• Publications
  • Influence
Cure: Strong Semantics Meets High Availability and Low Latency
Developers of cloud-scale applications face a difficult decision of which kind of storage to use, summarised by the CAP theorem. Currently the choice is between classical CP databases, which provideExpand
  • 77
  • 18
Antidote: the highly-available geo-replicated database with strongest guarantees
Cloud-scale applications need highly available, low latency responses to serve millions of users around the world. To meet this need, applications have to carefully choose a high performanceExpand
  • 3
  • 2
FMKe: a Real-World Benchmark for Key-Value Data Stores
Standard benchmarks are essential tools to enable developers to validate and evaluate their systems' design in terms of both relevant properties and performance. Benchmarks provide the means toExpand
  • 5
Global-Local View: Scalable Consistency for Concurrent Data Types
Concurrent linearizable access to shared objects can be prohibitively expensive in a high contention workload. Many applications apply ad-hoc techniques to eliminate the need for synchronous atomicExpand
  • 3
Transactions on Mergeable Objects
Destructible updates on shared objects require careful handling of concurrent accesses in multi-threaded programs. Paradigms such as Transactional Memory support the programmer in correctlyExpand
  • 3
Observing the consistency of distributed systems
In distributed Erlang systems temporary network issues are very likely; and can affect the consistency of the system. If the system is a distributed data store relying on a weak consistency model,Expand
  • 2
Highly-scalable concurrent objects
Synchronisation is a bottleneck in shared memory concurrent programs. In this paper, we present Mergeable Data Types (MDTs) for relaxing synchronisation in shared memory system analogously to CRDTsExpand
  • 1
Scalable Consistency in the Multi-core Era
The advent of heterogeneous many-core systems has increased the spectrum of achievable performance from multi-threaded programming. As the processor components become more distributed, the cost ofExpand
Performance simulation of buffer bloat in routers
Due to the dramatic price drop of memory network devices are configured with much larger buffer sizes than before in order to prevent packet drop. This undermines the assumptions of the TCPExpand
  • 1
Multi-view data types for scalable concurrency in the multi-core era
With the rapid growth of number of cores, together with the heterogeneous access latencies, the cost of synchronization and communication between distant components keeps growing. As more generalExpand