Exploiting Total Order Multicast in Weakly Consistent Transactional Caches

Abstract

Abstract-Nowadays, distributed in-memory caches are increasingly used as a way to improve the performance of applications that require frequent access to large amounts of data. In order to maximize performance and scalability, these platforms typically rely on weakly consistent partial replication mechanisms. These schemes partition the data across the nodes and ensure a predefined (and typically very small) replication degree, thus maximizing the global memory capacity of the platform and ensuring that the cost to ensure replica consistency remains constant as the scale of the platform grows. Moreover, even though several of these platforms provide transactional support, they typically sacrifice consistency, ensuring guarantees that are weaker than classic 1-copy serializability, but that allow for more efficient implementations.This paper proposes and evaluates two partial replication techniques, providing different (weak) consistency guarantees, but having in common the reliance on total order multicast primitives to serialize transactions without incurring in distributed deadlocks, a main source of inefficiency of classical two-phase commit (2PC) based replication mechanisms.We integrate the proposed replication schemes into Infinispan, a prominent open-source distributed in-memory cache, which represents the reference clustering solution for the well-known JBoss AS platform. Our performance evaluation highlights speed-ups of up to 40x when using the proposed algorithms with respect to the native Infinispan replication mechanism, which relies on classic 2PC-based replication.

Improving ZooKeeper Atomic Broadcast Performance When a Server ’orum Never Crashes

“…Zab performance therefore impacts that of Zookeeper. Furthermore, e cient atomic broadcast protocols have far wider applications, e.g., in coordinating transactions particularly in large-scale in-memory database systems [5,6]. In such applications, the atomic broadcast protocol typically operates in heavy load conditions and is expected to o er low latencies even at such extreme loads.…”

“…Broadcasting an acknowledgement is common in symmetric (leaderless) atomic broadcast protocols such as [5]. at it can help to avoid the leader broadcasting commit messages has been hinted by Zab authors themselves (e.g., [1]).…”

Improving ZooKeeper Atomic Broadcast Performance When a Server Quorum Never Crashes

“…Broadcasting an acknowledgement is common in symmetric (leaderless) atomic broadcast protocols such as [15]. That it can help to avoid the leader broadcasting commit messages has been hinted by Zab authors themselves (e.g., [10]).…”

Self-tuning in Distributed Transactional Memory

“…At their top most layer, existing DTM platforms expose APIs analogous to those provided by non-distributed TMs that allow to define a set of accesses to in-memory data to be performed within an atomic transaction. The actual API exposed by a DTM is ultimately influenced by the data model that it adopts; the range of data models explored in the DTM literature includes, besides the object-based [7] and word-based [5] ones (typically employed in non-distributed TMs), also popular alternatives in the NoSQL domain, like the key-value [13,19] model. Certain DTM platforms [20,21] that support partial replication schemes (i.e., which do not replicate data at every replica of the system) provide also dedicated API support to influence the policies employed to determine the placement of data (and its replicas) across the nodes of the system, with the goal of enhancing the data locality achieved by DTM applications.…”