ترغب بنشر مسار تعليمي؟ اضغط هنا

State-Machine Replication for Planet-Scale Systems (Extended Version)

72   0   0.0 ( 0 )
 نشر من قبل Vitor Enes
 تاريخ النشر 2020
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

Online applications now routinely replicate their data at multiple sites around the world. In this paper we present Atlas, the first state-machine replication protocol tailored for such planet-scale systems. Atlas does not rely on a distinguished leader, so clients enjoy the same quality of service independently of their geographical locations. Furthermore, client-perceived latency improves as we add sites closer to clients. To achieve this, Atlas minimizes the size of its quorums using an observation that concurrent data center failures are rare. It also processes a high percentage of accesses in a single round trip, even when these conflict. We experimentally demonstrate that Atlas consistently outperforms state-of-the-art protocols in planet-scale scenarios. In particular, Atlas is up to two times faster than Flexible Paxos with identical failure assumptions, and more than doubles the performance of Egalitarian Paxos in the YCSB benchmark.



قيم البحث

اقرأ أيضاً

Modern web applications replicate their data across the globe and require strong consistency guarantees for their most critical data. These guarantees are usually provided via state-machine replication (SMR). Recent advances in SMR have focused on le aderless protocols, which improve the availability and performance of traditional Paxos-based solutions. We propose Tempo - a leaderless SMR protocol that, in comparison to prior solutions, achieves superior throughput and offers predictable performance even in contended workloads. To achieve these benefits, Tempo timestamps each application command and executes it only after the timestamp becomes stable, i.e., all commands with a lower timestamp are known. Both the timestamping and stability detection mechanisms are fully decentralized, thus obviating the need for a leader replica. Our protocol furthermore generalizes to partial replication settings, enabling scalability in highly parallel workloads. We evaluate the protocol in both real and simulated geo-distributed environments and demonstrate that it outperforms state-of-the-art alternatives.
This paper revisits the ubiquitous problem of achieving state machine replication in blockchains based on repeated consensus, like Tendermint. To achieve state machine replication in blockchains built on top of consensus, one needs to guarantee fairn ess of user transactions. A huge body of work has been carried out on the relation between state machine replication and consensus in the past years, in a variety of system models and with respect to varied problem specifications. We systematize this work by proposing novel and rigorous abstractions for state machine replication and repeated consensus in a system model that accounts for realistic blockchains in which blocks may contain several transactions issued by one or more users, and where validity and order of transactions within a block is determined by an external application-dependent function that can capture various approaches for order-fairness in the literature. Based on these abstractions, we propose a reduction from state machine replication to repeated consensus, such that user fairness is achieved using the consensus module as a black box. This approach allows to achieve fairness as an add-on on top of preexisting consensus modules in blockchains based on repeated consensus.
Active replication is commonly built on top of the atomic broadcast primitive. Passive replication, which has been recently used in the popular ZooKeeper coordination system, can be naturally built on top of the primary-order atomic broadcast primiti ve. Passive replication differs from active replication in that it requires processes to cross a barrier before they become primaries and start broadcasting messages. In this paper, we propose a barrier function tau that explains and encapsulates the differences between existing primary-order atomic broadcast algorithms, namely semi-passive replication and Zookeeper atomic broadcast (Zab), as well as the differences between Paxos and Zab. We also show that implementing primary-order atomic broadcast on top of a generic consensus primitive and tau inherently results in higher time complexity than atomic broadcast, as witnessed by existing algorithms. We overcome this problem by presenting an alternative, primary-order atomic broadcast implementation that builds on top of a generic consensus primitive and uses consensus itself to form a barrier. This algorithm is modular and matches the time complexity of existing tau-based algorithms.
Atomic multicast is a communication primitive that delivers messages to multiple groups of processes according to some total order, with each group receiving the projection of the total order onto messages addressed to it. To be scalable, atomic mult icast needs to be genuine, meaning that only the destination processes of a message should participate in ordering it. In this paper we propose a novel genuine atomic multicast protocol that in the absence of failures takes as low as 3 message delays to deliver a message when no other messages are multicast concurrently to its destination groups, and 5 message delays in the presence of concurrency. This improves the latencies of both the fault-tolerant version of classical Skeens multicast protocol (6 or 12 message delays, depending on concurrency) and its recent improvement by Coelho et al. (4 or 8 message delays). To achieve such low latencies, we depart from the typical way of guaranteeing fault-tolerance by replicating each group with Paxos. Instead, we weave Paxos and Skeens protocol together into a single coherent protocol, exploiting opportunities for white-box optimisations. We experimentally demonstrate that the superior theoretical characteristics of our protocol are reflected in practical performance pay-offs.
Transactional memory (TM) facilitates the development of concurrent applications by letting the programmer designate certain code blocks as atomic. Programmers using a TM often would like to access the same data both inside and outside transactions, and would prefer their programs to have a strongly atomic semantics, which allows transactions to be viewed as executing atomically with respect to non-transactional accesses. Since guaranteeing such semantics for arbitrary programs is prohibitively expensive, researchers have suggested guaranteeing it only for certain data-race free (DRF) programs, particularly those that follow the privatization idiom: from some point on, threads agree that a given object can be accessed non-transactionally. In this paper we show that a variant of Transactional DRF (TDRF) by Dalessandro et al. is appropriate for a class of privatization-safe TMs, which allow using privatization idioms. We prove that, if such a TM satisfies a condition we call privatization-safe opacity and a program using the TM is TDRF under strongly atomic semantics, then the program indeed has such semantics. We also present a method for proving privatization-safe opacity that reduces proving this generalization to proving the usual opacity, and apply the method to a TM based on two-phase locking and a privatization-safe version of TL2. Finally, we establish the inherent cost of privatization-safety: we prove that a TM cannot be progressive and have invisible reads if it guarantees strongly atomic semantics for TDRF programs.
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا