failures from the application developer. Ideally the developer would see the abstraction of a single large machine that runs transactions sequentially and neve ...
ASSOC COMPUTING MACHINERY2019PaRiS: Causally Consistent Transactions with Non-blocking Reads and Partial Replication
Willy Zwaenepoel, Diego Didona, Kristina Spirovska
Geo-replicated data platforms are the backbone of several large-scale online services. Transactional Causal Consistency (TCC) is an attractive consistency level for building such platforms. TCC avoids many anomalies of eventual consistency, eschews the syn ...
IEEE COMPUTER SOC2019The PCL Theorem: Transactions cannot be Parallel, Consistent, and Live
Rachid Guerraoui, Victor Bushkov, Panagiota Fatourou
We establish a theorem called the PCL theorem, which states that it is impossible to design a transactional memory algorithm that ensures (1) parallelism, i.e., transactions do not need to synchronize unless they access the same application objects, (2) ve ...
ASSOC COMPUTING MACHINERY2019The Complexity of Reliable and Secure Distributed Transactions
The use of transactions in distributed systems dates back to the 70's. The last decade has also seen the proliferation of transactional systems. In the existing transactional systems, many protocols employ a centralized approach in executing a distributed ...
EPFL2018Locking Timestamps versus Locking Objects
Rachid Guerraoui, Junxiong Wang, Tudor Alexandru David
We present multiversion timestamp locking (MVTL), a new genre of multiversion concurrency control algorithms for serializable transactions. The key idea behind MVTL is simple: lock individual timestamps instead of locking objects. After presenting a generi ...
ASSOC COMPUTING MACHINERY2018Wren: Nonblocking Reads in a Partitioned Transactional Causally Consistent Data Store
Willy Zwaenepoel, Diego Didona, Kristina Spirovska
Transactional Causal Consistency (TCC) extends causal consistency, the strongest consistency model compatible with availability, with interactive read-write transactions, and is therefore particularly appealing for geo-replicated platforms. This paper pres ...
2018A Scale-out Blockchain for Value Transfer with Spontaneous Sharding
Bitcoin, as well as many of its successors, require the whole transaction record to be reliably acquired by all nodes to prevent double-spending. Recently, many blockchains have been proposed to achieve scale-out throughput by letting nodes only acquire a ...
IEEE2018Locking Timestamps Versus Locking Objects
Rachid Guerraoui, Tudor Alexandru David
We present multiversion timestamp locking (MVTL), a new genre of multiversion concurrency control algorithms for serializable transactions. The key idea behind MVTL is simple and novel: lock individual time points instead of locking objects or versions. Af ...
2017